Digitized by the Internet Archive

in 2018 with funding from

Kahle/Austin Foundation

https://archive.org/details/mmpi2mmpiinterpr0000gree
The MMPI-2 / MMPI:
An Interpretive Manual
The MMPI-2 / MMPI:
An Interpretive Manual

ROGER L. GREENE
Texas Tech University

ALLYN AND BACON

Boston London Toronto Sydney Tokyo Singapore
Copyright © 1991 by Allyn and Bacon
A Division of Simon & Schuster, Inc.
160 Gould Street
Needham Heights, Massachusetts 02194

All rights reserved. No part of the material protected by this copyright notice may
be reproduced or utilized in any form or by any means, electronic or mechanical, in¬
cluding photocopying, recording, or by any information storage and retrieval sys¬
tem, without written permission from the copyright owner.

Portions of this text appeared in The MMPI: An Interpretive Manual © 1980 by

The Psychological Corporation.

Library of Congress Cataloging-in-Publication Data

Greene, Roger L.
The MMPI-2/MMPI : an interpretive manual / Roger L. Greene,
p. cm.
“Portions of this text appeared in the MMPI: an interpretive
manual, c 1980’’- -T.p. verso.
Includes bibliographical references and index.
ISBN 0-205-12525-5
1. Minnesota Multiphasic Personality Inventory. I. Title.
II. Title: MMPI-two/MMPI.
[DNLM: 1. MMPI. WM 145 G799ma]
BF698.8.M5B75 1991
DNLM/DLC
for Library of Congress 91-13934
CIP

Printed in the United States of America

10 9 8 7 6 95 94
 To Grant and Leona Dahlstrom

for their professionalism,

scholarliness, and personal warmth

Contents

Preface ix Steps in Assessing MMPI-2 Validity 50

Acknowledgments xi Item Omissions 50
Note about the Text xiii Consistency of Item Endorsement 61
Accuracy of Item Endorsement 76
Chapter 1
Cutting Scores for Assessing Validity in
The Evolution of the MMPI 1
Psychiatric Settings 100
The Early History of Objective Personality Impression Management and Self-
Inventories 1 Deception 104
Construction of the MMPI 4 Traditional Validity Scales 106
Interpretation of Individual Item Content 9 Validity Scale Configurations 116
Assessment of Test-Taking Attitudes 10 Normal K+ Profiles 121
Current Developments 12 Simulation as Role Playing 122

Chapter 2 Detection of Response Sets 124

A dministration and Scoring 31 Obvious and Subtle Items 128

Administration 31
Test Forms 34
AT-Correction 37 Chapter 4
Linear T Scores 37 Clinical Scales 135
Normalized T Scores 39 Scale 1: Hypochondriasis (Hs) 136
Uniform T Scores 40 Scale 2: Depression (D) 138
Computer Scoring 41 Scale 3: Hysteria (Hy) 143
Errors in Scoring 42 Scale 4: Psychopathic Deviate
Interpreting the Profile 43 (Pd) 151
Effects of Demographic Variables on Scale 5: Masculinity-Femininity
MMPI Profiles 43 (Mf) 154
Scale 6: Paranoia (Pa) 159
Chapter 3
Validity Indexes and Validity Scale 7: Psychasthenia (Pt) 164
Configurations 49 Scale 8: Schizophrenia (Sc) 167

A Note on the Concept of Validity on the Scale 9: Hypomania (Ma) 172

MMPI-2 49 Scale 0: Social Introversion (Si) 175

VII
VII! Contents

Chapter 5 Appendix A
Supplementary Scales, Content Scales, MMPI-2: Item Composition of
Critical Items, and Short Forms 181 Validity, Clinical, and Supplementary
Scales 357
Supplementary Scales 181
Traditional Supplementary Scales 182
Appendix B
Content Scales 194
MMPI-2: TScore Conversion
Additional MMPI-2 Supplementary Tables 383
Scales 207
Additional MMPI Supplementary
Appendix C
Scales 213
MMPI-2: Critical Items 395
Critical Items 220
Short Forms 226
Appendix D
Prototypic Scores for Specific
Chapter 6
Codetypes in Psychiatric Settings 399
Co de types 231
MMPI Cookbook Interpretive
Appendix E
Systems 232
MMPI-2: Item Overlap among MMPI-2
Frequencies of Codetypes 235 Scales 457
Concordance between MMPI-2 and MMPI
Codetypes 245
Appendix F
Codetype Stability 250
MMPI-2 to MMPI Conversion
Relationship between Codetypes and Tables 469
Psychiatric Diagnoses 252
Prototypic Scores for Codetypes 253
References 475
Codetypes 258

Chapter 7 A uthor Index 501

Interpreting the MMPI-2 Profile 287
Preliminary Issues in Profile Subject Index 511
Interpretation 287
The Interpretive Process 288
Examples of Profile Interpretation 294

Chapter 8
Specific Groups: Adolescents, the
Aged, Blacks, and Other Ethnic
Groups 331
Adolescents 331
The Aged 336
Ethnic Groups 338
Preface

The restandardization of the MMPI that re¬ have been moved to Chapter 7 to make it ex¬
sulted in the MMPI-2 has opened a whole plicit that the information on the Supplemen¬
new era of clinical use and research on this tary and Content Scales now in Chapter 5 is
venerable instrument. Although some an¬ to be included in profile interpretation.
guish has been aired about specific issues in
the use of the MMPI-2, most of this concern
PREFACE TO THE ORIGINAL
reflects how central the MMPI has been to
MMPI MANUAL
personality assessment for almost fifty years.
It will be exciting to watch the empirical data This book arose out of the frustrations of try¬
amass as we make the transition to the MMPI-2. ing to find a single source that would contain
This revision of my MMPI Manual is in¬ all the information needed by a clinician
tended to provide clinicians who have used learning to interpret the MMPI (the Minne¬
the MMPI for years the necessary informa¬ sota Multiphasic Personality Inventory).
tion for making the transition to the MMPI- Some texts were too advanced and detailed
2. Consequently, this book contains the most for a clinician’s first exposure to the MMPI,
current summary of the research on the and often they omitted basic issues. Others
MMPI and the available information on the were too elementary to provide any instruc¬
MMPI-2. This “dual” coverage sometimes tion on how to interpret and use the results of
results in awkward referencing to the MMPI- the MMPI. Combining these two types of in¬
2 or the MMPI within the same section. formation often resulted in a hodgepodge of
MMPI-2 or MMPI will be used as most ap¬ material that was difficult for the clinician to
propriate to the specific material that is being organize and integrate.
discussed. Any reference to the use of the in¬ In addition there was the difficulty of
strument prior to 1989 is to the original trying to make the process of interpreting
MMPI. psychological tests explicit for clinicians, who
This revision also will serve as an intro¬ have often remarked that the basis for spe¬
duction to the MMPI-2 for clinicians with no cific statements or inferences in interpreta¬
prior experience with the test. As with the tions of MMPI profiles was not clear. Despite
original MMPI Manual, this revision is in¬ the empirical basis of the MMPI, there some¬
tended to provide a summary of the extant re¬ times was not a demonstrable correlate un¬
search on the instrument and a step-by-step derlying an inference that was nevertheless
procedure for profile interpretation. frequently used by clinicians.
The chapters have been reorganized In a single source this book provides an
slightly from my original MMPI Manual. introduction to some basic issues in the devel¬
The examples of clinical case interpretation opment and administration of the MMPI, as

IX
X Preface

well as a step-by-step procedure for interpret¬ The behaviorial or personality correlates

ing it. In addition, it discusses the use of the of each validity and clinical scale and all high-
MMPI with such special groups as adoles¬ point pairs are included to make the process
cents, the aged, and minorities, and it reviews of profile interpretation more explicit. The
the use and interpretation of critical items, two examples in Chapter 6 include a step-by-
special scales, and short forms. Although the step description of the procedure to be fol¬
text was written for the clinician with little lowed by the clinician. With the process of
prior exposure to the MMPI, experienced cli¬ profile interpretation made more explicit, cli¬
nicians should find it helpful as a review of nicians can direct their research toward ascer¬
MMPI interpretation and as an update on the taining which inferences are accurate and
research in this rapidly growing area. which need to be revised or eliminated.
Acknowledgments

I wish to express my appreciation to five vided access to huge MMPI datasets that are
anonymous reviewers who provided valuable cited frequently throughout the text. I also
feedback on an earlier draft of the manu¬ would like to thank countless clinicians and
script. Their comments and insights helped clients who have provided numerous insights
me to elucidate many issues and to clarify the on the nuances of the MMPI-2 and MMPI.
content. Of course, I accept the ultimate re¬ Finally, I wish to thank the many individ¬
sponsibility for any inadequacies that may re¬ uals who gave me permission to reproduce
main. copyrighted materials. Specific citations oc¬
I specifically would like to thank Dr. cur whenever such material is used in the
James Hedlund and Drs. Robert C. Colligan book.
and Kenneth P. Offord who generously pro¬

XI
Note about the Text

The terms clinician and client are used tions, numerous studies that were reviewed
throughout this book as generic labels to de¬ could not be included. To the best of my
scribe the person interpreting and the person knowledge, however, no major substantive
taking the MMPI-2. Since the MMPI-2 is em¬ article on the MMPI-2 or MMPI has been
ployed in a variety of professional settings omitted.
and administered to an equally large variety In order to avoid the expression MMPI-
of persons, these terms were adopted to pro¬ 2/MMPI, either MMPI-2 or MMPI will be
vide continuity. Occasionally, when the spe¬ used when a statement is appropriate to only
cific content being discussed requires it, one version of the test, and MMPI-2 will be
another term will be used to describe the per¬ used when the information is known or as¬
son taking the MMPI-2. sumed to be relevant to both versions. Clini¬
All research published in major psycho¬ cians should realize that the correlates re¬
logical journals on the MMPI-2 and MMPI ported are almost entirely based on the
through December 1990 were reviewed for in¬ MMPI and their application to the MMPI-2
clusion in the book. This review focused on awaits empirical validation, although it is
the years since 1980 to provide an update of probably safe to assume that most of these
the recent research. Because of space limita¬ correlates will generalize to the MMPI-2.

XIII
 CHAPTER 1

The Evolution of
the MMPI

The Minnesota Multiphasic Personality In¬ After a brief review of the history of ob¬
ventory (MMPI) is currently the most widely jective personality inventories, this chapter
used and researched objective personality in¬ will describe the rationale underlying the de¬
ventory. Dahlstrom, Welsh, and Dahlstrom velopment of the MMPI and the methods
(1975) include almost 6,000 references on the used for item selection and scale construc¬
clinical and research applications of the tion. Problems associated with interpreting
MMPI in An MMPI Handbook; Buros’ the content of individual items as well as as¬
(1978) The Eighth Mental Measurements sessing test-taking attitudes will be discussed.
Yearbook contains more than 5,000 citations The chapter will conclude with a discussion
on the MMPI; and Lubin, Larsen, of the appropriateness of the original norms
Matarazzo,-and Seever (1985) report that the for the MMPI for contemporary use (cf. Col-
MMPI is the most frequently used test in pro¬ ligan, Osborne, Swenson, & Offord, 1983,
fessional settings. Originally devised by 1989) and the development of the MMPI-2
Hathaway and McKinley in 1940, the MMPI (Butcher, Dahlstrom, Graham, Tellegen, &
provides an objective means of assessing ab¬ Kaemmer, 1989), the current revision of the
normal behavior. A person taking the MMPI MMPI.
sorts 550 statements into one of three catego¬
ries: “true,” “false,” or “cannot say.” The
THE EARLY HISTORY OF OBJECTIVE
person’s responses to these statements are
PERSONALITY INVENTORIES
then scored on 10 clinical scales that assess
major categories of abnormal behavior. In Personality assessment, like intellectual as¬
addition, 4 validity scales assess the person’s sessment, received its first major impetus
test-taking attitudes. Table 1-1 illustrates the during World War I when a need arose for as¬
scale names and numbers of the 10 clinical sessment procedures to screen large numbers
and 4 validity scales. A standard profile sheet of individuals. In response to this demand,
(see Profile 1-1) is used for plotting the Woodworth and Poffenberger developed the
person’s scores on these 14 scales. Woodworth Personal Data Sheet (Wood-

1
2 Chapter 1

TABLE 1-1 MMPI Validity and Clinical Scales it did identify those recruits who were emo¬
tionally unsuitable for service in the army
Abbre¬ Number
under wartime conditions.
Scale Name Number viation of Items
The success of psychological testing dur¬
Validity ing World War I stimulated the development
Cannot Say ? in the next decade of several personality in¬
Lie L 1 5 ventories similar to the Personal Data Sheet.
F (Infrequency) F 64
30
Probably the best known of these instru¬
K (Correction) K
ments is the Bernreuter Personality Inventory
Clinical (Bernreuter, 1933), which measures neuroti-
Hypochondriasis 1 Hs 33 cism, dominance, introversion, and self-suf¬
Depression 2 D 60
ficiency. Like other personality inventories of
Hysteria 3 Hy 60
Psychopathic this era, the Bernreuter Personality Inventory
Deviate 4 Pd 50 was constructed on a rational rather than an
Masculinity- empirical basis. That is, the test developer
Femininity 5 Mf 60 would include items on a particular scale
Paranoia 6 Pa 40
that, on the basis of clinical experience, were
Psychasthenia 7 Pt 48
Schizophrenia 8 Sc 78 thought to measure a specific trait or con¬
Hypomania 9 Ma 46 struct. Likewise, the test developer would de¬
Social termine the scoring direction for any particu¬
Introversion 0 Si 70 lar item on a rational basis. For example, if
the test developer felt that a “yes” response
to the item “Do you daydream a lot?” indi¬
cated neuroticism, that item would be added
worth, 1920), a self-rating scale for detecting to the neuroticism scale with “yes” as the
neurotic individuals. They assembled 116 “deviant” response. The total number of
items reflecting neurotic symptoms to which these “deviant” responses, responses that the
a person answered “yes” or “no.” The total test developer felt tapped the specific trait or
number of positive answers resulted in a score construct being assessed, became the score on
that indicated whether the person should be the scale.
interviewed individually by a psychiatrist. Strong critiques (cf. Landis & Katz,
Some items were considered so pathogno¬ 1934; Super, 1942) devastated the Bernreuter
monic that a “yes” response to any of them Personality Inventory and other rationally
prompted an individual interview. The items derived personality inventories of this era.
were heterogeneous in content since they For example, to investigate how certain
tapped every symptom of psychological ner¬ groups would perform on the Bernreuter Per¬
vousness that Woodworth and Poffenberger sonality Inventory, Landis and Katz (1934)
could identify. The items were chosen be¬ administered the inventory to 224 patients
cause Woodworth and Poffenberger thought with a known clinical diagnosis and examined
that they assessed psychological maladjust¬ their scores. On the neuroticism scale 39 per¬
ment; no empirical or theoretical perspective cent of the neurotic patients scored above the
was employed in selecting items to be in¬ 90th percentile; 23 percent of the schizophre¬
cluded on the test. Although the Personal nic patients and 21 percent of the manic-de¬
Data Sheet was developed too late to be very pressive patients, however, also scored above
useful in selecting recruits, since the United the 90th percentile. Thus, this scale is inade¬
States was already involved in World War I, quate since in addition to identifying some
 The Evolution of the MMPI 3

PROFILE 1-1
Name Tim Smith
MINNESOTA MULTIPHASIC
PERSONALITY INVENTORY
Address 547 Geneva Avenue
S R. Hathaway and I.O. McKinley

PROEIl E Occupation Clerk Date Tested 5/ 6/89

MINM.V A \V,‘ii’HAMi i'l KSPNAl INVfNIOhh Education 12th Age 47
'iH 'd"' HI l 'M!VI RSI1N 01 MINNI SO IA Minnesota Multiphasu Personality Inventory and MMPI are
U.' ‘m • • v\ i o i in • AUK |d (i 198.' Ah i mills user veil tiadervMiks owned t>v The University ot Minnesota
n-sf :mi,v I vr.ts.viv, \AMONAI OOMI'UU R SNSHMS INI' Martial Status Married Referred by Dr. Clark
II Ml . I 'si ' lh'ivt’1 Silv Ol MmiVSl'il.l
•hr rm .I’HiV Stall’s id Amon a
1 234567890 , FOR RECORDING
i i Hs+SK D Hy Pd+4k Ml Pa Pt+tK Sc-*-1K Ma+?K S1 A R Es MAC ADDITIONAL SCALES TorTc

MMPI Code

Scorer's
Initials HG

Raw Score jD _4_3il^281Z183D_827_81234 23 18 40 2D_ NATIONAL

COMPUTER
K to be added _6 _A till J2. SYSTEMS
Raw Score with K 13 22 38 19 15 *49 item version 27309
4 Chapter 1

neurotic patients correctly, it also misclassi- groups. Even more striking, the normal indi¬
fied several groups of psychotic patients as viduals endorsed 11 of the 50 items more fre¬
neurotic. quently than the schizophrenics did. Some of
Furthermore, analyzing responses to in¬ these 11 items were: “Are you often occupied
dividual items revealed additional problems. with your own thoughts?” “Do you think it
Bernreuter weighted a positive response to is possible for other people to influence your
the items in the neurotic direction; however, actions?” “Do your emotions change fre¬
Landis and Katz (1934) found that other quently without cause?”
groups endorsed some items as much or more Thus, the early personality inventories
frequently than neurotics. For example, the constructed on a rational basis were unsuc¬
item “Are you critical of others?” elicited a cessful outside of a wartime setting. This
“yes” response from 69 percent of the nor¬ should not be interpreted as an indictment of
mal sample as compared with 32 percent of the general procedure, however. In the last
the neurotic sample and 39 percent of the psy¬ three decades several widely used personality
chotic sample. Similarly, the item “Do you tests have been developed at least partly on a
daydream frequently?” was answered “yes” rational basis, such as the Edwards Personal
by 43 percent of the normal sample, 40 per¬ Preference Schedule (Edwards, 1959) and the
cent of the neurotic sample, and 31 percent of Personality Research Form (Jackson, 1968).
the psychotic sample. Wiggins (1966) also successfully constructed
Other studies (cf. Page, Landis, & Katz, 13 content scales for the MMPI on a rational
1934) demonstrated that identifying psy¬ basis, which have been validated as veridical
chotic individuals with a rationally derived self-reports of psychopathology (Jarnecke &
scale is also difficult. Page and associates Chambers, 1977; Lachar & Alexander, 1978).
constructed a rationally based scale by select¬ Chapter 5 provides further information on
ing 100 commonly accepted schizophrenic Wiggins’ content scales and the new content
symptoms and traits from the psychiatric lit¬ scales for the MMPI-2 (Butcher, Graham,
erature. The 50 traits considered schizophre¬ Williams, & Ben-Porath, 1989). Wiggins
nic by at least 10 of the 12 psychiatrists who (1973) provides an excellent, in-depth analy¬
reviewed the list were combined into a ques¬ sis of the relative merits of empirically and ra¬
tionnaire. Page and colleagues administered tionally derived scales.
the questionnaire to 125 schizophrenic pa¬
tients, 100 manic-depressive patients, and 240
CONSTRUCTION OF THE MMPI
normal individuals, who were matched for
intelligence and social status. They found lit¬ Out of the psychometric wilderness of the
tle difference in the average number of the early 1930s appeared two men, Starke
“schizophrenic” traits reported by the three Hathaway and J. C. McKinley, who, under
groups: schizophrenics (M — 17.60; SD = the banner of empiricism, waged a new battle
7.50), manic-depressives (M = 14.00; SD = for the scientific advancement of personality
7.35), and normals (AL = 18.00; SD = 6.35). assessment. They sought to develop a multi¬
The normal individuals, in fact, actually ac¬ faceted or multiphasic personality inventory,
knowledged having more of the “schizophre¬ now known as the MMPI, that would sur¬
nic” traits than the schizophrenic patients mount the shortcomings of the previous per¬
did. sonality inventories, some of which were
In examining responses to individual described above. Instead of using indepen¬
items, Page and associates (1934) also found dent sets of tests, each with a special purpose,
inadequacies. Only 14 of the 50 traits reliably Hathaway and McKinley included in a single
differentiated one group from the other two inventory a wide sampling of behavior of sig-
 The Evolution of the MMPI 5

nificance to psychologists. They wanted to TABLE1-2 Content Categories

create a large pool of items from which vari¬ for MMPI Items
ous scales could be constructed, in the hope
Number
of evolving a greater variety of valid person¬ Content Category of Items
ality descriptions than was currently avail¬
able. Social attitudes 72
To this end, Hathaway and McKinley Political attitudes, law
and order 46
(1940) assembled more than 1,000 items from
Morale 33
psychiatric textbooks, other personality in¬ Affect, depressive 32
ventories, and clinical experience. After de¬ Delusions, hallucinations,
leting duplicate items and items that they illusions, ideas of reference 31
considered relatively insignificant for their Family and marital 29
Phobias 29
purposes, they arrived at a sample of 5041
Affect, manic 24
items. The items were written as declarative Habits 20
statements in the first-person singular, and Religious attitudes 20
most were phrased in the affirmative. General neurologic 19
Hathaway and McKinley (1940) arbitrarily Sexual attitudes 19
Occupational 18
classified the items under 25 headings as a
Lie 15
convenience in handling and in an effort to Obsessive, compulsive 15
avoid duplication (see Table 1-2). However, Educational 12
they did not attempt to obtain any particular Cranial nerves 1 1
number of items for a category or to insure Gastrointestinal 1 1
Vasomotor, trophic, speech,
that an item was actually properly classified
secretory 10
in a category. Table 1-2 shows that some cat¬ General health 9
egories are heavily overrepresented and other Sadistic, masochistic 7
categories are underrepresented. Genitourinary 6
Using the 504 items, Hathaway and Mc¬ Motility and coordination 6
Cardiorespiratory 5
Kinley (1940) next constructed a series of
Sensibility 5
quantitative scales that could be used to diag¬
TOTAL 504
nose abnormal behavior. In selecting items
for a specific scale (e.g., Hypochondriasis), Note: The category names and sizes are from
they used an empirical approach. The items Hathaway and McKinley (1 940).
had to be answered differently by the crite¬
rion group (e.g., hypochondriacal patients) as nite diagnostic categories, and hypo¬
compared with normal groups. Since their ap¬ chondriacs also were one of the largest
proach was strictly empirical and no theoretical groups of patients available to McKinley and
rationale was posited as the basis for accepting Hathaway. Since the procedure for develop¬
or rejecting items on a specific scale, it is not ing Scale 1 typifies the procedure for most of
always possible to discern why a particular item the clinical scales, it will be described in de¬
distinguishes the criterion group from normal tail. Later, the development of the other clin¬
groups. Rather, items were selected solely be¬ ical scales will be described only in cases
cause the criterion group answered them differ¬ where the procedure differs.
ently than other groups. The first step in developing Scale 1 was
Scale 1 (Hypochondriasis) was con¬ to select an appropriate criterion group.
structed first (McKinley & Hathaway, 1940).2 Using a diagnostic classification as the basis
This choice was not simply fortuitous. Hypo¬ for the criterion group selection was logical
chondriasis is one of the simpler, more defi- since McKinley and Hathaway’s intent was to
6 Chapter 1

develop an inventory to aid in differential di¬ ease in the general wards of the University
agnosis. They defined hypochondriasis as an Hospitals. None of the patients had obvious
abnormal neurotic concern over bodily psychiatric symptomatology. The fourth gen¬
health, excluding the symptomatic occur¬ eral normative group consisted of 221 pa¬
rence of hypochondriacal features in psy¬ tients in the psychopathic unit of the Univer¬
chotic individuals. Using this definition, they sity Hospitals, regardless of diagnosis.
selected 50 cases of pure, uncomplicated hy¬ Once the criterion group and the other
pochondriasis as their criterion group. reference groups were established, the pro¬
The next step was to select groups of cess of item selection began. For the criterion
normal individuals. The primary normative group and each of the normal groups, the fre¬
group, which served as the reference group quency of “true” and “false” responses was
for determining the standard MMPI profile calculated for each item. An item was consid¬
for over 40 years, consisted of 724 individuals ered significant and was tentatively selected
who were friends or relatives of patients in for a scale if the difference in frequency of re¬
the University Hospitals in Minneapolis. The sponse between the criterion group and the
only criterion for exclusion was if an individ¬ normative or reference groups was at least
ual was currently receiving treatment from a twice the standard error of the proportions of
physician. This group reflected a fairly repre¬ “true/false” responses of the two groups
sentative cross section for sex and marital sta¬ being compared. For example, the response
tus of the Minnesota population aged 16 to 55 frequencies for two potential items for Scale
in the late 1930s. Dahlstrom and colleagues 1 are provided in Table 1-3. In this example,
(1975) reported that all of the persons in the only two groups, the criterion group of hypo¬
primary normative group were white since chondriacs and the original normative group,
very few members of any ethnic minority are compared; before any items were finally
other than American Indian resided in Min¬ selected, the criterion group was compared
nesota at that time. The current normative with the other normative groups as well.
groups for the MM PI-2 will be described at The following (Ferguson, 1971) was used
the end of Chapters 1 and 2, and the use of for the test of the significance of the differ¬
the MMPI with minority individuals will be ence between two independent proportions:
described in Chapter 8.
Four additional normative groups were z P\ ~ Pi

used in the development of Scale 1 and other V pq[{\/nx) + (1 /n2)]

clinical scales on the MMPI. Two groups
were formed to assess whether “nuisance” where
variables such as age, socioeconomic class, or
education were influencing differential item p = the proportion of “true”
endorsement by members of the criterion responses in the total group
group and the primary normative group. One px = the proportion of “true”
group consisted of 265 precollege high-school responses in the first sample
graduates who came to the University of p2 — the proportion of “true”
Minnesota Testing Bureau for precollege responses in the second sample
guidance. The other was composed of 265 q = 1 - p
skilled workers from local Works Progress nx = the number of persons in the
Administration projects. A third normative first sample
group consisted of 254 patients who were n2 = the number of persons in the
hospitalized for some form of physical dis¬ second sample
 The Evolution of the MMPI 7

TABLE 1 -3 Frequency of Response by Group for Two Possible Items

for Scale 1 (Hypochondriasis)

Group

Normals3 Hypochondriacs13

Item True False True False

1. 1 have few or no pains 21 1(81%) 51(19%) 17(34%) 33(66%)

2. Much of the time my head
seems to hurt all over. 10( 4%) 252(96%) 5(10%) 45(90%)

3 n = 262
bn = 50

Thus, the values of p and q for the first item Using the same procedure for the second
would be the following: sample item would result in substituting the
_ 211 + 17 _ 228 following values in the formula:
P “ 262 + 50 “ 312 “ * .04 - .10
Z
q = 1 - p = 1.0 - .73 = .27 \ (.048) (.952^ [(1/262)“ + (1/50)]
Substituting these values in the above for¬ -.06
mula results in the following: -1.82
.033
.81 - .34 This item would not be included on Scale 7
Z =
\ (.73) (.27) [(1/262) + (1/50)] since the proportions of endorsement are not
.47 significantly different between the two
- 6.81
.069 groups.
Checking a standard table of Z values shows Having selected items according to this
that this Z value would result in a probability procedure, Hathaway and McKinley then
less than .001. Hathaway and McKinley con¬ eliminated some of them for various reasons.
sidered significant any percentage difference First, the frequency of the criterion group’s
of at least twice the standard error of the in¬ response was required to be greater than 10
dependent proportions, or any Z equal to or percent for nearly all items; those items that
greater than + 2. Since a Z of +2 has a prob¬ yielded infrequent “deviant” response rates
ability slightly less than .05 using a two-tailed from the criterion group were excluded even
test, they essentially selected only items that if they were highly significant statistically
were significant beyond the .05 level. Thus, since they represented so few criterion cases.
the first item in the preceding example would Additionally, items whose responses ap¬
be tentatively included in Scale 7, and a peared to reflect biases on variables such as
“false” response would be the “deviant” an¬ marital status or attitudes toward one’s chil¬
swer since the hypochondriacal patients re¬ dren were excluded.
sponded more frequently in the “false” Finally, Hathaway and McKinley re¬
direction. If this item also differentiated the jected a few more of the tentatively selected
hypochondriacal group from the other nor¬ items that, after a rational inspection of the
mative groups using an identical procedure, it list, they concluded were not germane to the
would then be included on Scale 7. construct of hypochondriasis. Interitem cor-
8 Chapter 1

relations were not calculated nor were any The normative group with physical dis¬
other psychometric bases used in selecting ease also was used in developing Scale 7. This
items. The psychometric problems that later group scored more like the normal group
were discovered with some of the validity and than like the hypochondriacal group on the
clinical scales arose because these issues were corrected Scale 7. Thus, their actual physical
not considered when each scale was con¬ symptoms appeared to alter their total scores
structed. These problems will be discussed only moderately in the direction of hypo¬
below as appropriate when each scale is re¬ chondriasis.
viewed. More recently, Scale 7 was modified
The preliminary Scale 7 consisted of 55 again. In order to differentiate Scale 7 more
items that had been identified by this proce¬ clearly from Scale 3 (Hysteria), McKinley
dure. The next step was weighting or combin¬ and Hathaway (1944) eliminated from Scale 7
ing them into a scale. Evaluation of several those correction items that also appeared on
methods of weighting individual items Scale 3, thus arbitrarily making Scale 7 into a
showed no advantage over using unweighted somatic complaints scale. They also elimi¬
items. Therefore, each item simply received a nated some of the original items from Scale 7
weight of “one” in deriving a total score. In that did not separate hypochondriacs from
other words, a person’s score on Scale 7 is normals under subsequent analyses. This
equal to the total number of items that the in¬ final step resulted in the 33 items that are cur¬
dividual answers as the criterion group did. rently used on Scale 7.
The responses of the normative group Soon after the development of Scale 7
consisting of psychiatric patients helped to (McKinley & Hathaway, 1940), five other
refine Scale 7. A fair number of psychiatric clinical scales were developed: 2 (Depression)
patients obtained high scores on this scale al¬ (Hathaway & McKinley, 1942); 7 (Psychas-
though the psychiatric staff had not noted the thenia) (McKinley & Hathaway, 1942); and 3
presence of hypochondriasis. To eliminate (Hysteria), 4 (Psychopathic Deviate), and 9
this potential source of bias, the responses of (Hypomania) (McKinley & Hathaway, 1944).
50 patients who had no hypochondriacal The description of the construction of three
symptoms but who obtained the highest scale other clinical scales—5 (Masculinity-Femi¬
scores on the preliminary Scale 7 were con¬ ninity), 6 (Paranoia), and 8 (Schizophre¬
trasted with the original criterion group of 50 nia)—was not published until 1956
hypochondriacal patients. Items showing a (Hathaway, 1956), although these three
significant difference in frequency of en¬ scales had been used routinely for more than
dorsement between these two groups were lo¬ a decade. (More detailed information on each
cated and combined into a separate grouping, of these scales will be provided in Chapter 4.)
known as the correction of Scale 7. (This cor¬ Scale 5 (Masculinity-Femininity) was de¬
rection of Scale 7 should not be confused veloped somewhat differently than the other
with the K-correction of Scale 7, which will clinical scales. Some 55 items, mostly related
be discussed later.) For each of these correc¬ to sexual orientation, were added to the
tion items that an individual answered in the MMPI item pool after the data already had
nonhypochondriacal direction, one point was been collected from the original normative
subtracted from the total score on Scale 7. sample.3 Thus, the criterion group of male
Cross-validation revealed that the corrected homosexuals who were used in developing
score on Scale 7 was more effective in differ¬ Scale 5 could not be contrasted with the orig¬
entiating the groups than the original uncor¬ inal normative group on these 55 items. Con¬
rected score. sequently, 54 male soldiers were used as one
 The Evolution of the MMPI 9

of the normative groups for this scale, and always the case. For example, for a test item
items that distinguished them from the male such as “I have headaches frequently,” per¬
homosexuals were included on Scale 5. In ad¬ sons may interpret “frequently” to mean
dition, items that differentiated males from once a day, once a week, or once a month and
females within the normative sample were in¬ respond “true” or “false” accordingly. One
cluded on this scale. The effects of these dif¬ client might endorse this item as being “true”
ferent construction procedures for Scale 5 since he has headaches at least once a month;
will be explored more fully in Chapter 4. another might endorse this item as being
In 1946 Scale 0 (Social Introversion) was “false” since she has headaches only once a
added to the MMPI (Drake, 1946), complet¬ week. The ambiguity inherent in any test item
ing the standard MMPI clinical profile. Scale makes it extremely difficult to obtain a verid¬
0 also was constructed differently from the ical self-description since the person answer¬
other clinical scales. Drake selected MMPI ing a specific test item and an observer rating
items that differentiated 50 college students the person on that item’s content may each
who scored above the 65th percentile on the interpret the item somewhat differently.
Minnesota T-S-E Inventory (Evans & Mc¬ Second, although self-ratings provided
Connell, 1941) from 50 students who scored through item responses can be useful since di¬
below the 35th percentile. rect observations of behavior are often im¬
The Minnesota T-S-E Inventory assesses practical, impossible, or inefficient, indivi¬
introversion-extroversion in three areas: duals vary in their self-awareness and in their
thinking (T), social (S), and emotional (E). ability or willingness to report the appropri¬
Drake limited his initial work to the social ate behaviors. Third, the rational method of
introversion-extroversion area, or, more spe¬ test construction also requires that the test de¬
cifically, he investigated introversion-extro¬ veloper be knowledgeable about the relation¬
version only in the social area as assessed by ship between persons’ responses to individual
the Minnesota T-S-E Inventory. Although items and the construct being assessed. The
Drake conducted his analysis on males and fallacies and errors in earlier rationally de¬
females separately, their norms were so sim¬ rived personality inventories suggest that it is
ilar that he combined the normative data for difficult if not impossible for the test devel¬
the two sexes into a single group before fi¬ oper to have this depth of understanding of
nally incorporating it into the standard the dynamics of a personality inventory.
MMPI profile. (This issue will be explored These problems can be demonstrated by
more fully in Chapter 4.) the response of psychopaths to the MMPI
item “I have been quite independent and free
from family rule.” A test developer would
INTERPRETATION OF INDIVIDUAL
likely make the a priori assumption that psy¬
ITEM CONTENT
chopaths would respond “true” to this item.
Individuals sometimes fail to provide a verid¬ In fact, psychopaths answered this item
ical self-report (one that accurately reflects “false” more often than the normative
how others perceive their behavior) in re¬ groups. This response does not mean that this
sponding to personality inventory items. specific behavior is actually characteristic of
There are several possible reasons for their in¬ psychopaths; rather it means that psycho¬
accurate self-description. First, although per¬ paths say it is characteristic of them. As such
sons constructing test items generally assume it can be treated like any other statement an
that each item has essentially the same mean¬ individual makes. It indicates how the person
ing to all persons taking the test, this is not interprets the statement and how the person
10 Chapter 1

thinks, perceives, and feels even though it or underreport (deny) the behavior being as¬
may actually be untrue. Although this state¬ sessed by the test instrument, or a client may
ment is untrue, it still provides useful diag¬ respond randomly to the test items because of
nostic information about the individual. an unwillingness or inability to respond ap¬
Another example is the response of hy¬ propriately. In either case it is important for
pochondriacs to the MMPI item “I have few the interpreter of the test inventory to be
or no pains.” They answered this item aware of the possibility that the client has re¬
“false” more often than the normative sponded inappropriately. Previous test devel¬
group. Such a response does not necessarily opers often paid lip service to the importance
mean that hypochondriacs actually experi¬ of appropriate test-taking attitudes, but they
ence more pain than other persons, but it did not provide specific directions on how to
does show that they are more willing to say develop or maintain those attitudes. More
that it is true about themselves. important, they did not provide a means of
Although these issues unquestionably assessing whether those attitudes were actu¬
exist in the interpretation of item content on ally present. In the development of the
the MMPI, they do not invalidate it. The em¬ MMPI this problem was directly assessed
pirical approach to item selection used by through what are now called the validity
Hathaway and McKinley, in fact, freed the scales.
test developers of these problems since it as¬ Meehl and Hathaway (1946) were con¬
sumes that the client’s self-report is just that vinced of the necessity of assessing two
and makes no a priori assumptions about the dichotomous categories of test-taking atti¬
relationships between the client’s self-report tudes: defensiveness (“faking-good”) and
and the client’s behavior. Items are selected plus-getting (“faking-bad”).4 (These two cat¬
for inclusion in a specific scale only because egories will be called “underreporting” and
the criterion group answered the items differ¬ “overreporting” of psychopathology, re¬
ently than the normative groups irrespective spectively, throughout later sections of this
of whether the item content is actually an ac¬ book to avoid the connotations inherent in
curate description of the criterion group. Any the terms of “faking-good” and “faking-
correlates between clients’ responses to a bad,” since it is not always clear whether the
given item or scale and their behavior must be person’s motivation for distorting responses
demonstrated empirically. The interested is conscious or unconscious.)
reader should consult Meehl’s (1945) article, To assess these two categories of test¬
which explores this issue in greater depth, and taking attitudes, Meehl and Hathaway con¬
the section on critical items in Chapter 5. sidered three possible approaches. First, they
could give the client an opportunity to distort
the responses in a specific way and observe
ASSESSMENT OF TEST-TAKING
the extent to which the client did so. One way
ATTITUDES
of implementing this approach would be to
In addition to the accuracy of the self-report repeat items within the MMPI, phrased either
in reflecting a client’s actual behavior, the identically or in the negative rather than the
honesty or frankness with which the client at¬ affirmative. A large number of inconsistent
tempts to respond to the items is also impor¬ responses would suggest that the client was ei¬
tant. It is possible that a client might adopt a ther unable or unwilling to respond consis¬
test-taking attitude other than that desired by tently. Although Meehl and Hathaway re¬
the test developer. A client may decide, for jected this solution, the old MMPI group
whatever reason, to overreport (exaggerate) booklet form included 16 identically repeated
 The Evolution of the MMPI 11

items that could be used to detect inconsistent vided another reason for its use as a validity
responding (see the TR [Test-Retest] Index scale.
section in Chapter 3); however, these 16 items Third, Meehl and Hathaway considered
have been deleted in the MMPI-2 (see below). using an empirical procedure to identify
Second, Meehl and Hathaway consid¬ items that elicit different responses from per¬
ered providing an opportunity for the client sons taking the test in an appropriate fashion
to answer favorably when a favorable re¬ and those who have been instructed to
sponse would almost certainly be untrue. “fake” psychopathology. Gough’s Dissimu¬
This solution would involve developing a list lation scale (Gough, 1954, 1957), which was
of extremely desirable but very rare human based on this procedure, will be described in
qualities. If a client endorsed a large number Chapter 3.
of these items, it is highly probable that the Meehl and Hathaway adopted a variant
responses would be dishonest. The L (Lie) of this third approach in developing a third
scale was developed specifically for this pur¬ validity scale, the K scale. Their task was to
pose. Items for the L scale, based on the work differentiate abnormal persons who were
of Hartshorne and May (1928), reflect behav¬ hospitalized and yet obtained normal profiles
iors that, although socially desirable, are all from normal individuals who for some rea¬
rarely true of a given individual. A large son obtained abnormal profiles. They se¬
number of responses in the deviant direc¬ lected 25 male and 25 female patients diag¬
tion on the L scale indicates response dis¬ nosed as having psychopathic personalities,
tortion. alcoholism, and other behavior disorders
The Fscale was developed according to a who (1) had a T score of 60 or higher on the L
variant of this second approach for assessing scale, which would indicate some form of re¬
test-taking attitudes. Items for the F scale sponse distortion, and (2) had diagnoses indi¬
were selected primarily because they were an¬ cating that they should have abnormal pro¬
swered with a relatively low frequency by a files, but (3) had actual profiles in the normal
majority of the original normative group. In range. Based on a comparison of this group
other words, if a client endorsed a large num¬ with the original normative sample on all
ber of the Fscale items, that person would be items, 22 items were selected that showed at
responding in a manner that was atypical of least a 30 percent difference in the response
most people in the normative group. In addi¬ rates of the two groups.
tion, the items include a variety of content It was later found that these 22 items
areas so that any specific set of experiences or generally did an adequate job of identifying
interests for a particular individual would be defensiveness in most patients; however, de¬
unlikely to influence the person to answer pressed and schizophrenic patients tended to
many of the items in the deviant direction. score low. To counteract this tendency, 8
The F scale effectively identified individuals items were added and scored to differentiate
who were intentionally faking pathology; these two groups from the original normative
however, schizoid individuals and persons group. This final step resulted in the 30-item
who were overly pessimistic about themselves K scale, which is currently used. Meehl and
also obtained high scores. Therefore, addi¬ Hathaway also empirically determined the
tional procedures were needed to separate proportions of K that when added to a clini¬
these two groups of persons from those who cal scale would maximize the discrimination
faked their pathology or misunderstood the between the criterion group and the norma¬
items. Meehl and Hathaway thought the L tive group. Since Meehl and Hathaway deter¬
scale would serve this function, which pro¬ mined the optimal weights of K to be added
12 Chapter 1

to each clinical scale in a psychiatric inpatient rized by Pancoast and Archer. First, the
population, they warned that with malad¬ scores of normal individuals may have been
justed normal populations and other clini¬ slightly different from the original Minnesota
cal populations, other weights of K might normative group on the standard validity and
serve to maximize the identification of clinical scales since the MMPI was first devel¬
pathological individuals. This issue of the oped. Second, there have been only small
optimal weights to be added to each clinical changes in normal individuals across five de¬
scale in different populations will be dis¬ cades as reflected by their mean T scores on
cussed in Chapter 3 when the K scale is exam¬ the standard validity and clinical scales.
ined in more depth. Greene (1990) examined the changes in
the standard validity and clinical scales on the
MMPI within four frequently occurring
CURRENT DEVELOPMENTS
codetypes (Spike 4, 2-474-2, 2-7/7-2, and
The issue of whether the items and norms for 6-878-6) in samples of psychiatric patients
the MMPI developed in the early 1940s are over a span of 40 years. The mean and me¬
appropriate for contemporary use has been dian profiles were virtually identical within
raised repeatedly and debated widely (cf. all four codetypes for all four samples as can
Butcher, 1972; Colligan et al., 1983; be seen in Table 1-4. The range in scores
Faschingbauer, 1979). Since the typical indi¬ across all of the clinical scales in ail four sam¬
vidual in the original Minnesota normative ples was 2 T points for the Spike 4 codetypes,
group was “about thirty-five years old, was 4 T points in the 2-474-2 codetypes, 5 T
married, lived in a small town or rural area, points in the 2-777-2 codetypes, and 9 T
had had eight years of general schooling, and points in the 6-878-6 codetypes. The average
worked at a skilled or semiskilled trade (or difference between the highest and lowest
was married to a man with such an occupa¬ score on all of the clinical scales was 2.2, 3.0,
tion level)’’ (Dahlstrom, Welsh, & Dahl- 2.5, and 5.5 T points for these four
strom, 1972, p. 8), it seems apparent that codetypes, respectively. It appears that the
there have been numerous changes in our so¬ MMPI scale scores of psychiatric patients
ciety over the ensuing five decades. have been very stable over this time span.
Pancoast and Archer (1989) collated the Greene’s data did not address whether the
existing literature on the performance of nor¬ empirical correlates of these codetypes re¬
mal individuals on the MMPI to assess the mained unchanged across the 40 years that
adequacy of the norms based on the original the MMPI has been in use. However, the sta¬
Minnesota normative group. The mean bility of the MMPI scale scores across these
MMPI profile for these normal men (Profile years would at least suggest that the corre¬
1-2) and women (Profile 1-3) showed T lates probably have not changed. Of course,
scores near 55 for Scales K, 3 (Hysteria), 4 empirical data are needed to address this
(Psychopathic Deviate), and 9 (Hypomania). question.
Only on Scales L and 1 (Hypochondriasis) The finding that normal individuals and
did the mean T scores approach 50. Pancoast psychiatric patients have shown only minimal
and Archer found that studies as early as changes on the standard validity and clinical
1949 demonstrated that normal individuals scales of the MMPI across 40 years is very
showed generally small, but consistent varia¬ surprising, and would suggest that the MMPI
tions from the mean scores of the original may not be as outdated as many people have
Minnesota normative group. Two conclu¬ thought. The recent work of Colligan and as¬
sions can be drawn from the data summa¬ sociates (1983) in developing contemporary
 The Evolution of the MMPI 13

PROFILE 1-2
NAME_
MINNESOTA MULTIPHASIC™
PERSONALITY INVENTORY
S.R. Hathaway and J.C. McKinley ADDRESS_

PROFILE
OCCUPATION DATE TESTED L±
MINNESOTA MULTIPHASIC PERSONALITY INVENTORY
Copyright THE UNIVERSITY OF MINNESOTA Minnesota Muitiphasic Personality Inventory and MMPI are EDUCATION_ AGE
1943 Renewed 1970 This Profile Form 1948 1976 1982 All lights leseived trademarks owned by The University ot Minnesota
DistiiOuled Exclusively by NATIONAL COMPUTER SYSTEMS INC
Undei License trom The University ot Minnesota MARITAL STATUS__ REFERRED BY
Punted m the United States ot Amenca
1 23456789 o
TorTc L F K Hs+5K D Hy Pd+4K Ml Pa Pt+IK Sc+IK Ma+ Si A R Es MAC'

Various MMPI Normal Samples

Pancoast and Archer (1989)

T score __ 49 53 56 53 54 57 57 59 54 54 53 5£ 50_ NATIONAL

COMPUTER
SYSTEMS
49 item version 27309
 14 Chapter 1

PROFILE 1-3
NAME_
MINNESOTA MULTIPHASIC
PERSONALITY INVENTORY
S.R. Hathaway and J.C. McKinley ADDRESS_

PROFILE
OCCUPATION DATE TESTED

MINNES01A MULTIPHASIC PERSONALITY INVENTORY

Copyright ■ THE UNIVERSITY OF MINNESOTA Minnesota Multiphasic Personality Inventory" and "MtvtPI" are EDUCATION_ AGE
1942 Renewed 1970 This Profile Form 1948 1976 1982 All rights reserved trademarks owned by The University oi Minnesota
Disti itluted Exclusively by NATIONAL COMPUTER SYSTEMS, INC
Undet License from The University of Minnesota MARITAL STATUS, REFERRED BY
Printed in the United Stales of America
1 234567890
TocTc y L F K HS+5K 0 Hy Pd+ 4 K Ml Pa Pt+IK Sc+IK Ma+ 2K Si A fl Es MAC*

Various MMPI Normal Samples

Pancoast and Archer (1989)

T score _ 5Q 52 56 51 53 55 M 47 55 53 53 51 55 NATIONAL
COMPUTER
SYSTEMS
*49 item version 27309
 0)
c
+D
c
ou

CO
CD
■o
CD
O

CD
Q
3
O

CO
CO
o
o
<
CO
<D
Q.

>-1
-t
CD
"O
O

O
D

O
LL
C

£
CD
O

c
CD
E
o
CD
CL

LU
—I
CO
<
H

15
 9(Ma) 0(Si)

CO CO CD CD CD CD CD D D LD LD CO
CD CD CD CO CO CO CO CO CO CO CO CD CD CD CD CD

CO CO CO CO CO CO CO CO CM O ID CO CO CO
ID ID ID D D D D D r-« r-> I"'- r-
8(Sc)

T-
_ - T-
''t CO ^ CO D CO D ^ O T CD CD
l-v C" r-" CD CD CD CD O o CD CD
T
7{Pt)

ID CO D D co D CO D LD T“ CD CD CD
CD CD CD CD CO CD 00 00 r>. 00
5(Mf) 6(Pa)

LD LD D D CO D D O 00 CD 00 ID CM
CD CD CD CO co co co CO CD 00 00 00 CD CD CD CD

CM D CO cor- CO D O CD CD O CD 00
ID LD D D D D D D CO D CO LD CD LD ID LD
4(Pd)

I"' CD r-' O CD o CD CM r- -
T CD CD
CD CD CD co co CO C' 1^
3(Hy)

o co r- CD CM 00 D Tf CD c-
I"" co co co co CO D CO CD CD CD CD CD
2(D)

O CD O CO O CD CD CD O CD o O CD CD co CM
CD CO CD CD CD CO 00 00 r-" co C' 1^
l(Hs)

CD CO CO r" CO D CO D CO o CD O 00 O 00
CD CD CO co co co CO CO CO CO CO CD CD CD

CM CM CD 00 o CD CM CO CD co o ID CM
K

ID LD D D D "vt "3-

CD LD o CO O CO O CD 00 ■zt CD co CM CM
F

ID LD CO co co co CD CO 00 00 CD CD CD CD CD CD

O O CO O CO CD CO r- CO CO O CD CD (D CD
L

LD LD D D ^ ■'t 'Ct D LD ID "sj-

Gender

LD S LL u- ^ U- S u- ^ U- ^ LD 2 Li.
vP ^5 ■sO
p^ p^ c£ ^ 0s
LD LO r-" CO ID LD CD (D ^ i— CD CD >^|- CM 00
ID LD CM !-. LD CD co CM C0 co LD

00 CD CO CD CD CO
Age

d CO CM CM O CO
CO CO Tf CO CO CO CO CM

vP vP Co •sP vp 3
o'- 0s p^ 0s
(%)

CD CD CM LD CO O CM
CM CO CD 00 CO LD CD
CD CD CO CO ■'t CM CD

CM o
T_ CD ID
N

ID 00 CD 00 CM co 00 CM
r— CO CO CD CM
T~ T—

o o o o
ID CD LD CD
o o
CD CD <D CD
Sample

T- <“ o r— '— O
CD CD
CD CD
00 ro 0 r— 00
+-< c ■M
o
c c CD c +-■ c
O
C c CD c
0 § c 0 T- c CD
C- 0
CD CO CD ° r
° c C/5 c .5 § c r“ c , 0 </) c .5
I
IN CD CO TO CD CD TO
o ro TO c/3 CD TO m 0 cd TO 0 0 ~o o TO (f) 0 T3
CD c CD c 0 0 S 2 0 CD 2 0 £0 c 0 0 c 0 0 0 CD 0 CD
rs C ^ C ^ ^ CO c ^ X
< X S
UJ E ^ ^ CO S 2
CD
h- fN 1- CO ^ (-

16
 The Evolution of the MMPI 17

norms for the MMPI and the current profile (and the profile for the mean plus two
restandardization of the MMPI that has re¬ standard deviations [i.e., the profile equiva¬
sulted in the MMPI-2 also have examined the lent to a T score of 70]) for these contempo¬
changes that have occurred since the MMPI rary men and women plotted on the original
was developed originally. MMPI norms. As can be seen in these two
profiles, the men average 3 to 8 T score points
higher on the clinical scales and the women
A Contemporary Normative Study
average 1 to 6 T score points higher (except
of the MMPI
on Scale 5 [Masculinity-Femininity] where
Colligan and colleagues (Colligan et ah, they are 4 points lower) on the clinical scales
1983, 1989; Colligan, Osborne, Swenson, & than the original Minnesota normative
Offord, 1984) investigated whether the origi¬ group. The men’s profile equivalent to a T
nal MMPI norms were appropriate for con¬ score of 70 averages 2 to 7 T points higher on
temporary use. They essentially replicated the the clinical scales, whereas the women’s pro¬
data-collection procedures employed by Mc¬ file equivalent to a T score of 70 is fairly sim¬
Kinley and Hathaway (1940) and gathered a ilar to the original Minnesota normative
representative sample of individuals living group.
within 50 miles of the Mayo Clinic in Roches¬ Two basic points can be made based on
ter, Minnesota. “Persons having chronic dis¬ the data presented in Profiles 1-4 and 1-5.
eases (for example, diabetes) were excluded First, there are some differences in MMPI
from the study, as were patients receiving performance across the five decades that the
cancer treatment, those with rheumatoid or MMPI has been in use, although these differ¬
other types of arthritis, those described as ences are not as substantial as might have
being chemically dependent, having a learn¬ been expected, given the changes in our soci¬
ing disability, or being mentally retarded, ety in the last 50 years. If Profiles 1-4 and 1-5
and persons undergoing psychotherapy” are compared with the data reported by
(Colligan et al., 1983, pp. 74-75). Pancoast and Archer (1989) in Profiles 1-2
Their final sample consisted of 1,408 and 1-3, respectively, further support can be
white individuals (646 men and 762 women), provided for the statement that the scores of
whose mean age was in their mid-40s and normal individuals may have been slightly
who had a mean of 13 years of education. different from the original Minnesota norma¬
Nearly three-fourths of them were married. tive group on the standard validity and clini¬
These individuals were somewhat older and cal scales since the MMPI was first devel¬
better educated than the original Minnesota oped. Second, it appears that these changes
normative sample that has been described average less than one-half standard deviation
above. Colligan and associates also selected a (5 T points) and these small changes in profile
subset of these individuals “in proportion to elevation are not likely to have major impact
the age and sex in the general population of on the clinical interpretation of the MMPI.
adult whites in the United States, as deter¬ Colligan and colleagues (1983) contin¬
mined by the 1980 census” (1983, p. 87) so ued the procedure of using K-corrected scores
that they could make more direct compari¬ and they used the same correction weights on
sons with the original normative group since the same clinical scales (Scales 1 [Hypochon¬
the population of the United States had in¬ driasis], 4 [Psychopathic Deviate], 7 [Psych-
creased in age and become better educated in asthenia], 8 [Schizophrenia], and 9 [Hypoma-
the ensuing four decades. nia]) that had been suggested by Meehl and
Profiles 1-4 and 1-5 provide the mean Hathaway (1946). However, they made one
 18 Chapter 1

PROFILE 1-4

•\j fi'i r r > MINNESOTA MULT1PHAS1C

PERSONALITY INVENTORY
S.R. Hathaway and J.C. McKinley
NAME_

ADDRESS_

PROFILE
OCCUPATION DATE TESTED LI
MINNESOTA MUl MPHASIC PERSONALITY INVENIORY
Cnpv’iqlH I HE UNIVERSITY OF MINNESOTA Minnesota Multiphasic Personality Inventory' and MMPI are
EDUCATION_AGE,
IXi Renewed 1970 I his F‘i ofilo Form 1948 1976 1982 All rights ieserved trademarks owned by The University ot Minnesota

DisinbuleO Exclusively by NATIONAl COMPUTER SYSTEMS INC

Undei License tiom The Umveisily ol Minnesota
MARITAL STATUS_ REFERRED BY _
Pnnied m the United States ol Ainenca
1 234567890
fOf Tc "> L F K Hs+5K D Hy Pd+4K Ml Pa Pt+IK Sc+IK Ma+ 2K Si A R Es MAC*

T score _ 49 54 54 54 54 57 54 58 55 54 53 55^ 52_ NATIONAL

computer
SYSTEMS
*49 item version 27309
 The Evolution of the MMPI 19

PROFILE 1-5

MINNESOTA MULTIPHASIC PERSONALITY INVENTORY

Copyright THE UNIVERSITY OF^MINNESOTA Minnesota Multiphasic Personality Inventory and 'MMPI are EDUCATION_ AGE.
1943 Renewed 1970 'his Profile Form 1948 1976 1982 All rights reserved trademarks owned by The University ol Minnesota
Distributed Exclusively by NATIONAL COMPUTER SYSTEMS INC
Undei License from The Umveisity ol Minnesota MARITAL STATUS. REFERRED BY
Punted in the United Stales ot America
1 234567890

Contemporary Normative Sample

Colligan and Associates (1983)

Mean (T 50)

Mean + 2 SD (T70)

1 234567890

T score ___ 50 52 55 51 54 53 53 45 56 53 54 51 56_ NATIONAL

COMPUTER
SYSTEMS
*49 item version 27309
20 Chapter 1

major change when they decided to employ Over the ensuing years Colligan and col¬
normalized T scores. The linear transforma¬ leagues have provided contemporary norms
tions of raw scores into T scores that are used for Barron’s (1953) Ego Strength scale (Col¬
on the standard MMPI profile sheet assume ligan & Offord, 1987a), Welsh’s (1956) Anxi¬
an underlying normal distribution that is not ety and Repression scales (Colligan &
typical for most of the scales since T scores Offord, 1988a), Wiggins’ (1966) Content
greater than 69 occur quite frequently. The scales (Colligan & Offord, 1988b), and the
normalized T scores that were developed by F - K index (Osborne, Colligan, & Offord,
Colligan and colleagues produced a normal 1986).
distribution for each scale in their normative Tables for converting raw scores into T
sample. The issue of linear versus normalized scores so that the clinician can compare a
T scores will be explored further in the next client’s performance with a contemporary
chapter. adult sample are available in Colligan and as¬
There has been only limited research sociates (1983). Hsu and Betman (1986) have
with the Colligan and associates (1983) provided tables for converting the T scores
norms. Colligan, Osborne, Swenson, and for the original MMPI normative group into
Offord (1985) reported the frequency with Colligan and colleagues’ (1983) contempo¬
which codetypes occurred in four clinical rary norms and vice versa. Colligan and col¬
samples, and the concordance between their leagues (1983) also illustrate a standard pro¬
contemporary norms and the original MMPI file sheet for use with their contemporary
norms. Concordance of codetypes between norms (p. 421).
the two sets of norms ranged from 40 to 60
percent for women and from 50 to 70 percent
for men, whereas agreement on single scales
ranged from 66 to 79 percent for women and
The MMPI-2
from 69 to 79 percent for men.
Miller and Streiner (1986) reported the The MMPI-2 (Butcher et al., 1989) represents
concordance between profiles generated by the restandardization of the MMPI that
contemporary norms and the original MMPI marks the advent of a new era of clinical
norms in a large sample of psychiatric pa¬ usage and research of this venerable inven¬
tients. They found that 48.4 percent of the tory. Restandardization of the MMPI was
profiles showed no changes in the two highest needed to provide current norms for the in¬
clinical scales, and another 15.1 percent of ventory, develop a nationally representative
the profiles had the two highest clinical scales and larger normative sample, provide appro¬
reversed. Thus, 63.5 percent of the profiles priate representation of minority groups, and
had the same codetype using the two sets of update item content where needed. Continu¬
norms. In 23.6 percent of the profiles, the ity between the MMPI and the MMPI-2 has
highest clinical scale remained the same while been maintained since new criterion groups
another clinical scale became the second and item derivation procedures were not used
highest scale. A totally unique codetype was on the standard validity and clinical scales.
produced in 9.4 percent of the profiles. Al¬ Thus, the items on the validity and clinical
though it is important to know the concor¬ scales of the MMPI are essentially unchanged
dance between codetypes generated by the on the MMPI-2 except for the elimination of
two sets of norms, the primary issue remains 13 items based on item content (see Table
whether the original or contemporary norms 1-5) and the rewording of 68 items.
more accurately reflect external correlates. The profile forms for the original MMPI
 The Evolution of the MMPI 21

TABLE 1-5 Thirteen Items Dropped from the Standard Validity and Clinical Scales

FScale
1 4. I have looseness in my bowels (diarrhea) once a month or more.
53. A minister can cure disease by praying and putting his hand on your head.
206. I am very religious (more than most people).
258. I believe there is a God.

Scale 7 (Hs)
63. I have had no difficulty in starting or holding my bowel movements.

Scale 2 (D)
58. Everything is turning out just as the prophets of the Bible said it would.
95. I attend religious services almost every week.
98. I believe in the second coming of Christ.

Scale 5(Mf)
69. I am very strongly attracted by members of my own sex.
70. I used to like drop-the-handkerchief.
249. I believe there is a Devil and a Hell in afterlife.
295. I liked "Alice in Wonderland" by Lewis Carroll.

Scale 0 (Si)
462. I have had no difficulty starting or holding my urine.

Note: Reproduced from the MMPI by permission. Copyright © 1 943, renewed 1 970 by the Uni¬
versity of Minnesota. Published by the University of Minnesota Press. All rights reserved.

(Profile 1-1) and the MMPI-2 (Profile 1-6) change their item-scale correlations in most
also are virtually identical. A quick compari¬ cases (Ben-Porath & Butcher, 1989). The
son of Profile 1-1 and 1-6 will not reveal any Restandardization Committee then added
readily apparent differences between the two 154 provisional items that resulted in the 704
forms. Only on closer examination are any items in Form AX, which was used to collect
differences seen on the MMPI-2 profile the normative data for the MMPI-2.
form: the Cannot Say (?) scale has been When finalizing the items to be included
moved to the bottom of the page, T scores of on the MMPI-2, the Restandardization Com¬
65 are considered to be clinically significant mittee deleted 77 items from the original
instead of T scores of 70, and the T score dis¬ MMPI in addition to the 13 items deleted
tributions have been truncated at 30 so that T from the standard validity and clinical scales
scores below 30 do not occur. and the 16 repeated items. Consequently,
In the development of the MMPI-2, the most special and research scales that have
Restandardization Committee (Butcher et been developed on the MMPI are still capable
al., 1989) started with the 550 items on the of being scored unless the scale has an em¬
original MMPI. They reworded 141 of these phasis on religious content or the items are
550 items to eliminate outdated and sexist drawn predominantly from the last 100 items
language and to make these items more easily on the original MMPI. The content areas for
understood. Rewording these items did not these 77 items that were not retained plus the
 22 Chapter 1

PROFILE 1-6
Name Cherie Jones

MMPI-2
s K I l.ilIi.iw.i\ ,iiuI I t Mi Kink*\

Hhi/h'.K'ht Address 2894 Albion Way

ffr.nvur/Sty /mt’/tAvy - J
Occupation Secretary Date Tested 3/ 4/91
Profile for Basic Scales
Minnesota Muiiiphasic Personalii\ Inventors-2
Education 13th Age 27 Marital Status Single
( opwight 1 b\ llll RICil NTS OK III! I NIYKRSITY OK MINNESOTA
km; km; I renewed KJ7H). 1989. This Profile Korm 1989.
\ll rights reserved Distrihined e\eltisi\ el> h\ NATION A l COMPITKR SYSTKMS. INC
Referred by Dr. Smiley
under lieense from The l ni\ersil\ of Minnesota.

\1\1PI-; ;md ''Minnesota Mtiltiphasie Personalitv Inventon-2" are trademarks owned b\ MMPI-2 Code 26’43-1_078/59 F-KL
1 he l nix ersit\ of Minnesota. Printed in the l niled States oi America.
Hs- 5K Pd- 4K Pi-IK Sc+IK Ma+ 2K
1 4 7 8 9 o t or Tc Scorer's Initials HG

.
• In 12 ■
1 11 6
11 5
■

. in
IS 9 :

1 j!
Is 8 6 3
1-1 6 3
13 7 9 3
:: • 5 2
11 ■ 4 2

in 5 4 2
• 4 2
8 4 .1 2

6 3 2 l

s 3 2 i
4 2 2 l
3 2 i :
: 1 i 0
1 i 0 0
0 0 0 0

Raw Score _4 ^ 16 ^ 32 27 23 38 16 14 12 14 36 NATIONAL

Raw Score 0 COMPUTER
K to be Added 8 _6 16 16 SYSTEMS

Raw' Score with K 17 26 30 28 17 24001

 The Evolution of the MMPI 23

13 items deleted from the standard validity TABLE 1 -7 Changes in Items from the MMPI
and clinical scales can be seen in Table 1-6. to the MMPI-2
Levitt (1990) also has grouped these 77 items
Number
into logical content categories and provided
of Items
the actual items within each category.
The Restandardization Committee in¬ MMPI 566
cluded 68 of the 141 items that had been re¬
Drop 1 6 repeated items -16
written, and they incorporated 107 of the
Drop 1 3 items from the standard
provisional items to assess major content validity and clinical scales -13
areas that were not covered in the original Drop 77 items from the last
MMPI item pool. The rationale for including 1 67 items -77
and dropping items from Form AX that re¬ TOTAL 460
Add 89 items for the new
sulted in the 567 items on the MMPI-2 has
content scales + 89
not been made explicit to date. Table 1-7 il¬ Add 1 8 unscored items + 18
lustrates the changes that were made in the
transition from the 566 items on the MMPI to MMPI-2 567
the 567 items on the MMPI-2.
The MMPI-2 was standardized on a
sample of 2,600 individuals who were se¬ minority individuals, and occupational sta¬
lected from seven different states (California, tus. The individuals in the normative sample
Minnesota, North Carolina, Ohio, Pennsyl¬ for the MMPI-2 also are more representative
vania, Virginia, and Washington) to reflect of the United States as a whole since national
national census parameters on age, marital census parameters were utilized in their col¬
status, ethnicity, and so on. The demo¬ lection. However, they still varied from the
graphic characteristics of this sample can be census parameters on years of education and
seen in Table 1-8. occupational status. The potential impact of
The normative sample for the MMPI-2 the relatively high level of education and oc¬
varies significantly from the original norma¬ cupational status characteristic of the MMPI-
tive sample for the MMPI in a number of 2 normative group on the standard validity
areas: years of education, representation of and clinical scales and codetype interpreta¬
tion remains to be determined empirically.
Profiles 1-7 and 1-8 illustrate the aver¬
TABLE 1-6 Content Areas of MMPI Items age scores for men and women on the stan¬
Not Retained on the MMPI-2 dard validity and clinical scales in the MMPI-
2 normative sample when plotted on the
Content Area Number of Items
original MMPI norms. The MMPI-2 norma¬
Interests/hobbies 17 tive sample scores about 3 to 5 T points
Religion 16 higher on most of these scales. Only on Scales
Interpersonal relationships 14
L, 1 (Hypochondriasis), and 0 (Social Intro¬
Negative affects 12
Bodily functions 9 version) are their scores nearly identical.
Miscellaneous 5 Thus, the transition to the MMPI-2 norms
Sexuality 5 will mean that the new profiles are slightly
Sensory functions 4 less elevated when compared to the original
Substance abuse 3
MMPI norms. It also should be noted that
Blushing 3
Dreaming 2 the average scores for the standard validity
TOTAL 90 and clinical scales for the MMPI-2 corre¬
spond very closely to those reported by Col-
 24 Chapter 1

PROFILE 1-7
NAME
MINNESOTA MULTIPHASIC1
PERSONALITY INVENTORY
S.R. Hathaway and J.C. McKinley ADDRESS_

PROFILh
OCCUPATION. _DATETESTED L±
MINNl Si'’ A Ml‘l IIPHASK' PI RSONAL11Y INVENTORY
Copv- am THt UNIVERSI1V OF MINNESOIA Minnesota Mulliphasic Personality Inventory and MMPI are
EDUCATION AGE
194.1 Renewed l9/'0 1 Ins Piofile Form 194# 1976 1982 All nghls leseivecl trademarks owned by The University o( Minnesota

DiStiibviU'il t xrltisivt'ly Dv NA1 IONAl COMPUTER SYSTEMS INC

Under license tioni I he Univeisity ol Minnesota
MARITAL STATUS REFERRED BY
Punted in the United States ol Ammo
3 4 7 8 9
TofTc 1 l F K Hy Pd+ 4 K Pt+IK Sc+1 K Ma+2K

MMPI-2 Normative Sample

Butcher and Associates (1989)

T score 43 56 56 54 58 53 58 64 56 57 58 58 52 NATIONAL
COMPUTER
SYSTEMS
*49 item version 27309
 The Evolution of the MMPI 25

PROFILE 1-8

MINNtSOIA MULTIPHASIC PERSONALITY INVENTORY

Copyright THE UNIVERSITY OF MINNESOTA Minnesota Mulliphasic Personality Inventory and MMPI' are EDUCATION_ AGE.
-*43 Renewed 1970 T(113 Profile Form 1948 1976 198? All rights reserved trademarks owned by The University ol Minnesota
Distntu.'ind Exclusively by NATIONAL COMPUTER SYSTEMS INC
Undei License from The University of Minnesota MARITAL STATUS. REFERRED BY
Pnnlei m the United Stales of Amenca
1 234567890
TorTc 7 L F K Hs+5K 0 Hy Pd4 4K Ml Pa Pt+1K Sr.+lK Ma+ 2K Si A R Es MAC*

MMPI-2 Normative Sample

Butcher and Associates (1989)

T score _ 48 54 55 52 54 55 57 46 57 54 56 55 54_ NATIONAL

COMPUTER
SYSTEMS
*49 item version 27309
26 Chapter 1

TABLE 1 -8 Demographic Variables for the MMPI-2 Normative Sample

Variable N Percentage Census Variable N Percentage Census

Gender Education
Female 1462 56.2% Part high school 129 5.0% 33.5%
Male 1 1 38 43.8 High school 640 24.6 34.4
Part college 651 25.0 1 5.7
College graduate 700 26.9 8.7
Ethnicity Postgraduate 480 18.5 7.8
White 21 1 7 81.4% 85.0%
Black 314 12.1 10.5
Native-American 77 3.0 0.5 Marital Status
Hispanic 73 2.8 — Married 1717 66.0% 59.0%
Asian-American 19 .7 1.5 Never married 518 19.9 25.9
Other 2.5 Divorced 220 8.5 5.7
Widowed 89 3.4 7.2
Separated 56 2.2 2.3
Age
18-19 50 1.9% 5.4%
20-29 641 24.7 25.2 Occupation
30-39 769 29.6 19.3 Professional 1060 40.8% 1 5.6%
40-49 401 1 5.4 14.0 Managerial 277 10.7 1 1.2
50-59 321 1 2.3 14.3 Skilled 215 8.3 2 5.5
60-69 277 10.7 1 1.5 Clerical 365 14.0 27.2
70-79 120 4.6 7.5 Laborer 205 7.9 20.6
80-89 21 .8 3.1 None of the above 463 1 7.8 —

(Missing data) 15 .6 —

Note: Adapted from Butcher et al. (1 989).

ligan and associates (1983). (Profile 1-4 scales. It does not appear that the relatively
should be compared to Profile 1-7 and Pro¬ high level of education and occupational sta¬
file 1-5 to Profile 1-8.) tus characteristic of the MMPI-2 normative
It also is possible to compare the average group adversely affected scores on the obvi¬
scores of the MMPI-2 normative group on ous subscales since these scores are very sim¬
the Wiener and Harmon (Wiener, 1948) Ob¬ ilar to the original Minnesota normative
vious and Subtle subscales (see Chapter 3 for group. Again, it can be noted that the
a discussion of these subscales) with the orig¬ scores of the MMPI-2 normative group on
inal Minnesota normative group (see Profiles the Wiener and Harmon Obvious and Sub¬
1-9 and 1-10). It is readily apparent in Pro¬ tle subscales correspond very closely to
files 1-9 and 1-10 that the MMPI-2 norma¬ those reported by Colligan and associates
tive group and the original Minnesota norma¬ (1983).
tive group have almost identical scores on the The /C-correction procedure, which will
obvious subscales (excluding Ma-O), whereas be described in detail in Chapters 2 and 3, will
their scores are very different on the subtle continue to be used with the MMPI-2 with
subscales (excluding D-S). Thus, the differ¬ the same /f-weights being used on the same
ences between the MMPI-2 normative group clinical scales. However, in a major depar¬
and the original Minnesota group in Profiles ture from the MMPI, the MMPI-2 will have
1-7 and 1-8 predominantly reflect the influ¬ raw scores converted to uniform T scores for
ence of the subtle items on these five sub¬ all clinical scales except Scales 5 (Masculinity-
 The Evolution of the MMPI 27

PROFILE 1-9

MINNESOTA MU LTIP H ASIC'M

name IVIIVIPI -2 Normative Sample
PERSONALITY INVENTORY
S.R. Hathaway and J-C. McKinley ADDRESS

WIENER-HARM ON SUBTLE-OBVIOUS
SUBSCALES PROFILE OCCUPATION DATF TFSTFD / /
MINNESOTA MULTIPHASIC PERSONALITY INVENTORY
Copyrighl F THE UNIVERSITY OF MINNESOTA EDUCATION AGE
1943. Renewed 1970 This Prolile Form 1948 <976. 1982. 1986 All rights reserved Minnesota Multiphasic Personality Inventory'' and MMPI' are
trademarks owned by The University of Minnesota
Distributed Exclusively by NATIONAL COMPUTER SYSTEMS INC
Under License from The University ol Minnesota
Printed in the United States ol America
MARITAL STATUS_REFERRED BY

MMPI-2 Normative Sample

Butcher and Associates (1989)

Scorers
Initials_

Raw Score 9 11 _6 16 5 11 T 7 6 10 NATIONAL

COMPUTER
SYSTEMS
 28 Chapter 1

PROFILE 1-10
NAME.
MINNESOTA MULTIPH ASIC"
PERSONALITY INVENTOR V
S.R. Hathaway and J.C. McKinley ADDRESS.

W1ENHR-HARMON SL’BTLE-OBVIOUS
SUBSCALES PROFILE OCCUPATION. .DATE TESTED.

MINNESOTA MUITIPHASIC PERSONALITY INVENTORY

Copyright ■ THE UNIVERSITY OF MINNESOTA EDUCATION. AGE
1943 Renewed 1970 This Prolile Form 1948. 1976. 1982. 1986 All righls reserved. Minnesota Multiphasic Personality Inventory" and MMPI" are
trademarks owned by The University of Minnesota
Distributed Exclusively by NATIONAL COMPUTER SYSTEMS. INC
Under License Irom The University of Minnesota
MARITAL STATUS. REFERRED BY
Printed in the United States ol America

MMPI-2 Normative Sample

Butcher and Associates (1989)

Scorers
Initials_

Raw Score NATIONAL

COMPUTER
SYSTEMS
 The Evolution of the MMPI 29

Femininity) and 0 (Social Introversion). tors. These new content scales will be de¬
Scales 5 and 0 will retain their linear T scores. scribed in Chapter 5.
The uniform T scores for the clinical scales Since uniform T scores change the rela¬
were developed based on the composite dis¬ tionships among the clinical scales, the rela¬
tribution for each clinical scale by gender so tive frequencies with which high-point scales
that the percentiles will be equivalent across and codetypes occur will be changed. Con¬
the scales. Thus, a uniform T score of 75 on cordance rates for codetypes between the
Scale 2 (Depression) will be equivalent to a T MMPI-2 and MMPI are reported to range
score of 75 on Scale 6 (Paranoia). The issues from 57 to 70 percent in clinical samples
involved in the change to uniform T scores (Butcher et al., 1989), and appear to be sim¬
will be discussed in Chapter 2. ilar to those reported for the Colligan and as¬
A number of supplementary scales are sociates (1983) norms when compared to the
available to assist the clinician in interpre¬ original MMPI norms. This issue will be ex¬
ting the standard validity and clinical scales on plored in more depth in Chapter 6, which also
the MMPI-2. Most of these supplementary will report the specific codetype concordance
scales will be familiar to the clinician who has rates between the MMPI-2 and MMPI.
used the MMPI: Barron’s (1953) Ego Strength An important question about the
scale; Gough, McClosky, and Meehl’s (1951) MMPI-2 will involve whether the correlates
Dominance scale; Gough, McClosky, and of the individual scales and the codetypes
Meehl’s (1952) Social Responsibility scale; that were derived on the MMPI can be ap¬
Kleinmuntz’s (1961a) College Maladjustment plied to the MMPI-2. For example, if a client
scale; MacAndrew’s (1965) Alcoholism scale; has a 2-4/4-2 codetype on the MMPI-2, can
Megargee, Cook, and Mendelson’s (1967) the clinician use the correlates of a 2-4Z4-2
Overcontrolled Hostility scale; and Welsh’s codetype that were developed on the MMPI
(1956) Anxiety and Repression scales. These for interpreting the profile? An associated
scales have had a few minor changes made in question is even more difficult to answer. If
them at the item level. Seven new scales are the client has a 2-8/8-2 codetype and would
represented in the supplementary scales: two have had a 2-4/4-2 codetype if linear rather
gender-role scales (Gender Role-Femininine; than uniform T scores had been used, are cor¬
Gender Role-Masculine), two Post-trauma- relates of the former or latter codetype, or
tic Stress Disorder scales (Post-Traumatic even some other codetype, more appropriate
Stress Disorder—Keane; Post-Traumatic for this client? Clearly such questions require
Stress Disorder—Schlenger), and three valid¬ an empirical answer. Studies of the correlates
ity scales (Back F; True Response Inconsis¬ of the MMPI-2 will begin to appear over the
tency; Variable Response Inconsistency). The next few years once it becomes readily avail¬
three new validity scales will be discussed in able to clinicians. In the interim, clinicians
Chapter 3; the remainder of the supplemen¬ will need to be very cautious in using MMPI
tary scales will be reviewed in Chapter 5. correlates to interpret MMPI-2 profiles.
A total of 15 new content scales have At this time the MMPI-2 is to be used
been developed for the MMPI-2 (Butcher et only with adults 18 years of age and older,
al., 1989). These scales are: Anxiety, Fears, since normative data do not exist for persons
Obsessiveness, Depression, Health Concerns, under 18 years of age. Adolescents are still to
Bizarre Mentation, Anger, Cynicism, Antiso¬ be tested with the original MMPI. Work is
cial Practices, Type A, Low Self-Esteem, So¬ currently underway to develop a restandardized
cial Discomfort, Family Problems, Work version of the MMPI for use with adoles¬
Interference, and Negative Treatment Indica¬ cents, the MMPI-A, which should be avail-
30 Chapter 1

able in late 1991 or 1992. The MMPI-A will codetypes is needed to address this critical
be a new form designed specifically for ado¬ issue.
lescents, although it will retain the standard
validity and clinical scales of the original ENDNOTES
MMPI. 1. Hathaway and McKinley did not provide
a rationale for deleting insignificant items. Al¬
though potentially useful items may have been dis¬
A Note on Research with the MMPI-2
carded, this procedure was acceptable at the time
The advent of the MMPI-2 may have a quick since they used an empirical method of item selec¬
appeal to researchers who would like to re¬ tion. The issue of their rationale for deletion of
port the relationships among the new MMPI- items, however, has become more relevant since
item content is sometimes important in current
2 norms, the Colligan and associates (1983)
usage of the MMPI. Wiggins’ (1966) content
norms, and the original MMPI norms
scales are an example.
(Hathaway & Briggs, 1957). The important
2. It is now customary to identify each scale
issue with these three sets of norms is not by its number rather than its name. The use of the
whether they produce different elevations on scale number reduces the emphasis placed on diag¬
the various scales and/or codetypes, since nostic labels like hypochondriasis, schizophrenia,
such differences would be expected or there and so on, and encourages the clinician to be
would be no reason for their publication. aware of the empirical correlates of specific scores
Rather, the issue is whether the behavioral on each scale.
and clinical correlates are more related to any 3. The addition of 55 items to the original
specific set of norms. Unfortunately, the lat¬ 504 items on the MMPI would produce an item
ter research is more difficult and time-con¬ pool of 559 items. Since the MMPI contains only
550 items, it is not clear what happened to the
suming than the former, which requires only
other 9 items (W. G. Dahlstrom, personal commu¬
that the researcher calculates three sets of
nication, 1979).
norms for a given MMPI-2 and reports their
4. The term plus-getting describes the pro¬
concordance rates. Hopefully, researchers cedure of making a deviant response to an item on
and journal editors will realize that a careful a scale, thus adding plus one to the total score on
investigation of the clinical correlates of these the scale.
various sets of norms within specific
 CHAPTER 2

A dministration
and Scoring

Administering and scoring the MMPI-2 are book addresses issues in the administration
usually straightforward procedures that can of the MMPI, the clinician should realize that
be handled by a competent psychometrician. the suggestions also are appropriate for the
The apparent ease of MMPI-2 administration MMPI-2.
sometimes leads clinicians to underestimate Once administered, the MMPI-2 can be
the importance of establishing appropriate scored either by hand or by computer. This
conditions for taking the MMPI-2, clarifying chapter will describe the procedures for scor¬
the test instructions if necessary, and unob¬ ing and profiling the MMPI-2; examples of a
trusively monitoring the client’s progress. clinician’s interpretation of the MMPI-2 pro¬
Occasionally clinicians inappropriately rele¬ file and several computer interpretive services
gate the task of MMPI-2 administration to a will be provided in Chapter 7. This chapter
secretary or clerk, who may administer the also will address several related issues, such
test incorrectly. The ease of MMPI-2 admin¬ as the various forms of the MMPI-2; T score
istration does not absolve the clinician of the derivation; linear, normalized, and uniform
responsibility for insuring that it is handled T scores; and the effects of various demo¬
properly. graphic variables on MMPI performance.
Before administering the inventory for
the first time, the clinician should read the
ADMINISTRATION
MMPI-2 Manual (Butcher, Dahlstrom, Gra¬
ham, Tellegen, & Kaemmer, 1989) and Chap¬ Reading level is a crucial factor in determin¬
ter 1 of An MMPI Handbook (Dahlstrom, ing whether or not a person can complete the
Welsh, & Dahlstrom, 1972). The Handbook, MMPI-2; inadequate reading ability is a
the definitive reference on the MMPI, is par¬ major cause of inconsistent patterns of item
ticularly useful when the clinician anticipates endorsement. Butcher and colleagues (1989)
any unusual circumstance in administration suggest that most clients who have had at
or whenever a more general text omits an¬ least eight years of formal education can take
swering any question. Although the Hand¬ the MMPI-2 with little or no difficulty since

31
32 Chapter 2

the items are written on an eighth-grade level reading skills or an IQ below 70, the clinician
or less. The clinician should note this eighth- need not automatically abandon the idea of
grade reading level on the MMPI-2 since the administering the MMPI-2, since such per¬
MMPI was considered to have a sixth-grade sons sometimes can complete the inventory if
reading level. However, Ward and Ward it is presented orally. An audiotaped version
(1980) found that readability for some MMPI of the MMPI-2, available from National
scales reached the seventh-grade level, and Computer Systems (NCS), P.O. Box 1416,
Blanchard (1981) reported that nine years of Minneapolis, MN 55440 (800-627-7271),
education was necessary for a criterion of 90 serves this purpose. Dahlstrom and associates
percent comprehension of the MMPI items. (1972) reported that taped administrations of
This issue of the reading level required to the MMPI were effective with IQs as low as
complete the MMPI-2 becomes more serious 65 and reading levels as low as the third
when it is realized that most freshman-level grade.
college texts are written at the ninth-grade Few explicit guidelines exist regarding
level. Despite the increased years of educa¬ how much assistance the clinician can pro¬
tion of most people, reading level can still be vide for a client who has marginal reading
a potential problem. Clients with less educa¬ skills. It seems reasonable to give standard
tion may be able to take the MMPI-2 if their dictionary definitions of terms if the client
reading level is adequate. If there is any rea¬ asks. The clinician should refrain, however,
son to suspect that a person’s reading level from administering the MMPI-2 by reading
may be deficient, the clinician should ascer¬ the items aloud since this practice is a signifi¬
tain the person’s reading level, administering cant change from standardized administra¬
a brief reading test such as the Gray Oral tion and its effects are unknown. Hopefully,
Reading Test (Gray & Robinson, 1963) if nec¬ future research will compare the responses
essary. Reading level becomes especially im¬ obtained when the clinician reads items aloud
portant among minority individuals who, de¬ with responses obtained in the standard ad¬
spite their fluency in speaking English, may ministration. Until such research is con¬
be unable to read English. Anyone who can¬ ducted, the clinician should consider admin¬
not read may be defensive about revealing istering some other personality instrument to
this deficiency; at times persons have re¬ persons with limited intelligence or reading
sponded to all 567 items even though they ability unless standard definitions provide
were unable to read them. adequate clarification or the taped version
Although less important than reading can be used.
level, a person’s age and intelligence also af¬ Psychiatric impairment rarely precludes
fect ability to complete the MMPI-2. The taking the MMPI-2 unless the client is agi¬
Manual (Butcher et al., 1989) states that the tated and unable to sit still long enough to
MMPI-2 can be given to clients 18 years and complete the test. Clients who are severely
older. Presently, only the original MMPI depressed or noncommunicative frequently
should be administered to adolescents 17 can complete the MMPI-2, providing the cli¬
years of age and younger. There is no upper nician with valuable information that might
age limit for the MMPI-2 as long as the read¬ otherwise be unavailable. Such clients usually
ing level is adequate. Persons who score feel pleased that they can complete the task
below a Wechsler Adult Intelligence Scale— and relieved that other clients have had expe¬
Revised (Wechsler, 1980) IQ of 70 probably riences similar to their own, as evidenced by
will be unable to complete the MMPI-2. items referring to such experiences on the
Even when a person has inadequate MMPI-2.
 Administration and Scoring 33

After determining that the person is ca¬ MMPI-2, so there should be few objections
pable of completing the MMPI-2, the clini¬ to item content. Clients who raise objections
cian should insure that the individual is to item content frequently can be reassured
seated comfortably and provided with a pen¬ by being told that their answers will remain
cil for taking the test. If a pencil is used in confidential and that their answers to groups
taking the MMPI-2, the client can easily of items, rather than individual items, are
change any responses if so desired. Since few what is important. If this reassurance is insuf¬
problems generally occur in administration, ficient, the client may be allowed to omit an
the MMPI-2 can be given to small groups of objectionable item. The number of such
individuals as long as there is sufficient room omissions must be minimized, however, since
so that each person’s privacy can be re¬ the validity of the entire test becomes an issue
spected. if the client omits more than 10 items (see
Giving a brief explanation of why the Chapter 3).
test is being administered and what uses will If the client still objects to many items,
be made of the results, as well as answering the clinician may need to use some other per¬
any other questions that the client may have sonality instrument. Exploring more fully the
about testing, will pay tremendous dividends reasons for the client’s reluctance to complete
in avoiding invalid profiles. Clients also the MMPI-2 also might be useful. Clients are
should be told whether they will be provided more likely to object to item content when the
with feedback on their results. In most, if not MMPI-2 is being used for personnel selec¬
all, instances, clients should be provided with tion; they may have legitimate questions
such feedback which will enhance their moti¬ about the relevance of such items to job per¬
vation to complete the MMPI-2 appropri¬ formance. In such situations, the clinician
ately. Friedman, Webb, and Lewak (1989, should explain how the results from the
pp. 43-47) provide illustrations of common MMPI-2 will be used and how they are rele¬
questions asked about taking the MMPI with vant to personnel selection.
their suggested answers. The client should Once or twice during the test session, the
read the instructions, the clinician should an¬ clinician should unobtrusively check on the
swer any questions about them, and then the client’s progress. If possible, and particularly
client should proceed at his or her own pace if the client appears confused, the clinician
in completing the test. More than 90 percent should verify that the client is placing the an¬
of the persons taking the MMPI-2 will not swers correctly on the answer sheet. The clini¬
need any explanation of the instructions, and cian should be available throughout the test
they will complete the test in 60 to 90 min¬ session in case any questions arise.
utes. Although it is preferable to have the cli¬
One common question asked by clients is ent complete the MMPI-2 in a single session,
whether they should report prior feelings or it is not mandatory to do so. In such a case
current ones. The clinician should clarify that the client should be encouraged to complete
the client should report current feelings and the MMPI-2 within a few days at most in
experiences. Clients also occasionally ques¬ order to minimize the possibility of any sig¬
tion the appropriateness of the content of nificant changes in the client’s current status
some MMPI items, particularly those related during the testing period. Some clients are re¬
to gender, bodily functions such as elimina¬ lieved to know that they may complete the
tion, and religion (Butcher & Tellegen, 1966; MMPI-2 across several days since its length
Walker & Ward, 1969). Most of these objec- may appear formidable if it is to be com¬
tional items have been deleted from the pleted in one session. Awareness of the
34 Chapter 2

client’s need to complete the MMPI-2 across 7271). Since the MMPI-2 is a restricted test,
several days will increase the likelihood of only professionals with appropriate training
obtaining a valid MMPI-2. may purchase the test materials. The qualifi¬
cations for ordering the test materials are out¬
lined in the NCS catalog of tests or can be
Computer Administration
obtained directly from NCS.
The MMPI-2 can also be administered by MMPI-2 booklets are available in either
computer wherein the client sits in front of a softcover or hardcover. Both MMPI-2 book¬
terminal and responds to the items as they are lets have the 567 items in the same item order,
presented on the screen. Because the MMPI-2 unlike the MMPI where the Group Booklet
items are copyrighted, clinicians must lease form (softcover) and Form R (hardcover)
the software to be used on the computer from have the last 200 items in different orders.
the test distributor. At this time the only au¬ The softcover booklet of the MMPI-2 is pref¬
thorized distributor of the software for com¬ erable in situations where clients may mark
puter administration of the MMPI-2 is NCS on or otherwise deface the booklet, since it is
(P.O. Box 1416, Minneapolis, MN 55440, less expensive than the hardcover booklet.
800-627-7271). The hardcover booklet is useful in situations
Honaker (1988) has provided a critical such as a hospital where the client may not
review of the comparability of hand and have a desk or table readily available on
computer ft/IMPI administrations and con¬ which to work.
cluded that (1) computer administration is Several answer sheets can be used with
generally viewed in a more positive light and the MMPI-2 booklets, some of which are de¬
takes less time to complete, (2) individuals are signed for hand or computer scoring and oth¬
ranked similarly across the two procedures, ers which are designed to be read by optical
and (3) computer administration may pro¬ scanners. When purchasing MMPI-2 test ma¬
duce lower overall profiles. Honaker de¬ terials for the first time, the clinician should
scribes a number of methodologic issues that insure that the answer sheet is appropriate for
must be addressed before clinicians can as¬ the test booklet and hand or computer scor¬
sume that hand- and computer-administered ing as desired.
MMPIs or MMPI-2s yield equivalent scores. Once the client has completed the
Honaker (1988; Honaker, Harrell, & Buf- MMPI-2, the clinician should inspect the an¬
faloe, 1989) and Butcher (1987) provide ex¬ swer sheet for any problems, such as an omit¬
cellent overviews of the issues that are in¬ ted item, an item marked both “true” and
volved in the use of computers with the “false,” marking all or most of the items as
MMPI that should be consulted by the clini¬ “true” or “false,” or an item for which the
cian interested in this topic. client changed the response but failed to
show clearly which answer was intended. Oc¬
casionally, a client will even omit an entire
TEST FORMS
column of items. Usually the client can
The MMPI-2 exists in many different forms: readily correct these problems; then scoring
softcover, hardcover, audiotape, and com¬ of the responses can begin.
puter. All of these forms have the items in the
same order with the same item numbers. All
Hand Scoring
MMPI-2 test forms, answer sheets, and pro¬
file sheets are available from NCS, P.O. Box Scoring can be accomplished either by com¬
1416, Minneapolis, MN 55440 (800-627- puter (discussed later in this chapter) or by
 Administration and Scoring 35

hand. The first step in hand scoring is to ex¬ Second, each column on the profile sheet
amine the answer sheet carefully and indicate is used to represent the raw scores for a spe¬
omitted items and double-marked items by cific scale. Each dash represents a raw score
drawing a line through both the “true” and of 1 with the larger dashes marking incre¬
“false” responses to these items with a brightly ments of 5. Thus, the clinician notes the
colored ink pen. Also, cleaning up the answer client’s raw score on the scale being plotted
sheet is helpful and facilitates scoring. Re¬ and makes a point or dot at the appropriate
sponses that were changed need to be erased dash. Since the client has a raw score of 7 on
completely if possible, or clearly marked with the L (Lie) scale (see Profile 2-1), the clini¬
an “X” so that the clinician is aware that the cian finds the dark dash marked 5 on this
item has not been endorsed by the client. scale and then counts up 2 more dashes and
There is one scale that must always be makes a point or dot at 7. A similar proce¬
scored without a template: The ? (Cannot dure is followed for the other two validity
Say) scale score is the total number of items scales. Once the clinician has plotted the
not marked and double marked. All of the client’s scores on the three validity scales, a
validity and clinical scales on all hand-scored solid line is drawn to connect them. The raw
answer sheets are scored by placing a plastic score on the ? (Cannot Say) scale is merely re¬
template over the answer sheet with a small corded in the proper space in the lower left-
box drawn at the scored (deviant) response— hand corner of the profile sheet.
either “true” or “false”—for each item on A similar procedure is followed to plot
the scale. The total number of such items the ten clinical scales except that five of the
marked equals the client’s raw score for that clinical scales (/ [Hypochondriasis], 4 [Psy¬
scale; this score is recorded in the proper chopathic Deviate], 7 [Psychasthenia], 8
space on the answer sheet. One scale—Seale 5 [Schizophrenia], and 9 [Hypomania]) are K-
(Masculinity-Femininity)—is scored differ¬ corrected and a fraction of K is added to the
ently for males and females, and unusually raw score before the client’s score is plotted.
high or low scores on this scale might indicate (The rationale and procedure for the K-coy-
that the wrong template was used. Among fe¬ rection will be described in the next section.)
males, for example, a raw score of less than For these five scales that are A'-corrected, the
30 is unusual, and such raw scores should at clinician plots the raw score on the scale with
least arouse a suspicion that the wrong tem¬ K added. Thus, on Scale / the client’s raw
plate was used in scoring the scale. All scor¬ score plus one-half (.5) of the raw score of K
ing templates are made of plastic and they is 15 (see Profile 2-1), so the clinician finds
must be kept away from heat. the dark dash marked 15 and makes a point
Plotting the profile is the next step in the or dot there. Once the clinician has plotted
scoring process. In . essence, the clinician the client’s scores on the ten clinical scales,
transfers all the raw scores from the answer another solid line is drawn to connect them.
sheet to the appropriate column of the profile The clinician should note that the validity
sheet (see Profile 2-1). Some precautions and clinical scales are not connected, since
must be taken and data calculations per¬ the determination of the validity of the
formed. hirst, separate profile sheets are used MMPI-2 is independent of or precedes the
for males and females, as with the scoring evaluation of the clinical scales.
templates for Scale 5; an unusually high or The left and right columns of the profile
low score plotted for Scale 5 should alert the for basic scales provide the T score equiva¬
clinician to the possibility that the wrong pro¬ lents for the raw scores on each scale (see
file sheet was selected. Profile 2-1). For example, the client’s raw
 36 Chapter 2

PROFILE 2-1
Name John Brown

MMPI-2
s U I l.nli.m.iv ,iiuI I i \K Kmlr\

44
%////<\» ’ 1 *hi/U'ff/ti r.vt' Address 041 1 Chicago Street
/iv/mt’/iAvy -J
Occupation Janitor Date Tested 4/ 8/91
Profile for Basic Scales
\li;iik'M'i.i Mtiliiplusic IViM'iuilii\ In\enioi \ - '
Education 9th Age 47 Marital Status Married
i.'inndii • In INI KU.IMSOI INI 1 \|\l RSI h Oi MINM SOI \
!'»II'M’' (renewed I'i'iM. Hus Profile \ orm I msm
\ll 11 Mi is leaned I )isi i ilnned e\eliiM\el> In \\ll()\\l C'OMI’l II K S'! SI I MS. INC Referred by Dr. Nichols
under lieeiise horn Ilk- I imcrsiis of Minnen'Ui

MMI’I ? .nkl MiMik'sou Mull ipli.isic Person.ilil\ I in enloin - _v are trademarks owned In MMPI-2 Code 96-145872/0:3 F’L-K
ll\- l imoisti\ oi Minnesota Punted in iIk* l lined Suites of Mnene.i

TorTcL F K
Hs• :>K
1
D
2
H\
3
P,’-4K
4
Ml
5
P,i
6
Pi* Ik
7
bi -'K
8
Mil->
9
Si
OTorTc Scorer's Initials HG

c E

M'

0
0
ll 0

Raw score _7 11 14 8 18 15 20 29 13 14 16 24 22 NATIONAL

COMPUTER
k to be Added 7 6 14 14 3 SYSTEMS

Raw Score with k 15 26 28 36 27 24001

 Administration and Scoring 37

score of 7 on the L (Lie) scale is equivalent to K to be added to each of the A'-corrected

a T score of 65. Similarly, the client’s raw scales (see Profile 2-1).
score with AT of 15 on Scale 1 (Hypochondria¬
sis) is equivalent to a T score of 57. The pro¬
LINEAR T SCORES
file form for the basic scales provides a direct
means of converting the raw scores on the Inspection of the completed profile sheet pro¬
standard validity and clinical scores into the vides the clinician with a standard score (T
appropriate T scores. The development and score) as well as a raw score for each scale. T
use of T scores will be described below. scores among normals have a mean of 50 and
a standard deviation of 10. Thus, a T score of
70 indicates that a score is two standard devi¬
ations above the mean, and a T score of 30 is
/(-CORRECTION
two standard deviations below the mean.
As noted in Chapter 1, Meehl and Hathaway Knowing the client’s T score on a scale is im¬
(1946) developed the K scale to identify indi¬ portant for two reasons. First, it shows how
viduals who were defensive in endorsing the the client scored compared to the group of
MMPI items. They determined that the raw normals on whom the MMPI-2 was standard¬
scores on five scales, when transformed into ized. For example, knowing that a male client
A7corrected form, enhanced the ability of has a raw score of 26 on Scale 5 (Masculinity-
these scales to discriminate their respective Femininity) means virtually nothing, but the
criterion groups from other groups of respon¬ fact that the client has a T score of 50 on this
dents (see Chapter 3). This Tf-correction pro¬ scale tells the clinician that the client is no dif¬
cess is a standard step in plotting the MMPI ferent on this scale than the MMPI-2 norma¬
and it was not changed in the restandardiza¬ tive group.
tion of the MMPI-2. Second, T scores enable the clinician to
The standard profile sheet can be used compare the client’s scores on the various
only to plot directly AT-corrected profiles. In scales with one another. For example, know¬
order to use this profile sheet correctly, the ing that a male client has a raw score of 26 on
clinician must add to five of the clinical scales Scale 5 and 39 on Scale 0 (Social Introver¬
the proper fractions of the client’s raw score sion) means very little; knowing that the cli¬
on the K scale. The five clinical scales in ques¬ ent has a T score of 50 on Scale 5 and 65 on
tion and their AT-corrections are 1 (Hypo¬ Scale 0, however, allows the clinician to de¬
chondriasis) -I- .5AT; 4 (Psychopathic Deviate) termine the relative deviation and the inter¬
+ AK\ 7 (Psychasthenia) + 1 K\ 8 (Schizo¬ pretive importance of these two scales in this
phrenia) + 1AT; and 9 (Hypomania) + .2K. specific client.
Thus, in Profile 2-1 where K = 14, the fol¬ Since the T score equivalent of each raw
lowing amounts were added to Scales /, 4, 7, score can be read directly from the profile
8, and 9, respectively: 7, 6, 14, 14, and 3. sheet when the raw score is plotted, the clini¬
The clinician need not calculate the val¬ cian need not perform any calculations to
ues of the AT-correction since a table on the convert raw scores to T scores. The following
left side of the profile sheet provides all equation was used to obtain the linear T score
needed fractions of K for all possible raw equivalents of each raw score:
scores on the K scale. Circling the raw score
of the K scale and the other numbers on this
same row in the table on the profile sheet fa¬
T = 50 + 10
cilitates locating the appropriate fractions of
38 Chapter 2

where: The original T scores for the MMPI clin¬

ical scales were developed using a linear
transformation of the raw scores. (The T
X = the client’s raw score
scores for the validity scales were assigned ar¬
M = the mean score on the scale in
bitrarily so it is not clear what transformation
normals
of these scores was involved [Hathaway &
SD = the standard deviation on the
McKinley, 1983].) As noted above, the trans¬
scale in normals
formation of raw scores to T scores allows
clinicians to know how much a client’s score
In converting raw scores to T scores, frac¬ on a specific clinical scale deviates from the
tions were rounded to the nearest whole num¬ mean of the original Minnesota normative
ber. For example, if a male answered 35 items group and to compare a client’s scores on the
on Scale 5 (Masculinity-Femininity) of the various scales with one another.
MMPI-2 in the deviant direction or like the Inherent in the use of T scores on the
criterion group on whom the scale was con¬ MMPI is the assumption that they have sim¬
structed, substituting 35 in the above formula ilar meanings from one clinical scale to the
would produce a T score of 68 (see Profile next (i.e., a T score of 75 on Scale 4 (Psycho¬
2-1): pathic Deviate) has the same probability of
occurrence as a T score of 75 on Scale 8
35 - 26.01
T = 50 + 10 ( [Schizophrenia]). This assumption is valid,
5.08
however, only if the scales involved have sim¬
- 50 + 10(1.8)= 50 + 18 = 68
ilar distributions. If the underlying raw score
distributions for each scale are not similar,
Similarly, if a male answered 21 items on then a T score of 70 will not be equivalent to
Scale 5 in the deviant direction, substituting the 97.7 percentile.
this raw score in this formula would produce Colligan, Osborne, and Offord (1980)
a T score of 40: examined the raw score distributions of the
standard validity and clinical scales in the
21 - 26.01
T = 50 + 10 ( ) Hathaway and Briggs (1957) “purified” sam¬
5.08 ple of the original Minnesota normative
= 50 + 10(- 1.0)= 50 - 10 = 40 group that was used to derive the T scores on
the standard profile sheet for the MMPI.
The above described procedure was used They found that most of these scales showed
to develop T scores for all of the validity and significant skewness to the right, with from
clinical scales on the original MMP1. These T 4.8 to 8.0 percent of these normal individuals
scores are known as linear T scores since they scoring above a T score of 70 instead of the
are linear transformations of the raw scores 2.3 percent that would be expected with a
that maintain the underlying distributions of normal distribution. Colligan and colleagues
the raw scores. The MMPI-2 retained linear recommended that Scales 1 (Hypochondria¬
T scores for the standard validity scales, sis), 2 (Depression), 7 (Psychasthenia), 8
Scales 5 (Masculinity-Femininity) and 0 (So¬ (Schizophrenia), and 9 (Hypomania) in
cial Introversion), and the Supplementary women and Scales 1 (Hypochondriasis), 2
scales (see Chapter 5). However, the remain¬ (Depression), 4 (Psychopathic Deviate), 7
der of the clinical scales on the MMPI-2 and (Psychasthenia), 8 (Schizophrenia), and 9
all of the content scales (see Chapter 5) are (Hypomania) in men should be interpreted
uniform T scores. more conservatively because of the increased
 Administration and Scoring 39

frequency with which elevations occurred in transformation could result in different

normal individuals. groups of clients with extreme scores on a
given scale. Third, Hsu wondered whether
the normalized transformations of Colligan
NORMALIZED T SCORES
and associates (1983) were tested to see
Colligan and associates (1980) suggested that whether significant skewness and/or kurtosis
normalized transformations of the raw scores was present. Fourth, he questioned the as¬
on each MMPI scale should replace linear sumption that equality of percentile ranks for
transformations to insure that the same T normalized T scores was a more valid indica¬
score elevation would have similar meaning tor of psychopathology than equality of lin¬
across scales. However, no further research ear T scores since no empirical data were pre¬
was generated on this issue until Colligan, sented. Finally, he noted that all of the existing
Osborne, Swenson, and Offord (1983, 1984, literature on the MMPI is based on linear T
1989) developed normalized T scores for their scores and switching to normalized T scores
contemporary normal sample, described in would make these data irrelevant since the el¬
Chapter 1. These normalized T scores were evations of the individual scales and the
developed by determining the transformation rank ordering of the scales would differ.
of the raw scores on each validity and clinical Colligan, Osborne, and Offord (1984)
scale of the MMPI that would result in a nor¬ provided a point-by-point rebuttal to Hsu
mal distribution. They found that square (1984). First, Colligan and colleagues stated
root and log transformations of the raw that the assumption of a normal distribution
scores were necessary to produce these nor¬ for dimensions of personality are accepted
mal distributions. widely in the social sciences and Hsu did not
The interested clinician can see Table 39 present any data to indicate that the assump¬
(Colligan et al., 1983, pp. 206-207) or Table tion was untenable. Second, they agreed with
24 (Colligan et al., 1989, pp. 60-61) for the Hsu that skewness and kurtosis did differ
specific transformation used on each scale. from scale to scale and as a result they used
These transformations resulted in 0.6 to 3.6 different transformations as appropriate.
percent of the normalized T scores of their Third, they did not state specifically whether
census-matched samples on the MMPI valid¬ they tested the kurtosis of their normalized
ity and clinical scales being greater than 70 transformations, although they presented
compared with the expected percent of 2.3. data on the skewness of each transformation.
The more recent publications of Colligan and Fourth, they believed that the more reliable
colleagues seemed to spur further interest in scores produced by normalized transforma¬
this issue that continues unabated. tions would enable clinicians to realize that
Hsu (1984) presented five reasons for true differences were being reflected when
questioning Colligan and colleagues’ (1983) two scales were being compared, not some ar¬
recommendation to use normalized transfor¬ tifact of skewness. Finally, they suggested that
mations of the raw scores on the MMPI. the possible invalidation of existing MMPI re¬
First, he questioned whether the underlying search is a serious problem that needed to be
dimensions of psychopathology were nor¬ evaluated empirically. If this empirical research
mally distributed that would justify a nor¬ were to demonstrate that contemporary norms
malized transformation of the raw scores were more appropriate, then clinicians will
measuring the dimension. Second, he argued need to consider utilizing current methodology
that the differences in skewness and kurtosis rather than rely on an aging database no mat¬
of the normalized as compared to the linear ter how extensive it might be.
40 Chapter 2

Since some clinicians may prefer using T First, non-/f-correeted linear T score dis¬
scores based on linear transformations of the tributions were determined by gender for
raw scores, Osborne and Colligan (1986) pro¬ each of the eight clinical scales. This step re¬
vided equations for computing linear T sulted in 16 distributions of linear T scores.
scores for the Colligan and associates’ (1983) Second, the associated linear T score was de¬
contemporary normative sample. These lin¬ termined for each percentile in each of the 16
ear T scores tended to vary less than four distributions. Finally, composite or average
points from the normalized T scores for this T scores were determined for each percentile.
contemporary sample except on the F scale, These composite T scores were then used to
which differed by 9 points in men and 11 create uniform T scores for each of the eight
points in women. clinical scales. Uniform T scores result in a
similar probability of occurrence of a partic¬
ular T score across these eight clinical scales
UNIFORM T SCORES
while maintaining the underlying positive
The MMPI-2 Restandardization Committee skew in the distribution. A easy means of un¬
(Butcher et al., 1989) realized the importance derstanding uniform T scores is to realize that
of insuring that the same T score elevation they are equivalent percentiles across these
would have similar meaning or equal proba¬ scales.
bility of occurring across scales, and they de¬ Table 2-1 provides an illustration of the
veloped uniform T scores to meet this need. process of creating uniform T scores for two
Uniform T scores were developed for all clin¬ scales. In this illustration, the composite or
ical scales on the MMPI-2 (except Scales 5 average T score for the 92nd percentile was
[Masculinity-Femininity] and 0 [Social Intro¬ 65, which would become the uniform T score
version], which retained linear T scores) and for the 92nd percentile on both scales. Simi¬
the new content scales (see Chapter 5). Linear larly, the composite T score for the 96th per¬
T scores were retained on Scales 5 and 0 be¬ centile was 70, which would become the uni¬
cause these two scales were derived in a dif¬ form T score for that percentile.
ferent manner than the other clinical scales These uniform T scores produce slight
and the distribution of raw scores was less changes in the overall elevation of the profile
skewed (Butcher et al., 1989). These uniform and may alter the rank ordering of the clinical
T scores were developed in a three-step pro¬ scales, which would result in a different code¬
cess. type than would be obtained with linear T

TABLE 2-1 Illustration of the Process of Creating Uniform T Scores

Scale 7 Scale 2
(Hypochondriasis) (Depression)
Uniform
T Score Percentile Men Women Men Women

80 99 82 79 81 80
75 98 76 74 75 74
70 96 70 69 72 68
65 92 66 67 64 64
60 85 60 61 59 61
 Administration and Scoring 41

scores. Since all of the interpretive issues tion, (2) normalized transformations will re¬
raised about normalized T scores also are ap¬ sult in fewer scores above a T score of 65
plicable to uniform T scores, they will not be than uniform transformations, and (3) the rela¬
reiterated here. tionships among any pair or set of scales will
Table 2-2 illustrates the relationship be altered substantially by these changes in
among linear, normalized, and uniform T relative elevation and distribution of the T
scores. Linear and uniform T scores are very scores. Again, clinicians are cautioned against
similar throughout the distribution on Scale 1 casually assuming that correlates of scales
(Hypochondriasis), whereas uniform T scores from the MMPI will generalize directly to the
are four or more points lower above T scores of MM Pi-2. Such generalizations need to be
70 on Scale 2 (Depression). Normalized and verified empirically, it would appear that
uniform T scores are virtually identical until a T there are sufficient research opportunities
score of about 65 on both scales; the uniform with the MMPI-2 to keep clinicians busy for
transformation produces slightly to signifi¬ several decades.
cantly higher T scores above that point.
These comparisons among these three
COMPUTER SCORING
procedures for transforming raw scores into
T scores illustrate the points that were made Computer scoring of the MMPI-2 eliminates
above: (1) normalized and uniform trans¬ the need for the clinician to go through all of
formations reduce the overall profile eleva¬ the above steps for scoring and plotting the

TABLE 2-2 Comparisons among Linear, Normalized, and Uniform T Scores

for a Male Client

Scale 1 (Hypochondriasis) Scale 2 (Depression)

T Score T Score

Raw Score Linear Normalized Uniform Linear Normalized Uniform

10 47 43 42 34 26 32
12 51 49 48 39 33 36
14 56 54 54 43 39 40
16 60 59 59 48 44 45
18 64 63 64 52 49 50
20 69 66 68 56 53 54
22 73 69 73 61 57 59
24 78 72 77 65 61 62
26 82 75 81 70 64 66
28 87 78 86 74 67 70
30 91 80 90 78 70 74
32 96 82 94 83 73 78
34 100 85 99 87 75 81
36 105 86 103 92 77 85
38 109 88 108 96 80 89
40 1 14 90 1 12 100 82 93

Note: Raw scores on Scale 1 are /(-corrected. The uniform and linear T scores may be found in Ap¬
pendices A and K of Butcher et al. (1 989), respectively. The normalized T scores may be found in
Table 44 (p. 227) of Colligan et al. (1 983) or Appendix C (p. 1 1 6) of Colligan et al. (1 989).
42 Chapter 2

profile. Computer scoring also encourages swer sheets for two samples of MMPI cli¬
the clinician to use the content and supple¬ ents—Clinic Clients and University Stu¬
mentary scales (see Chapter 5) since no extra dents—had no errors on any of the 14 validity
time is required. Various computer scoring and clinical scales, and another 12 to 14 per¬
services are available through NCS, P.O. Box cent had errors on only one scale (see Table
1416, Minneapolis, MN 55440 (800-627- 2-3). For a third group—Medical Patients—
7271); clinicians will need to determine which whose answer sheets were computer scored,
service is most appropriate for their clinical 100 percent of the answer sheets w^ere scored
setting. There are three basic types of com¬ perfectly. The computer is not always 100
puter scoring services: the MMPI-2 can be percent accurate, however; Dahlstrom and
administered, scored, and interpreted (if de¬ associates (1972) and Fowler and Coyle
sired) on a personal computer in the (1968b) have reported errors in computer¬
clinician’s office; the clinician can use a per¬ scoring services. Nevertheless, errors occur
sonal computer and a modem to transmit the infrequently in hand scoring the MMPI and
client’s responses to the MMPI-2 to NCS and even more rarely in computer scoring.
have the results transmitted back to the When an error did occur, in the above
clinician’s personal computer; or the clinician two samples, it was most likely a result of the
can mail the answer sheet to NCS for com¬ clinician counting only one fewer deviant
puter scoring. Clinicians need to insure that item than the client actually answered (see
they are using an answer sheet for the MMPI- Table 2-4). Consequently, errors in scoring
2 that is compatible with whatever computer should have a negligible effect on the inter¬
scoring service is to be used, since different pretation of the profile. This statement is not
answer sheets are employed. meant to suggest that clinicians do not need
to be concerned about scoring the MMPI or
the MMPI-2; when clinicians exercise reason¬
ERRORS IN SCORING able care, however, few substantial errors in
Few if any errors in scoring generally occur scoring occur.
with the MMPI. Occasionally, the clinician Other than miscounting the number of
miscounts the number of deviant responses deviant responses, the other likely source of
on a specific scale. Greene (1980) reported error, as already mentioned, is using the tem¬
that at least 70 percent of the hand-scored an¬ plate for the opposite gender in scoring Scale

TABLE 2-3 Percentage of Answer Sheets Scored Incorrectly

Group3

Number of Scales Clinic University Medical

Scored Incorrectly Clients Students Patients

0 70% 82% 100%

1 14 12 0
2 6 2 0
3 6 4 0
4+ 4 0 0

Note: See Greene (1 980, pp. 22-25) for a more complete description of these groups.
an = 50 for each group.
 Administration and Scoring 43

TABLE 2-4 Percentage of Answer Sheets next two chapters, the effects of demographic
Scored Incorrectly with Errors variables such as age, gender, and education
of Varying Magnitude on MMPI profiles will be discussed.

Group
EFFECTS OF DEMOGRAPHIC
Magnitude Clinic University VARIABLES ON MMPI PROFILES
of Error3 Clients Students
Clinicians generally are aware that a number
-3 2.6% 0.0%
of demographic variables such as age, gen¬
-2 7.7 7.1
- 1 82.1 85.7 der, education, ethnicity, and environmental
+ 1 2.6 7.1 setting may have an potential effect on
+2 5.1 0.0 MMPI scores. However, there is little system¬
atic research that has investigated the influ¬
aA negative magnitude indicates that the clini¬
cian counted fewer deviant items than the cli¬ ence of any single demographic variable on
ent actually answered. MMPI profiles, let alone combinations of
these variables. Only the variables of age in
the comparison of adolescents and adults,
5 (Masculinity-Femininity). Unusually high and ethnicity in the comparison of blacks and
or low scores on Scale 5 should alert the clini¬ whites have been explored in any real depth.
cian to the possibility that the wrong template The research in these two specific areas will
was used, particularly if such a score seems be summarized in Chapter 8. This section will
inappropriate for the individual being tested. be limited to a more general discussion of the
effects of single demographic variables on
MMPI profiles. Gynther (1983) has provided
a recent review of this area. Little empirical
INTERPRETING THE PROFILE
research exists at this time to know whether
Once the MMPI-2 has been scored and plot¬ similar results would be found on MMPI-2
ted on the standard profile sheet, the process profiles, although it would be expected that
of interpreting the profile can begin. The first these results should generalize directly to the
step involves translating the T scores on each MMPI-2.
scale into more usable information. The next
two chapters will present interpretations of
Age
various T score elevations on each of the va¬
lidity and clinical scales both individually and The primary interest in the effects of age on
in some combinations with each other. Chap¬ MMPI profiles has involved the comparison
ter 6 will provide the correlates of the two of adolescents and adults, as was noted ear¬
clinical scales with the highest elevation at or lier. Little interest has been focused on the ef¬
above a T score of 65 (i.e., the codetype or fects of age across the adult life span.
high-point pair of the profile). The correlates Colligan and colleagues (1983, 1989) and
of the codetype are the core of the process of Swenson, Pearson, and Osborne (1973) have
interpreting the profile. Finally, Chapter 7 provided data on the effects of age on MMPI
will present illustrations of integrating all of profiles in adults. Table 2-5 provides the
this information to complete the process of mean scores for the standard validity and
profile interpretation. clinical scales for six age groups in the Col¬
Before turning to the interpretations of ligan and colleagues’ contemporary norma¬
the various validity and clinical scales in the tive sample of adults. Increases in mean
44 Chapter 2

TABLE 2-5 Effects of Age on MMP1 Scale Scores in the Colligan et al. (1 983) Contemporary
Normative Sample

Age N L F K 1(Hs) 2(D) 3(Hy) 4(Pd) 5(Mf) 6(Pa) 7( Pt) 8(Sc) 9(Ma) 0(Si)

Men
20-29 75 45.3 55.9 52.7 50.7 51.7 54.4 56.0 58.6 56.1 54.8 55.6 60.0 52.1
30-39 61 47.1 55.1 54.5 51.7 54.6 57.9 57.3 59.3 57.4 55.0 53.9 56.0 52.6
40-49 44 48.3 53.1 56.8 52.7 53.2 56.7 54.3 58.7 54.3 52.4 51.4 53.4 51.8
50-59 45 49.9 53.6 54.8 55.7 55.7 59.0 52.3 58.0 54.4 52.8 50.6 49.8 54.1
60-69 35 53.0 53.3 55.5 55.4 57.5 56.7 50.8 55.7 52.3 52.8 50.3 49.4 54.3
70 + 27 56.4 53.9 55.5 59.1 59.5 56.8 49.8 52.8 52.9 51.9 51.4 48.7 55.2

Women
20-29 77 46.1 51.9 55.0 48.8 52.0 52.0 55.0 45.6 56.4 54.3 54.4 54.1 53.7
30-39 62 47.0 51.3 56.1 49.0 51.9 53.0 54.8 44.8 56.5 52.8 53.3 51.7 53.8
40-49 44 48.6 51.7 55.4 50.9 52.4 52.9 52.7 45.4 56.7 52.9 53.2 51.0 56.0
50-59 49 50.7 52.4 55.7 53.2 54.4 54.6 52.9 47.1 54.2 52.6 53.2 49.6 55.9
60-69 42 53.7 51.5 56.3 52.3 54.6 52.7 49.4 49.1 52.3 52.1 52.5 46.1 58.7
70 + 44 56.1 54.4 52.1 55.7 57.6 54.4 48.5 51.6 53.9 53.8 53.3 48.8 57.7

Note: Data are adapted from Colligan et al. (1 983), Appendix J, pp. 340-352. These data are expressed as
means of the linear T scores based on the original Minnesota normative group.

scores as a function of age occurred fre¬ Femininity) increased in women. Five scales
quently and were five T points or more on were relatively unaffected by age: Scales F,
Scales L, 1 (Hypochondriasis), and 2 (De¬ K, 1 (Hypochondriasis), 2 (Depression), and
pression) in both men and women, and on 0 (Social Introversion). These small decreases
Scale 5 (Masculinity-Femininity) in women. in elevation would mean that there was a
Decreases in mean scores of 5 T points or slightly smaller probability of a clinical scale
more occurred only on Scales 4 (Psycho¬ being elevated above a T score of 70 with in¬
pathic Deviate) and 9 (Hypomania) in men creasing age in these medical outpatients.
and women, and on Scale 5 in men. These Both Colligan and associates (1983,
changes in mean scores were nearly 10 T 1989) and Swenson and associates (1973)
points (one standard deviation) on Scales L, found that mean scores decrease consistently
7, and 9 (men only). with increasing age on Scales 4 and 9, and in¬
Swenson and associates (1973) reported crease on the L scale. There also was a ten¬
the MMPI data from 50,000 medical outpa¬ dency for scores to decrease on Scales 7 and 8
tients, and Table 2-6 summarizes the mean with increasing age, although these decreases
profiles on the standard validity and clinical were smaller in the normal sample than in the
scales for six age groups. They found system¬ medical sample. Since both studies used a
atic effects as a function of age, with most cross-sectional design (i.e., they sampled
scales showing consistent small decreases in their individuals at one point in time rather
elevation. The decreases in elevation across than following a single group of individuals
these age groups were five T points or larger across time), it is not possible to know for
on Scales 3 (Hysteria), 4 (Psychopathic Devi¬ sure whether the prevalence of psychopathol¬
ate), 7 (Psychasthenia), 8 (Schizophrenia), ogy actually decreases with age or whether
and 9 (Hypomania) in both men and women. older individuals with less psychopathology
Only the L scale increased in elevation in both are simply more likely to participate in stud¬
men and women, and Scale 5 (Masculinity- ies such as these. Regardless of the explana-
 Administration and Scoring 45

TABLE 2-6 Effects of Age on MMPI Scale Scores in Medical Patients

Age N L F K 1(Hs) 2(D) 3(Hy) 4(Pd) 5(Mf) 6(Pa) 7(Pt) 8(Sc) 9(Ma) 0(Sh

Men
20-29 1298 48.7 53.7 57.1 60.9 60.2 63.1 60.2 57.2 55.6 59.6 56.3 57.8 51.3
30-39 2905 48.7 52.6 57.6 62.4 60.8 63.6 58.4 56.7 55.5 57.8 53.7 55.2 51.0
40-49 5379 49.1 52.3 57.0 63.0 60.9 63.2 57.1 56.2 54.8 56.0 52.2 53.3 51.3
50-59 7097 50.0 52.3 55.9 62.4 61.5 61.7 55.7 55.6 54.0 54.7 51.1 51.9 52.3
60-69 5315 51.9 52.4 55.6 60.7 61.1 59.2 53.6 54.7 52.6 53.1 50.5 50.0 53.1
70 + 1733 53.4 52.8 55.1 60.5 61.3 58.1 52.0 54.3 52.1 52.8 50.7 48.8 53.6

Women
20-29 1690 49.7 53.3 57.1 59.4 58.7 61.4 58.6 48.2 57.1 58.6 57.8 54.6 53.7
30-39 3474 50.5 52.4 57.4 62.0 60.6 63.3 57.4 48.4 56.9 58.4 56.8 51.9 54.0
40-49 5955 51.2 51.9 56.7 62.0 60.0 62.5 55.5 49.4 55.4 56.4 55.4 51.1 54.2
50-59 7209 52.4 51.4 56.5 60.7 59.4 60.3 54.0 50.0 54.2 55.2 54.3 50.5 54.1
60-69 5229 54.3 51.3 55.7 59.3 59.0 58.0 51.8 52.2 53.3 53.7 52.8 49.0 54.4
70 + 1471 56.3 52.0 55.2 58.7 58.7 56.4 50.3 54.2 52.7 53.1 52.3 48.2 55.1

Note: Data are adapted from Swenson et al. (1 973), Table I, p. 1 3. These data are expressed as means of the
linear T scores based on the original Minnesota normative group.

tion of these differences, however, these and 8 (Schizophrenia) in the younger pa¬
changes in mean scores as a function of age tients. Newmark and Hutchins (1980) found
reveal that the probability of obtaining a pro¬ that a discriminant function developed on
file with a clinical scale elevated at a T score younger schizophrenics could identify only
of 70 or higher will decrease slightly as the 22 percent of older (age 44-54) schizophre¬
client’s age increases. nics.
There have been few studies of the ef¬ Hedlund and Won Cho (1979) and
fects of age on MMPI profiles in psychiatric Schenkenberg, Gottfredson, and Christensen
settings. Aaronson (1958) reported the (1984) have provided T scores for the stan¬
changes in the highest clinical scale by age for dard validity and clinical scales by decade
the cases in Hathaway and Meehl’s An Atlas (see Tables 2-7 and 2-8). These patients
for Clinical Use of the MMPI (1951). Peaks showed consistent decreases in T scores
on Scales 1 (Hypochondriasis) and 2 (Depres¬ across the decades on Scales F, 4 (Psycho¬
sion) were more likely in the older patients pathic Deviate), 6 (Paranoia), 8 (Schizophre¬
and peaks on Scales 4 (Psychopathic Deviate) nia), and 9 (Hypomania). These decreases

TABLE 2-7 Effects of Age on MMPI Scale Scores in Psychiatric Patients

Age N L F K 1(Hs) 2(D) 3(Hy) 4(Pd) 5(Mf) 6(Pa) 7(Pt) 8(Sc) 9(Ma) 0(Si)

20-29 2956 50.9 68.5 50.5 60.5 67.8 62.6 75.2 57.6 68.4 68.6 74.8 66.1 57.6
30-39 2359 51.0 65.1 50.8 61.9 69.0 63.3 73.5 57.1 66.3 66.9 70.7 63.3 57.8
40-49 2006 51.1 62.3 51.1 63.2 69.5 64.0 70.7 55.7 63.6 65.1 66.6 60.9 57.7
50-59 1481 51.5 61.1 51.3 64.9 69.5 64.3 68.4 55.0 62.8 63.8 65.0 59.8 57.0
60-69 566 53.3 60.7 51.5 65.2 69.8 63.3 65.9 55.5 62.4 63.3 64.0 57.8 58.1
70 + 60 52.8 64.0 52.2 65.1 69.7 62.5 68.1 52.0 62.7 65.1 68.5 59.8 59.1

Note: Data are from Hedlund and Won Cho (1 979). These data are expressed as means of the linear T scores
based on the original Minnesota normative group.
46 Chapter 2

TABLE 2-8 Effects of Age on MMPI Scale Scores in Psychiatric Patients

Age N L F K UHs) 2(D) 3(Hy) 4(Pd) 5{Mf) 6(Pa) 7(Pt) 8(Sc) 9(Ma) 0(Si)

20-29 245 49.2 64.3 48.3 59.3 68.0 62.7 76.4 63.7 65.3 63.8 66.3 64.8 58.8
30-39 264 48.4 64.9 48.2 62.1 70.7 65.4 77.2 63.2 64.4 64.4 66.6 62.6 60.2
40-49 225 48.2 60.8 48.6 63.4 71.8 66.4 75.5 62.1 62.1 62.9 63.7 61.4 58.8
50-59 1 92 48.6 59.3 48.2 66.8 70.8 67.6 70.3 60.9 60.3 61.3 61.7 59.1 58.2
60 + 63 50.3 59.0 50.0 66.7 69.2 66.4 64.4 59.8 60.5 59.1 60.9 57.3 58.3

Note: data are adapted from Schenkenberg et al. (1 984), p. 1421. These data are expressed as means of the
linear T scores based on the original Minnesota normative group.

were particularly notable on Scales 4 and 9 sion in three age groups (20-39, 40-59, and
where they were approximately 10 and 8 T 60 +). The youngest group of patients em¬
score points, respectively. Only on Scale 1 phasized the intrapsychic nature of depres¬
(Hypochondriasis) did these patients show an sion, the middle-aged group felt vulnerable in
increase in T scores with increasing age. interpersonal relationships and approached
Several trends are evident in these psy¬ stresses in a less active manner, and the oldest
chiatric patients with increasing age that were group of patients highlighted frustration and
noted in the normal individuals and the med¬ anger in their relationships.
ical patients: (1) the decreases in scores on a Probably their most important finding
number of clinical scales will result in fewer was that the oldest group of patients did not
profiles reaching or exceeding a T score of 70, express concern over declining physical well¬
(2) Scales 4 and 9 decreased, and (3) only being which has been conjectured to produce
Scale K was unaffected by age. elevations on Scales 1 (Hypochondriasis) and
There also were age effects in one sample 2 (Depression) (cf. Swenson, 1961). Hyer,
that was not evident in other samples: (1) the Harkey, and Harrison (1986) reported that
L scale increased in the normal and medical older patients tended to respond more con¬
samples, but it did not change in the psychiat¬ servatively and to show less pathology on a
ric samples, (2) Scale 2 increased in the nor¬ number of the 77 supplementary scales that
mal sample but it did not change in either the they examined.
medical or psychiatric samples, (3) Scale 7 de¬ It is apparent that age does affect MMPI
creased in medical patients but it did not profiles, although these effects are a function
change in the normal or psychiatric samples, of the sample and more complex than might
(4) Scale 1 increased in both the normal and be expected. The next question is what can be
psychiatric samples but it did not change in done about these effects. Colligan and associ¬
the medical sample, and (5) the F scale de¬ ates (1983, 1989) suggested one solution when
creased in the psychiatric samples but it did they provided age-specific normalized T
not change in the medical or normal samples. scores for all of the standard validity and
Several investigators have examined the clinical scales. Another solution would be to
effects of age on individual clinical scales of provide adjustments to only those scales that
the MMPI or special scales. Dye, Bohm, An- are most affected by increasing age. For ex¬
derten, and Won Cho (1983) factor-analyzed ample, 5 to 10 T points could be added to
Scale 2 (Depression) and identified signifi¬ Scales such as 4 (Psychopathic Deviate) and 9
cant differences in the expression of depres¬ (Hypomania) that consistently decrease with
 Administration and Scoring 47

age. Graham (1979) suggested such a solution (1965) Alcoholism scale where men score
for a variety of demographic variables but to about two raw-score points higher than women
date there has not been a single study of such (see Chapter 5; Greene & Garvin, 1988).
a correction process. Aaronson (1958) and Webb (1971) re¬
Both of these solutions also will change ported that men were more likely to have
the relationships among the clinical scales high-points on Scales 1 (Hypochondriasis)
and the resulting codetypes. These potential and 7 (Psychasthenia), whereas women were
changes in codetype may leave the clinician more likely to have high-points on Scales 3
without any empirical basis on which to de¬ (Hysteria) and 6 (Paranoia). Gender differ¬
cide which codetype and what correlates are ences also occur in the frequency with which
most appropriate for a specific profile. MMPI codetypes are found (see Chapter 6,
Until empirical data are available to sug¬ Tables 6-1, 6-2, 6-3, and 6-4). Some
gest what scales should be adjusted and by codetypes (2-0/0-2, 3-8Z8-3, 4-6/6-4) are
what amount, clinicians will need to remain more frequent in women; others (7-2/2-/,
aware of the potential effects of age on 4-9/9-4) are more frequent in men. There has
MMPI and MMPI-2 scores. been little published research on whether gen¬
der differences occur in the empirical corre¬
lates of codetypes. This dearth of research
Gender
could reflect that gender differences are
Hathaway and McKinley (1983) were aware rarely found and hence are considered unim¬
of the effects of gender on MMPI profiles portant to report or the absence of research.
since they developed separate norms for all of Gynther, Altman, and Sletten (1973)
the clinical scales except Scales 6 (Paranoia), found gender differences in the empirical cor¬
9 (Hypomania), and 0 (Social Introversion). relates of only 1 (2-4/4-2) of 14 codetypes,
They did not develop separate norms for any which would support the former position.
of the validity scales, although males score However, Kelley and King (1978, 1979a,
slightly higher on Scale F and K, and lower 1979b) reported consistent patterns of gender
on Scale L. differences in their studies of codetypes in an
Because separate T score norms are used university mental health setting, which would
for men and women, gender is less likely to support the latter position. This issue of
have any appreciable effect on MMPI or whether demographic variables such as gen¬
MMPI-2 profiles. In fact, if gender does af¬ der affect the empirical correlates of a scale
fect MMPI or MMPI-2 profiles, that would or codetype needs to be investigated further.
be one basis for arguing that the T score con¬ Hopefully, the ensuing years will see more re¬
versions are somehow inappropriate. Clini¬ search on demographic variables being pub¬
cians who are reporting research on mixed lished.
samples of men and women should be sure to
use non-TGcorrected T scores in analyzing
Ethnicity
their data to minimize any potential effects of
gender. However, raw scores are preferable As noted above, the effects of ethnicity on
for analyzing data when only men or women MMPI profiles will be covered in Chapter 8.
are being studied (Butcher & Tellegen, 1978).
Clinicians will need to be aware of the
Education/Intelligence/Socia! Class
potential effects of gender when raw scores
are reported, as is typical for some supple¬ The demographic variables of education, in¬
mentary scales such as the MacAndrew telligence, and social class tend to be reported
48 Chapter 2

as if they were interchangeable in MMPI re¬ and Scale 5 (Masculinity-Femininity) in men

search despite their different referents. Typi¬ were significantly correlated with years of ed¬
cally a researcher will use one of these three ucation. Butcher did suggest that the inter¬
variables because the information is readily pretation of scores on Scale 5 in men with less
available. than a high school education or with post¬
Several studies have indicated that this graduate training should be adjusted slightly
set of demographic variables can have a sig¬ to account for the small effects of education.
nificant effect on MMPI profiles. For exam¬
ple, ethnic effects on MMPI profiles can be
accentuated or eliminated depending on Summary
whether or not these variables are controlled It is clear that some demographic variables do
(cf. Cowan, Watkins, & Davis, 1975; Penk, affect MMPI profiles, and may have similar
Roberts, Robinowitz, Dolan, Atkins, & effects on MMPI-2 profiles. However, there
Woodward, 1982; Penk, Robinowitz, Rob¬ is not a simple, direct relationship between
erts, Dolan, & Atkins, 1981; Rosenblatt & any single variable and scores on a given
Pritchard, 1978). Similarly, Heilbrun (1979)
scale. It also appears that the more data that
found that an index of psychopathy based in
are available on a single variable, the more
part on the raw score on Scale 4 (Psycho¬
complex are the relationships that are identi¬
pathic Deviate) was directly related to the fre¬ fied. Research in this area is still in the pre¬
quency of violent crimes in a sample of white liminary stages where single variables are
prisoners, but only in those prisoners with an examined at a time. Research that looks at
IQ less than 95.
the potential interaction of two or more of
As noted in Chapter 1, the MMPI-2 nor¬
these variables simply is nonexistent. It also is
mative group averaged nearly 15 years of ed¬
necessary for researchers to progress beyond
ucation, while the original MMPI normative
merely reporting mean T scores as a function
group averaged around 8 years of education.
of some demographic variable and begin to
This difference in level of education between
examine whether the correlates of a given
the MMPI and the MMPI-2 has been a source
MMPI-2 scale or codetype change as a func¬
of concern (cf., Caldwell, 1990). Butcher
tion of the demographic variable.
(1990) categorized the MMPI-2 normative
Stated simply, the question is whether a
group into five groups based on years of edu¬
man with an eighth-grade education and an
cation and concluded that the MMPI-2 T
IQ of 90 will have the same correlates of a
scores showed minimal impact from educa¬
2-7/7-2 codetype as a woman with a college
tion. Most of the mean scale scores for all
degree and an IQ of 120. It is time to begin to
five levels of education fell at or very near a T
look for empirical answers to basic questions
score of 50. Only Scale K in men and women
such as these.
 CHAPTER 3

Validity Indexes and

Validity Configurations

The MMPI was one of the first personality lates of each traditional validity scale also
tests to offer a means of directly assessing a will be provided. Finally, the research on sim¬
client’s test-taking attitudes. Thus, the first ulation of psychopathology, detection of re¬
step in interpreting an MMPI-2 profile is to sponse sets, and subtle and obvious items on
examine the various validity scales and in¬ the MMPI will be reviewed.
dexes to determine the client’s test-taking at¬
titude. If these validity indexes reveal an
A NOTE ON THE CONCEPT
inappropriate attitude, the entire profile may
OF VALIDITY ON THE MMPI-2
be invalid and the interpretation of the pro¬
file should be tentative at best. The concept of validity traditionally has
In this chapter the process for assessing meant the degree to which a test actually mea¬
the validity of an MMPI-2 profile will be de¬ sures what it purports to measure (Anastasi,
scribed, which will emphasize additional 1968). For example, a graduate-school apti¬
MMPI-2 measures of test-taking attitudes tude test is valid to the extent that it can iden¬
such as the Variable Response Inconsistency tify students who will succeed in graduate
(VR1N) scale (Butcher, Dahlstrom, Graham, school. The test’s validity would be assessed
Tellegen, & Kaemmer, 1989), and the Wiener by the relationship between scores on the test
and Harmon (Wiener, 1948) Obvious and and some index of success in graduate school,
Subtle subscales. Several MMPI measures of such as grade point average or completion of
test-taking attitudes, such as the Test-Retest a graduate degree. Similarly with the MMPI-
(TR) index (Dahlstrom, Welsh, & Dahlstrom, 2, the overall relationship of the test to some ex¬
1972), the Carelessness scale (CLS: Greene, ternal criterion (i.e., the accuracy with which
1978a), the Gough Dissimulation scale (Ds-r: the MMPI-2 can predict some other variable
Gough, 1954, 1957), and the Positive Malinger¬ such as length of hospitalization or psychiatric
ing (Mp) scale (Cofer, Chance, & Judson, diagnosis) would be a measure of its validity.
1949), also will be reviewed even though they The concept of validity on the MMPI-2
are not available on the MMPI-2. The corre¬ also has a second, somewhat different mean-

49
50 Chapter 3

ing. It describes the test-taking attitudes of an that involves multiple steps, which need to be
individual client, that is, whether or not the carried out in a sequential manner.1 An
client has endorsed the test items in some dis¬ overview of these steps is provided in Figure
torted manner. If the client has provided a 3-1. The clinician will see the various mean¬
consistent and accurate self-appraisal when ings of the concept of validity that are rais.ed
responding to the MMPI-2 items, the profile at each of these steps and their differential ef¬
is considered to be valid. Consequently, it is fects on overall profile validity as they are ex¬
possible for a client to provide a valid MMPI- plained below.
2 on one occasion, an invalid MMPI-2 at an¬ The clinician also may be surprised that
other time, and a valid MMPI-2 on a third the traditional validity scales of the MMPI-2
testing. Because in this second sense validity (Scales L, F, and K) are introduced at a very
actually can refer to the consistency with late stage in this process. Hopefully, the ra¬
which the client has endorsed the items, as tionale for this revised process for determin¬
well as the accuracy with which the client has ing the validity of an individual administra¬
described himself or herself, the clinician tion of the MMPI-2 will become clearer in the
needs to be aware of the multiple meanings of next few pages. The first step in assessing the
the concept of validity on the MMPI-2. validity of this specific administration of the
The usage of the concept of validity to MMPI-2 is to evaluate the number of items
refer to the consistency of item endorsement omitted (see Figure 3-1), which is discussed in
within a single administration of the MMPI-2 the next section.
would be described more appropriately by
the term of reliability. Specifically, one type
of reliability—internal consistency—has been ITEM OMISSIONS
used traditionally to refer to the extent to
which a person responds in a consistent man¬
Cannot Say (?) Scale
ner throughout a test or scale.
The usage of the concept of validity also The ? scale consists of the total number of
to refer to the accuracy of the self-appraisal items that the client omits, that is, fails to an¬
by the client further complicates this issue, swer or answers both “true” and “false.”
since a client can provide a consistent pattern Therefore, the ? scale is not composed of a
of item endorsement that is distorted in some specific set of items as the other validity and
manner so as to make himself or herself look clinical scales are; the client potentially can
more or less psychopathologic. The term va¬ omit any one or combination of the 567
lidity of the MMPI-2, however, has a long items. Thus, the term scale is a misnomer
history of usage, and attempting to convince since it comprises no specific items.
several generations of clinicians to use more In standard scoring procedures omitted
appropriate terms is probably unrealistic. items are considered to be answered in the
Therefore, the clinician needs to understand nondeviant direction since only items an¬
the multiple meanings of the concept of va¬ swered in the deviant direction are counted.
lidity on the MMPI-2. Thus, the effect of omitted items is poten¬
tially to lower the elevation of the overall pro¬
file and of any scale on which the items were
STEPS IN ASSESSING
omitted, since if the client had answered the
MMPI-2 VALIDITY
item a deviant response might have been
Assessing the validity of a specific adminis¬ given.
tration of the MMPI-2 to a client is a process Clopton and Neuringer (1977b) pro-
 Validity Indexes and Validity Configurations 51

FIGURE 3-1 Steps in Assessing MMPI-2 Validity

vided data to show how randomly omitting spectively, when these six quantities of ran¬
six different numbers of items (5, 30, 55, 80, dom items were omitted. More important,
105, and 130) affects MMPI profile elevation the codetype of the profile (the two highest el¬
and distorts profile configuration. The clini¬ evated clinical scales) changed in 1, 8, 4, 8,
cal scales dropped an average of .45, 2.74, 10, and 17 profiles, respectively, from a
5.61,7.70, 9.09, and 11.54 T score points, re¬ group of 30 profiles at each level of item
52 Chapter 3

omission. Thus, omitting only 30 items The number of MMPI items omitted by
changed the codetype for more than 25 per¬ specific clients has not been extensively exam¬
cent of their profiles, and when 130 items ined. Clopton and Neuringer (1977b) re¬
were omitted, more than half of the ported the mean number of item omissions in
codetypes changed. Consequently, profile three different populations: psychiatric pa¬
distortion seems possible when 30 or more tients at a large Veterans Administration
items are omitted, even though profile eleva¬ (VA) hospital, outpatients of a regional men¬
tion may be reduced only slightly. tal health center, and applicants for jobs as
It is clearly preferable to minimize the police officers and firefighters. The percent¬
number of omitted items. Initially explaining age of persons who omitted five or fewer
to the client the reasons for and importance items from these three populations was 85.6,
of completing the MMPI-2 will help obtain 88.9, and 93.7, respectively, and the corre¬
the client’s full cooperation in completing the sponding mean number of omitted items was
test. If the client omits an excessive number 4.68, 2.24, and 2.45. Clopton and Neuringer
of items (11 or more), the clinician can ask did not report the specific items that were
the client to review the omitted items and re¬ omitted.
spond to them based on whether each is Rosen (1958) and Tamkin and Scherer
mostly true or mostly false. If the client still (1957) also reported median ? scale scores of
omits an excessive number of items, the clini¬ less than 4 in two other VA male psychiatric
cian can question the client about the reasons samples. Rosen noted that only 15 of his 307
for not responding. A final solution is to con¬ patients (4.9 percent) omitted more than 30
struct an augmented profile (discussed later items. Ball and Carroll (1960) analyzed the ?
in this chapter). scale scores in groups of high-school fresh¬
A major issue concerning omitted items men and adolescent delinquents. The mean ?
is the client’s motivation for doing so. Is the scale scores for these adolescent populations
client unwilling to answer the omitted items ranged from 1.07 to 6.74 and were signifi¬
or is the client unable to answer them? Distin¬ cantly lower for girls than boys.
guishing these two categories of omitted Tables 3-1 and 3-2 present the number
items would be useful (Brown, 1950). In the of MMPI items omitted in normal individu¬
former instance the client probably would als and psychiatric patients, respectively. As
have given a deviant response if he or she had can be seen in Table 3-1, 5 or fewer items
actually answered the items; when this as¬ were omitted in 76 to 98 percent of these nor¬
sumption is tenable, construction of an aug¬ mal individuals, and more than 30 items were
mented profile might be considered. In the omitted by less than 1 percent, except for
latter instance ignoring the omitted items Greene’s (1986) sample of adult males. There
seems preferable since they are probably ir¬ is a consistent trend for men to omit slightly
relevant for this particular client. Unfortu¬ more items than women in all samples except
nately, no one, including Brown (1950), has Colligan, Osborne, Swenson, and Offord
provided criteria by which these two catego¬ (1983). The modal number of items omitted
ries of omitted items can be distinguished. A in these normal individuals was 0, and the av¬
clinician may be able to make this distinction erage number of omitted items was less than
by interviewing the client extensively. This 4.
procedure, however, is time-consuming and The psychiatric samples reported in
often impractical or impossible. Fortunately, Table 3-2 omitted slightly more MMPI items
most clients omit few items, and so the prob¬ than the normal individuals. Five or fewer
lem rarely occurs. items were omitted in 58 to 62 percent of
 xO
OX
00 CD
rx CD o
CD CD o

0 CM vp
CD C\j
CO
+-> E rx r- 00 o in in
c
(D
"O
£ c- o
rx cm
o d O

3 x$
ox
■(->
rx r- in
CO
0 ID rx’ CD
O) CD CD CD
_o 00
o xP
0s
o
CJ CNJ CD 00 ,3- in CD
CO d rx.’ CM o in
00 rx
o xp CO
OX 3
rx cd TO
C
ID r- o
c k-
0)
w
„
-S> CO
00
xO
oo CD o
<X>
£
OX
(3 CD CM O
00 CD o D Q.
I ii CD d in 00 o CM L.
IL _ D CM 0
_0

■*co-» o

3 rx CO 3
C
*D
< CO d CM 0
rx CO CD _C
oo
CO xP
OX
.0

0 CM CD CD CD rx CO
0
II in O T~ 00
■*->
00
c
Z D CM \- 0
xO o
OX
CM 00 a.
Q.
00 in CD 0
CO CD a> CD jQ
CD 4—.
oxo O

00 00 r- rx in oo
0 II lo rx cm d
rx r- oc
00 E
cn CO
4—1
w_
CM o
3
T3
c <
O
a>
CD
o
<D CD
■>
CD * i xO
OX
0
.0 O
rx
E
*0 T~~ 4— CO in rx 0
■M
^ || in d "sT CM CD
CO CM CD
CD
.1 %

00
o CM +-<
0
T— CD CD
CO CO CD
„ Cm 0
CO -S? co xP
O'
cd rx O r- CD CM 00
x:
+->
00
o I ii in cd O 00 in
>
c
u. _ 00 r— o
CO x?
O' ■O
4-1
CD 3 CO 00 in
0
k_
+-> ■D 0
a) cd d
c
< 00 co CD 0
CO
CD x?
OX xO
ox
’c
o>
CO O 00 o CM in CD
E
T3
o d in in o oo 0
O
^r oo
Tr
00
CD

>- “O 0 0 0
o
•O® *->00c in - > 00
>
+
£ o ■* E
0 «) I 0 I
' CD
ro o
Q
cn
3 £ cn 3 — 3 00
z o E £
3 3
CJ o CJ

53
 54 Chapter 3

TABLE 3-2 Frequency of Omitted Items on the MMPI in Psychiatric Patients by Gender

Psychiatric Patients (Hedlund & Won Cho, 1979)

Adults Adolescents
Number of
Omitted Male Female Male Female
Items (N = 8,646) (N = 3,743) ('N = 693) (N = 290)

0 29.8% 28.4% 28.0% 27.9%

1 -5 32.3 30.4 31.0 30.8
Cumulative % 62.1 % 58.8% 59.0% 58.7%
6-10 16.8 1 7.2 21.1 21.7
Cumulative % 78.9 76.0 80.1 80.4
1 1 -30 14.7 1 7.4 14.6 1 3.1
Cumulative % 93.6 93.4 94.7 93.5
31 + 6.4 6.6 5.3 6.5

M 8.7 9.3 7.7 8.4

SD 18.1 19.2 1 3.1 1 5.1

these patients, and more than 30 items were consuming, more research is sorely needed in
omitted by 5 to 6 percent. There do not ap¬ a variety of clinical populations.
pear to be any gender or age (adult versus ad¬ In addition to the number of items omit¬
olescent) differences in the number of items ted, the specific items most frequently omit¬
omitted in this psychiatric sample. The find¬ ted also are of interest. Dahlstrom, Welsh,
ing that psychiatric patients omit slightly and Dahlstrom (1975) have summarized in
more items than normal individuals high¬ Appendix A of their Handbook the fre¬
lights the importance of assessing item omis¬ quency with which each MMPI item was
sions in clinical populations. omitted by the purified Minnesota adult sam¬
Although high ? scale scores occur infre¬ ple and three different samples of college stu¬
quently, research is needed to establish the dents. Gravitz (1967) reported the frequency
causes of excessive omissions when they do with which specific items were omitted by
occur. The work in this area to date has re¬ normal men and women taking the MMPI as
sulted in conflicting findings. Dahlstrom and preemployment screening for vocational po¬
colleagues (1972) suggested that defensive sitions. Gravitz concluded that the omitted
procedures are one significant cause of ele¬ items were those that probed personal and
vated ? scale scores. In an empirical study, private feelings. Ball and Carroll (1960)
however, Tamkin and Scherer (1957) found found that items omitted by adolescents fell
that high ? scale scores did not seem to repre¬ into four content areas: items not applicable
sent a defensive, evasive attitude in their psy¬ to adolescents, religion, sexual and bodily
chiatric sample. Moreover, Eaddy (1962) functions, and unwillingness or inability of
found that intolerance of ambiguity and de¬ the adolescents to make a positive decision
sire for uncertainty were not related to level about themselves.
of item omission on the ? scale among college Gravitz (1967) reported that men omit¬
sophomores. Although the relative infre¬ ted 25 different MMPI items and women
quency with which most clients omit items omitted 16 different items 5 percent of the
makes this area of research slow and time¬ time or more from the standard validity and
 Validity Indexes and Validity Configurations 55

clinical scales. The Scales from which items and psychiatric samples: Group Booklet
were omitted most frequently were 2 (Depres¬ items 17, 53, 58, 70, 98, 168, 232, 295, 400,
sion), 5 (Masculinity-Femininity), and 0 (So¬ 476, 483, 513, and 558. The content of these
cial Introversion). items is provided in Table 3-5. There is a reli¬
MMPI items that are omitted frequently gious content to five of these items (53, 58,
in normal and psychiatric samples are pre¬ 98, 476, and 483), all of which were elimi¬
sented in Tables 3-3 and 3-4, respectively. If nated from the MMPI-2; the remainder ap¬
items omitted over 5.0 percent of the time are pear to reflect a variety of content areas. It
used as a criterion, normal individuals are should be noted that item 17 has been
slightly more likely to omit items than psychi¬ reworded on the MMPI-2 (item 6), which
atric patients. The actual MMPI items that may make this item more understandable.
are omitted appear to be similar in the normal Most of the other frequently omitted items on

TABLE 3-3 MMPI Group Booklet Items Frequently Omitted in Normal Samples by Gender

Normal Adults Normal Adults

(Greene, 1986) (Gravitz, 1967)

Male Female Male Female

(N = 163) (N = 238) (N = 7,149) (N = 4,816)

Item % Item % Item % Item %

70 35.6 58 24.4 70 12.8 58 18.4
58 19.6 70 19.3 295 9.9 70 17.6
483 1 5.3 441 1 5.5 400 9.8 255 14.8
513 14.1 513 13.4 1 27 8.9 287 8.5
471 13.5 483 13.0 564 8.7 310 8.4
1 15 10.4 1 15 1 1.3 255 8.2 20 7.9
295 10.4 98 10.9 502 7.3 295 7.4
258 9.2 249 10.9 547 7.0 297 6.7
65 8.0 17 9.2 549 7.0 237 6.4
249 8.0 558 9.2 98 6.9 306 6.2
566 8.0 255 8.8 41 5 6.7 98 6.1
98 7.4 471 8.8 232 6.5 400 5.9
17 6.7 65 7.6 1 15 6.4 17 5.5
220 6.1 168 7.6 249 6.4 289 5.5
387 6.1 295 7.6 451 6.2 127 5.3
19 5.5 476 7.6 455 5.9 564 5.3
231 5.5 1 1 7.1 287 5.4 302 5.0
239 5.5 19 7.1 306 5.4
255 5.5 53 6.7 521 5.3
101 6.7 58 5.1
20 6.3 299 5.1
52 6.3 297 5.1
310 6.3 101 5.1
232 5.9
236 5.9
220 5.5
258 5.5

Note: Gravitz (1 967) only reported omissions for items on the standard validity and clinical scales.
56 Chapter 3

TABLE 3-4 MMPI Group Booklet Items Frequently Omitted in Psychiatric Samples by Gender

Psychiatric Patients (Hedlund & Won Cho, 1979)

Adults Adolescents

Male Female Male Female

(N = 8,646) (N = 3,743) (N = 693) (N = 290)

Item % Item % Item % Item %

513 1 5.5 513 13.1 58 14.3 513 14.8
58 1 2.7 58 1 1.7 255 1 1.7 58 12.1
483 8.7 558 9.3 513 10.2 255 1 1.0
558 7.9 441 8.7 483 9.2 441 9.3
287 7.6 476 6.1 70 8.2 42 8.6
295 7.0 287 6.0 558 6.9 98 6.9
400 7.0 483 5.8 53 6.8 483 6.9
53 6.8 400 5.7 57 6.3 17 6.6
41 5 6.6 485 5.7 98 6.1 558 6.6
98 6.4 42 5.6 236 6.1 1 1 5 6.2
70 6.3 232 5.5 42 5.8 133 6.2
249 6.2 295 5.4 306 5.8 476 5.9
232 5.6 168 5.2 476 5.8 19 5.5
168 5.5 53 5.2 232 5.6 57 5.5
476 5.4 41 5 5.2 17 5.2 232 5.5
373 5.3 98 5.0 168 5.2 249 5.5
364 5.1 400 5.2 65 5.2
413 5.1 41 5 5.1 168 5.2
299 5.0 209 5.2
562 5.0 295 5.2
485 5.2
519 5.2

the MMPI have been eliminated: 70, 400, viewed, if possible, to evaluate their inability
476, 483, and 513. However, several of these to endorse these specific items since most cli¬
items are still found on the MMPI-2 ents are capable of doing so.
(168[180], 232[211], 400[345], and 558[470]). For example, suppose that a male psy¬
It would be expected that clients will omit chiatric patient omitted the following 22
slightly fewer items on the MMPI-2, since a MMPI items: 4, 17, 53, 58, 61, 70, 73, 77, 87,
number of the objectionable MMPI items 98, 115, 116, 135, 168, 173, 232, 295, 318,
have been dropped. 400, 513, 558, and 562. Since 78.9 percent
In scoring a client’s MMPI, the clinician (see Table 3-2) of male psychiatric patients
should ascertain how many of the omitted omit 10 or fewer items, this patient has omit¬
items are among those commonly omitted ted more items than most male psychiatric
items from Tables 3-3 or 3-4. These items patients. When Table 3-4 is consulted, he
probably have no particular significance for omitted 11 items that male psychiatric pa¬
this client and can be ignored as long as they tients commonly omit (53, 58, 70, 98, 168,
are not excessive (< 11). Clients who omit id¬ 232, 295, 400, 513, 558, and 562). (Item 17
iosyncratic items, that is, those items that are also could be included in this category since it
not omitted commonly, should be inter¬ is omitted commonly in other samples.) He
 Validity Indexes and Validity Configurations 57

TABLE 3-5 Content of Frequently Omitted also omitted 10 idiosyncratic items: 4, 61,73,
MMPI Items 77,87, 115, 116, 135, 173, and 318. The clini¬
cian could interview the patient about his in¬
MMPI
Item
ability to endorse these latter 10 items.
Number Item Content Data will need to be collected to deter¬
mine what items are commonly omitted on
17 My father was a good man. the MMPI-2. It does appear that several
[My father is a good man, (if MMPI-2 items (180, 211, 345, and 470)
your father is dead) my father
was a good man.]
would be expected to fall into the category of
commonly omitted items.
53 A minister can cure disease by
praying and putting his hand on Hathaway and McKinley (1967) arbi¬
your head. [A minister or priest trarily converted raw scores on the ? scale of
can cure disease by praying and the MMPI to T scores. They assigned a raw
putting a hand on your head.] score of 30 on the ? scale to a T score of 50;
58 Everything is turning out just thus 30 omitted items should be an average
like the prophets of the Bible score for the reference population. Assuming
said it would.
that a client could omit 1 item in 5 (110 items
70 1 used to like drop-the-handker-
total) and still have an interpretable profile,
chief.
Hathaway and McKinley assigned a raw
98 1 believe in the second coming
of Christ.
score of 110 to a T score of 70.
In light of the findings reported above,
168 There is something wrong with
my mind. Hathaway and McKinley’s estimate of the
232
number of items a client could omit without
1 have been inspired to a pro¬
gram of life based on duty which distorting the profile is extremely high. The
1 have since carefully followed. findings reported above suggest that a raw
295 1 liked "Alice in Wonderland" score of 30 on the ? scale occurs about 5 per¬
by Lewis Carroll. cent of the time; this means that a raw score
400 If given the chance ! could do of 30 is nearly a T score of 70, not a T score
some things that would be of of 50 as Hathaway and McKinley (1967) as¬
great benefit to the world. sumed. Similarly, the research reported indi¬
476 1 am a special agent of God. cates that a raw score of 100 on the ? scale
483 Christ performed miracles such occurs about .5 percent of the time; this is ap¬
as changing water into wine. proximately a T score of 80, not 70 as
513 1 think Lincoln was greater than Hathaway and McKinley suggested. Thus,
Washington.
the ? scale T score equivalents on the stan¬
558 A large number of people are dard profile sheet of the MMPI are of ques¬
guilty of bad sexual conduct.
tionable accuracy, and the clinician should
Reproduced from the MMPI by permission. consider using the T score equivalents pre¬
Copyright ©1 943, renewed 1 970 by the Uni¬ viously described that are more accurate esti¬
versity of Minnesota Press. Published by the mates of the frequency with which items are
University of Minnesota Press. All rights re¬
omitted.
served. MMPI-2 revisions of items are in
brackets. There are two small changes on the
MMPI-2 that may affect the ? scale. First, the
? scale is not included on the standard profile
sheet for the MMPI-2 (see Profile 2-1); in¬
stead the number of omitted items is simply
58 Chapter 3

noted at the lower lefthand corner of the TABLE 3-6 Interpretations of Cannot Say
sheet. Second, the instructions for adminis¬ (?) Scale Elevations
tering the MMPI-2 discourage clients from
Raw
omitting items. These two changes may cause
Score Interpretation
clinicians to underestimate the importance of
checking for item omissions. Consequently, 0 1. Low. These clients are able and
clinicians will need to be more intentional in willing to respond to all of the
checking each client’s answer sheet to deter¬ items. This is the expected
score for most samples.
mine the exact number of item omissions. 1-5 2. Normal. Scores in this range in¬
No reliability data on the ? scale have dicate clients who are omitting
been reported. The interpretation of four lev¬ a few items characteristically
els of elevation of the ? scale are summarized omitted by the individual's ref¬
erence group or omitting a few
in Table 3-6.
items that have idiographic sig¬
The first step in determining the validity nificance. The specific items
of this specific administration of the MMPI-2 omitted and their content
involves ascertaining the number of omitted should be scanned. There is lit¬
items, as outlined in Figure 3-1. Tables 3-45 tle probability of profile distor¬
tion unless all omitted items are
(MMPI-2) and 3-46 (MMPI) (pages 100-101)
from a single scale.
provide a summary of the quantitative data 6-30 3. Moderate. Scores in this range
that are needed to use the flowchart for the indicate clients who have omit¬
first step. If only the first 370 items of the ted more items than is typical of
MMPI-2 are administered because the clini¬ most individuals. The scales
cian is only interested in obtaining a client’s from which the items have been
omitted should be checked.
scores on the standard validity and clinical There is a distinct possibility as
scales, the clinician should not count the non- the number of omitted items
administered items as being omitted. approaches 30 that the profile
A summary of the potential causes of ex¬ may be distorted as to configu¬
cessive item omissions is presented in Table ration or codetype. Construct¬
ing an augmented profile might
3-7 as well as some possible solutions for
be considered. The profile may
these problems. Most of these solutions allow be of questionable validity as
for the client to correct the problem of omis¬ the number of omitted items ap¬
sions on the MMPI-2 so that the clinician proaches 30.
then can proceed with the assessment of the 31+ 4. Marked. THE PROFILE IS VERY
LIKELY TO BE INVALID. These
consistency of item endorsement. Conse¬ clients are unwilling or unable to
quently, this step ensures that an excessive complete the MMPI-2 in an ap¬
number of items have not been omitted once propriate manner. They may be
any reasons that the client might have for overly cautious in trying not to
omitting items have been identified and cor¬ reveal any significant informa¬
tion about themselves, obses-
rected.
sionally unable to come to any
decision about numerous
items, or simply so defiant and
Constructing an Augmented Profile uncooperative that they will
A clinician occasionally will be faced with a not answer the items. If possi¬
ble, the clinician should have
situation in which a client omitted a sizable
the client complete the omitted
number of items and it is not possible to have items or retake the entire test.
the client complete these items. For example, A client who fails to complete
the MMPI-2 does not automati¬
cally fall into this category.
 Validity Indexes and Validity Configurations 59

TABLE 3-7 Potential Causes of and the profile. The first is based on the assump¬
Solutions for Excessive Item Omissions tion that the client systematically avoided an¬
swering the omitted items and that, if the cli¬
Cause Solution
ent had responded to them, the response
1. Client is unsure 1. Emphasize to cli- would have been in the deviant direction. The
of frequency of ent that item re- process basically involves determining which
occurrence of sponses indicate items were omitted from which scales and
item content. "mostly" true
adding one point for each omitted item to the
and "mostly"
false. raw score on the appropriate scale. (Appen¬
2. Client has been 2. Encourage the dix E of this book lists the scales on which
careless in com- client to take each MMPI-2 item is scored. Karol [1985]
pleting the test. sufficient time to has provided this same information in tabu¬
respond to all
lar form for the MMPI that may facilitate de¬
items.
3. Client is very de- 3. Explain that an- termining the scales on which each item is
fensive and un- swers to individ- scored.)
willing to en- ual items are less Since some items are scored on more
dorse any items important than than one scale and may even be scored in the
that might be scores on scales;
opposite direction from one scale to the next,
"controversial." reassure client
that responses it is possible for the omitted items to be
to items are con¬ scored as if the client responded both “true”
fidential. and “false” to them. This “double” scoring
4. Client is not fa- 4. None; actually of items is a logical inconsistency in this
miliar with item very few of such
method of constructing an augmented pro¬
content. items for most
people. file; however, if it appears that the client sys¬
tematically avoided the omitted items, this
method represents an appropriate attempt to
salvage as much clinical data as possible from
an otherwise lost cause.
when the MMPI-2 is administered as a An example of this method of augment¬
screening procedure, it may not be possible to ing the profile might be that a male client
interview the client again or have the client re¬ omitted 10 items when taking the MMPI-2
take the omitted items. One way of handling (see Table 3-8). By checking Appendix E, the
such circumstances is to augment the ob¬ clinician can determine which scales con¬
tained profile. There are no well-defined cri¬ tained these items. To construct the aug¬
teria as to when a profile should be mented profile, the clinician would then add
augmented. Augmenting is probably unnec¬ one point for each omitted item to the raw
essary if fewer than 5 items have been omit¬ score of the appropriate scale. For example,
ted, and probably inappropriate if more than two points would be added to the raw score
30 items have been omitted. Consequently, for Scale 7 (Psychasthenia) since 2 omitted
augmenting the profile may be considered be¬ items (9 and 242) appear on this scale. A sim¬
tween these two extremes if it is not possible ilar procedure would be followed for the
to have the client retake the omitted items, al¬ other items and scales (see Table 3-8). The
though clinicians should be well aware that clinician would then plot the augmented pro¬
there are no empirical data to justify this pro¬ file on the standard profile sheet. If the K
cedure. scale has been augmented, all the K-corrected
There are two methods of augmenting scales (/, 4, 7, 8, and 9) also would have to be
60 Chapter 3

TABLE 3-8 Example of the Procedure the proportion of items on each of these
for Constructing an Augmented Profile scales that the client answered in the deviant
direction. Then for each scale the clinician
MMPI-2 Scale would multiply this proportion times the
Item
Items Number True False number of omitted items, assuming that the
client would answer the same proportion of
1. 9 2, 7 the omitted items in the deviant direction.
2. 42 F, 8 Using the same example as above (see
3. 65 3
Tables 3-8 and 3-9), the client answered 27 of
4. 95 2
5. 1 34 2 the 57 items on Scale 2 (Depression) in the de¬
6. 148 2 viant direction and omitted 5 items. Thus, the
7. 167 0 3, 4 client answered 27/52, or .519 of the items in
8. 1 77 5 8 the deviant direction. Multiplying .519 times
9. 189 2
the number of omitted items (5) yields 2.60;
10. 242 7, 8, 9
this figure, which would be rounded to the
nearest whole number (3), would be added to
the raw score for Scale 2. A similar process
would be followed for each scale on which
items were omitted. As with the first method
augmented if a A'-corrected profile is being
of augmenting the profile, the clinician
plotted.
would then plot the augmented profile, ad¬
In addition, the clinician should indicate
justing the //-corrections if the K scale was
on the profile sheet that the profile has been
augmented and noting on the profile sheet
augmented using standard procedures. In this
that the profile was augmented by this proce¬
example, in which only 10 items were omit¬
dure.
ted, augmenting the profile changed the two
highest clinical scales from Scales 7 (Psychas-
thenia) and 0 (Social Introversion) to Scales 2
(Depression) and 7 (see Table 3-9). Since the
TABLE 3-9 Standard and Augmented
codetype based on which clinical scales are
Scale Scores
most elevated in the profile is a central fea¬
ture of MMPI-2 interpretation, a shift such Standard Profile Augmented Profile
as this in the codetype can have a significant
effect on the interpretation of the profile. Raw score Raw score
Scale (with K) T score (with K) T score
The second method of augmenting the
profile is based on the assumption that the L 1 39 1 39
client did not systematically avoid the omit¬ F 7 58 8 61
K 1 1 41 1 1 41
ted items and would not have answered all the
1 (FIs) 17 62 17 62
omitted items in the deviant direction. This
2 (D) 27 68 32 78
method would be appropriate if, for exam¬ 3 (Hy) 22 52 24 57
ple, the client had insufficient time to com¬ 4 (Pd) 28 62 29 64
5 (Mf) 31 60 32 62
plete the MMPI-2 and omitted a number of
6 (Pa) 13 61 13 61
the later items. 7 (Pt) 39 77 41 81
After determining, in the fashion de¬ 8 (Sc) 37 69 40 74
9 (Ma) 16 41 17 43
scribed above, which scales contained the
0 (Si) 45 72 46 73
omitted items, the clinician would calculate
 Validity Indexes and Validity Configurations 61

CONSISTENCY OF ITEM use of these measures on the MMPI, this sec¬

ENDORSEMENT tion will begin with a review of the measures
of the consistency of item endorsement on the
After the number of items omitted has been
MMPI. The clinician who is already familiar
checked and found to be in the acceptable
with this information can proceed directly to
range or corrected in one of the manners de¬
the discussion of VRIN and TRIN (pages 68-
scribed above, the next step in the process of
76).
assessing the validity of this administration
of the MMPI-2 is to verify the consistency of
item endorsement (see Figure 3-1). Consis¬
Test-Retest (77?) Index (MMPI)
tency of item endorsement verifies that the
client has endorsed the items in a reliable One measure of the consistency of item en¬
manner. It is necessary to insure that the cli¬ dorsement on the MMPI is the Test-Retest
ent has endorsed the items consistently before (TR) index (Dahlstrom et al., 1972). This
it is appropriate to determine the accuracy Index is the total number of the 16 repeated
with which the client has endorsed the items. items on the MMPI that the client has en¬
Another way of understanding the dif¬ dorsed inconsistently. These 16 repeated
ference between the consistency and accuracy items from Scales 6, 7, 8, and 0 were added to
of item endorsement, which may clarify why the MMPI to facilitate scoring of the IBM an¬
they are distinct steps in this process of as¬ swer sheet. The Group Booklet form num¬
sessing the validity of an individual profile, is bers for these MMPI items are provided in
helpful. Conceptualize the assessment of con¬ Table 3-10. These 16 repeated items were
sistency of item endorsement as being inde¬ omitted from the MMPI-2, presumably to
pendent of or irrelevant to item content, and shorten the test, so the TR index cannot be
the assessment of accuracy of item endorse¬ scored on the MMPI-2.
ment as being dependent on or relevant to Buechley and Ball (1952) reported that
item content. Thus, measures of the consis¬ the TR index was helpful in discriminating
tency of item endorsement assess whether the between adolescents with a T score of 70 or
individual has provided a reliable pattern of higher on the F scale who were unable or un¬
responding to the items throughout the test willing to respond consistently to the MMPI,
regardless of their content, whereas measures and adolescents whose responses were consis¬
of the accuracy of item endorsement assess tent. They concluded that a score of three or
whether the individual has attempted to dis¬ more on the TR index indicated an invalid
tort his or her responses to the items in some profile. Dahlstrom and associates (1972) sug¬
specific manner. gested that four or more inconsistent re¬
The consistency of item endorsement on sponses indicate questionable response reli¬
the MMPI-2 is assessed by the Variable Re¬ ability, although they did not provide a basis
sponse Inconsistency (VRIN) and True Re¬ for this decision rule.
sponse Inconsistency (TRIN) scales. The con¬ Greene (1979) investigated response con¬
sistency of item endorsement on the MMPI is sistency on the TR index in four samples: psy¬
assessed by the Test-Retest (TR) Index chiatric patients at a large VA hospital, cli¬
(Dahlstrom et al., 1972) and the Carelessness ents at a university psychology clinic, young
scale (CLS: Greene, 1978a). adolescents seen at a juvenile probation of¬
Since the function of measures of the fice, and university students enrolled in intro¬
consistency of item endorsement on the ductory psychology. The frequency of incon¬
MMPI-2 will be clearer after exploring the sistent responses varied substantially, both
62 Chapter 3

TABLE 3-10 Percentage of Inconsistent Responses for the Test-Retest (77?) Index Items
by Sample

Percentage of Inconsistent Responses by Sample3

MMp, ---------—— -
Group Booklet Psychology College Juvenile
Item Numbers VA Clinic Students Probation

1. 8-318 20% 8% 12% 42%

2. 13-290 6 6 16 18
3. 15-314 28 20 28 34
4. 16-315 2 12 16 26
5. 20-310 12 6 6 20
6. 21-308 6 2 20 14
7. 22-326 10 14 18 18
8. 23-288 4 14 0 14
9. 24-333 20 14 14 34
10. 32-328 12 14 16 30
11. 33-323 16 10 14 28
1 2. 35-331 8 14 0 28
13. 37-302 10 16 6 28
14. 38-311 4 4 2 32
15. 305-366 8 8 4 28
16. 317-362 1 2 12 12 22

Note: From "Response consistency on the MMPI: The TR index," by R. L. Greene, Journal of
Personality Assessment, 1979, 43, p. 70. Copyright © 1979 by the Society for Personality
Assessment, Inc. Reprinted by permission.
an = 50 for each sample.

within and among samples. The mean num¬ The more stringent cutting score of three or
ber of inconsistent responses by each sample more inconsistent responses would classify 68
was 1.86 (SD = 1.97) for VA psychiatric pa¬ percent of their profiles as invalid.
tients, 1.90 (SD = 1.71) for psychology clinic Comparing TR index scores with other
clients, 1.86 (SD = 1.78) for college students, validity indicators such as the F scale yielded
and 4.14 (SD — 2.84) for juvenile probation mixed results in these four samples (Greene,
adolescents. 1979). Among profiles in all four samples
The high frequency of inconsistent re¬ ruled invalid by a T score greater than 80 on
sponses in the juvenile probation sample the F scale, approximately 4 to 14 percent of
probably reflected the general uncooperative¬ them were valid according to the TR index,
ness and poor motivation of these adoles¬ using a cutting score of four or more incon¬
cents since they had an appropriate reading sistent responses. More importantly, among
level to take the MMPI. The juvenile proba¬ profiles that would be considered valid be¬
tion sample also had more invalid profiles cause of a T score less than 70 on the Fscale,
than the other three groups. When a cutting which has been a traditional criterion for call¬
score of four or more inconsistent responses ing a profile valid, 8 to 16 percent were in¬
was used, 52 percent of the juvenile proba¬ valid according to the TR index. Thus, the
tion profiles would be classified as invalid. TR index seems to identify some invalid pro-
 Validity Indexes and Validity Configurations 63

files that would go undetected by the F scale four or fewer inconsistent responses. Adult
alone, and it identifies some profiles as valid and adolescent psychiatric patients scored in
that the F scale alone would consider to be in¬ a very similar manner on the TR index, which
valid. is somewhat unexpected since adolescents
The frequency with which normal and generally are seen as being less compliant.
psychiatric samples endorsed the items incon¬ The presence of psychopathology did not
sistently on the TR index are provided in Ta¬ preclude these patients from being able to en¬
bles 3-11 and 3-12, respectively. Normal dorse the items on the TR index in a consis¬
samples were very consistent in their endorse¬ tent manner.
ment of the TR items, with 95 percent of One of the advantages of the TR index
these individuals making four or fewer incon¬ compared to the F scale in the assessment of
sistent responses. Males made slightly more the consistency of item endorsement is that
inconsistent responses than females. Gravitz the TR index is not affected by the presence
and Gerton (1976) reported similar levels of of psychopathology as the Fscale is. The sim¬
inconsistent responses in their 2,000 normal ilarity among Greene’s (1979) three samples
individuals undergoing preemployment (excluding juvenile probation) in frequency
screening; however, they found that women of inconsistent responses to individual items
made slightly more inconsistent responses and in the distribution of total scores suggests
than men. that the TR index is relatively independent of
Psychiatric samples also were very con¬ the type or degree of psychopathology (see
sistent with 85 percent of the patients making Tables 3-10 and 3-12). This contention is

TABLE 3-11 Frequency of Inconsistent Responses on the Test-Retest (77?) Index for Normal
Samples by Gender

Colligan et al. (1983) Greene (1986)

Adults Adults College Students

iMumDer ot
Inconsistent Mate Female Male Female Male Female
Responses fN = 646) (N - 762) (N 163) (N - 238) (N - 208) (N = 224)

0 32.2% 40.3% 27.6% 42.9% 25.5% 36.6%

1 29.7 29.5 41.7 36.6 32.2 33.0
2 19.4 14.1 16.0 8.8 20.2 1 7.9
3 9.7 7.6 3.1 7.6 14.4 6.7
4 4.2 3.9 6.1 2.9 4.3 4.0
Cumulative % (0 -4) 95.2% 95.4% 94.5CVo 98.8% 96.6% 98.2%

Cumulative % (5 + ) 4.8 4.6 5.5 1.2 3.4 1.8

5 1.1 1.7 3.1 0.8 1.9 1.3

6 1.7 0.9 0.0 0.4 0.5 0.5

7 0.6 0.4 0.0 0.0 0.5 0.0
8+ 1.4 1.6 2.4 0.0 0.5 0.0

M 1.48 1.28 1.47 0.95 1.53 1.14

SD 1 .76 1.72 1.84 1.15 1.43 1.22
64 Chapter 3

TABLE 3-12 Frequency of Inconsistent Responses on the Test-Retest (77?) Index for
Psychiatric Patients by Gender and for Stimulus Avoidant Patterns

Stimulus
Psychiatric Patients (Hedlund & Won Cho, 1979) Avoidant Patterns

Adults Adolescents
Number of Stimulus
Inconsistent Male Female Male Female Random Avoidant
Responses (N = 8,646) (N = 3,743) (N = 693) (N = 290) (N = 100) m = 436

0 18.9% 15.0% 17.7% 22.1% 0.0% 0.0%

1 23.9 21.9 21.9 26.2 0.0 0.0
2 19.7 20.1 18.0 19.0 0.0 1.4
3 13.6 1 5.6 16.5 1 1.0 0.0 0.9
4 9.0 9.9 8.7 8.3 1.0 6.2
Cumulative % (0 -4) 85.1% 82.5% 82.8% 86.6% 1.0% 8.5 °A

Cumulative % (5 + ) 14.9 17.5 17.2 13.4 99.0 91.5

5 6.0 6.9 4.9 4.1 3.0 17.0

6 3.3 4.3 4.2 3.4 6.0 13.8

7 2.4 2.7 3.3 2.8 12.0 17.9
8+ 3.2 3.6 4.8 3.1 78.0 42.9

M 2.40 2.67 2.79 2.64 8.06 7.24

SD 1.74 1.77 ’1.85 1.78 2.00 2.20

Note: Stimulus avoidance patterns were defined as sequences of responses such as TFTF,
TTFTTF, FFTFFT, and so on. See Nichols, Greene, and Schmolck (1989) for a complete descrip¬
tion.

supported by the findings of Jones, Neurin- that the client has endorsed the items consis¬
ger, and Patterson (1976) and Coche and tently and not necessarily accurately, since
Steer (1974), who reported small mean differ¬ the client could consistently overreport or un¬
ences between various diagnostic groups. derreport psychopathology (see Accuracy of
Although elevations on the F scale can Item Endorsement later in this chapter).
represent inconsistent patterns of item en¬ Moreover, since the TR index assesses only
dorsement, or the client’s acknowledgment the consistency of the client’s responses, it
of the presence of psychopathology, or the will not detect “all true’’ or “all false’’ re¬
client’s overreporting of psychopathology sponse sets, which are consistent but non-
(see Accuracy of Item Endorsement below), veridical test-taking sets. The Carelessness
the TR index is relatively unaffected by the scale (Greene, 1978a), which will be discussed
type and severity of psychopathology, so it in the next section, is more sensitive to these
can provide an independent estimate of the response sets on the MMPI. The TR index
consistency of item endorsement on the will easily detect “random’’ response sets
MMPI. that would yield a score of approximately 8
The TR index is useful in identifying in¬ (see Table 3-12; Rogers, Dolmetsch, & Cav¬
consistent patterns of item endorsement, but anaugh, 1983; Rogers, Harris, & Thatcher,
the clinician should keep in mind that-an ac¬ 1983).
ceptable score on the TR index indicates only The TR index will detect some MMPI
 Validity Indexes and Validity Configurations 65

profiles with inconsistent responses that ond occurrence is in error (p. 13). Clinicians
would be considered valid by traditional va¬ should check their MMPI scoring templates
lidity indicators such as theFscale, and it can to insure that the first occurrence of each pair
demonstrate that the client has been endors¬ of TR items is being scored.
ing the items consistently despite elevated
scores on theFscale (Evans & Dinning, 1983;
Carelessness Scale (CIS) (MMPI)
Maloney, Duvall, & Friesen, 1980). These
findings indicate that the TR index and the The Carelessness scale (CLS: Greene, 1978a)
traditional validity indicators are not measur¬ has been developed as an alternative indica¬
ing identical processes in test-taking attitudes tor of consistency of item endorsement on the
(Fekken & Holden, 1987). MMPI. Prior development of a carelessness
An interesting issue that has not yet been scale for the Addiction Research Center In¬
examined is the elevation and configuration ventory (Haertzen & Hill, 1963) indicated
of the clinical scales in those persons who that psychologically opposite items are more
achieve a T score of 70 or below on the Fscale sensitive than repeated items in detecting the
(i.e., a valid profile) and who have six or inability or unwillingness of a client to com¬
more inconsistent responses on the TR index plete the inventory appropriately.
(i.e., an invalid profile). Research also is Greene’s (1978a) CLS scale on the
needed to examine the effects of various lev¬ MMPI consists of 12 pairs of empirically se¬
els of inconsistency of item endorsement on lected items that were judged to be psycho¬
profile validity using independent or nontest logically opposite in content. Greene pro¬
validity criteria. vided normative data for three samples: VA
The 16 pairs of items on the TR index are psychiatric inpatients, clients at a university
a potential source of discrepancies in scoring psychology clinic, and college students. The
the standard MMPI validity and clinical MMPI Group Booklet item numbers for
scales, since historically there was no consis¬ these 12 pairs of items, the direction of scor¬
tent convention on whether the first or sec¬ ing for a deviant response, and the frequency
ond occurrence of each pair of items was to of deviant responses by sample are presented
be scored. Some scoring templates and com¬ in Table 3-13.
puter scoring services used the first occur¬ The maximum total score on the CLS
rence of each pair of items, whereas others scale is 12 if the client answers all 12 pairs of
used the second occurrence (McGrath, items inconsistently. Greene suggested that a
O’Malley, & Dura, 1986). Any time the TR cutting score of four or more deviant re¬
index for a client does not equal zero, there is sponses is optimal in identifying invalid pro¬
the possibility of a discrepancy in scoring files; however, he did not use any external cri¬
when different MMPI templates or computer teria to validate this cutting score. The mean
scoring services are compared. These poten¬ number of inconsistent responses by each
tial discrepancies will be most serious on sample was 1.76 (SD = 1.45) for VA psychi¬
Scale 8 (Schizophrenia) since 13 pairs of TR atric patients, 2.20 (SD = 1.28) for psychol¬
items are scored on that scale. ogy clinic clients, and 1.48 (SD = 1.34) for
The current Manual for Administration university students.
and Scoring of the MMPI (Hathaway & Mc¬ The more subtle nature of the CLS
Kinley, 1983) states that any templates scale—based on items that are not simply re¬
marked with a 1983 copyright or later score peated but are psychologically opposite—
the first occurrence of each pair of TR items. should enable the clinician to detect a pattern
The Manual also states that scoring the sec¬ of inconsistent item endorsement on the
66 Chapter 3

TABLE 3-13 Item Composition, Direction of Scoring, and Frequency of Deviant Responses
by Sample for the Carelessness (CLS) Scale

Frequency of Deviant Responses3

MMPI
Item Group Booklet Deviant Psychology
Pair Item Numbers Responses VA College Clinic

1. 10-405 Same 22% 8% 18%

2. 17- 65 Different 4 8 26
3. 18- 63 Different 26 34 20
4. 49-113 Same 2 6 6
5. 76-107 Same 12 12 34
6. 88-526 Same 12 8 24
7. 137-216 Same 20 14 30
8. 177-220 Different 4 6 8
9. 178-342 Same 22 20 16
10. 286-312 Different 18 12 10
1 1. 329-425 Same 24 12 18
1 2. 388-480 Different 10 8 14

Note: From "An empirically derived MMPI Carelessness scale," by R. L. Greene, Journal of Clinical
Psychology, 1978, 34, p. 408. Copyright ©1978 by the Journal of Clinical Psychology. Re¬
printed by permission.
an = 50 for each sample.

MMPI in sophisticated clients who might rec¬ sistent responses. This point will be even
ognize the existence of identical repeated clearer when the potential causes of inconsis¬
items and consequently go undetected by the tent responding are summarized in Table
TR index. The CLS scale also is useful in de¬ 3-23 (page 76).
tecting “all true” and “all false” response The frequency with which normal and
sets since either response set would result in a psychiatric patients endorsed the CLS items
total score of seven deviant responses (see are provided in Tables 3-14 and 3-15, respec¬
Table 3-13). In addition to detecting clients tively. Normal samples were very consistent
who are unwilling to answer the MMPI ap¬ in their endorsement of the CLS items, with
propriately, the CLS scale also seems to de¬ 97 percent of these individuals making four
tect clients who are psychologically confused or fewer inconsistent responses. There did
and unable to answer the MMPI appropri¬ not appear to be any gender differences in the
ately. In such a case, an interview with the cli¬ endorsement of the CLS items. Psychiatric
ent usually enables the clinician to recognize samples also were very consistent, with 85
the client’s mental confusion. percent of the patients making four or fewer
Bond (1986) and Fekken and Holden inconsistent responses. Adult and adolescent
(1987) have noted that carelessness is not the psychiatric patients scored in a very similar
primary cause of inconsistent responding to manner on the CLS scale. It appears that the
the CLS items in normal students. It may be CLS scale performs in a very similar manner
more appropriate to call the CLS scale an in¬ as the TR index in assessing consistency of
consistency scale, which does not imply any item endorsement.
motivation by the client for his or her incon¬ As with the TR index, further research is
 Validity Indexes and Validity Configurations 67

TABLE 3-14 Frequency of Inconsistent Responses on the Carelessness (CLS) Scale

for Normal Samples by Gender

Colligan et al. (1983) Greene (1986)

Adults Adults College Students

Number of
Inconsistent Male Female Male Female Male Female
Responses (N = 646) (N = 762) (N - 163) (N = 238) (N = 208) (N = 224)

0 27.4% 31.9% 40.5% 43.3% 26.4% 35.3%

1 35.8 31.5 25.2 30.7 30.8 31.7
2 19.6 22.8 18.2 19.3 24.5 20.1
3 1 1.6 8.4 9.2 4.6 9.6 9.8
4 3.6 4.0 4.4 1.3 6.3 2.2
Cumulative % (0 -4) 98.0% 98.6% 97.5% 99.2% 97.6% 99.1%

Cumulative % (5 -1-) 2.0 1.4 2.5 0.8 2.4 0.9

5 1.2 0.6 2.5 0.8 1.4 0.9

6 0.6 0.8 0.0 0.0 1.0 0.0

7 0.2 0.0 0.0 0.0 0.0 0.0
8+ 0.0 0.0 0.0 0.0 0.0 0.0

M 1.35 1.26 1.19 0.92 1.47 1.15

SD 1.24 1.21 1.31 1.03 1.31 1.13

needed on the CLS scale to validate it exter¬ and CLS scale, with 97 percent of these indi¬
nally. Information is needed on how fre¬ viduals making eight or fewer inconsistent re¬
quently “inconsistent profiles” detected by sponses and most of them making five or
each of these measures are, indeed, invalid. fewer inconsistent responses. There did not
Meanwhile, both the TR index and the CLS appear to be any gender differences. Psychi¬
scale can be used to identify some MMPI atric samples also were very consistent, with
profiles that should be interpreted cau¬ 85 percent of the patients making eight or
tiously, if at all. fewer inconsistent responses. Adult and ado¬
Since the TR index and the CLS scale are lescent psychiatric patients scored in a very
both relatively short scales with 16 and 12 similar manner on the sum of the TR index
items, respectively, their reliability as a mea¬ and the CLS scale.
sure of the consistency of item endorsement Nichols, Greene, and Schmolck (1989)
can be enhanced by summing the score on the have developed a set of decision rules based
two scales. The frequency with which normal on the TR index and the CLS scale to identify
and psychiatric samples made inconsistent re¬ inconsistent patterns of item endorsement on
sponses on the sum of the TR index and the the MMPI. They also compared the use of
CLS scale are provided in Tables 3-16 and these rules with more traditional measures of
3-17. validity such as the Fand K scales and found
Normal samples were very consistent in that their rules were generally superior. They
their endorsement of the sum of the TR index recommended that their decision rules be used
68 Chapter 3

TABLE 3-15 Frequency of Inconsistent Responses on the Carelessness (CLS) Scale

for Psychiatric Patients by Gender and for Stimulus Avoidant Patterns

Stimulus Avoidant
Psychiatric Patients (Hedlund & Won Cho, 1979) Patterns

Adults Adolescents
Number of Stimulus
Inconsistent Male Female Male Female Random Avoidant
Responses (N - 8,646) (N = 3,743) (N = 693) (N = 290) (N = 100) (N = 436)

0 12.6% 8.8% 9.7% 10.0% 0.0% 0.0%

1 22.1 19.7 16.9 19.0 0.0 0.5
2 23.5 22.8 21.9 24.1 0.0 1.8
3 1 7.8 19.0 20.6 16.9 2.0 5.5
4 1 1.8 14.9 1 2.7 14.1 4.0 10.6
Cumulative % (0 -4) 87.8% 85.2% 81.8% 84.1 % 6.0% 18.4%

Cumulative % (5 + ) 12.2 14.8 18.2 1 5.9 94.0 81.6

5 6.4 8.0 9.1 9.0 16.0 1 3.1

6 3.4 3.9 5.2 3.8 16.0 28.9

7 1.7 1.8 2.9 2.8 28.0 19.0
8+ 0.7 1.1 1.0 0.3 34.0 20.4

M 2.40 2.67 2.79 2.64 5.90 6.08

SD 1.74 1.77 1.85 1.78 1.71 1.71

Note: Stimulus avoidance profiles were defined as sequences of responses such as TFTF, TTFTTF,
FFTFFT, and so on. See Nichols, Greene, and Schmolck (1989) for a complete description.

in clinical and research settings to identify in¬ dorses item 6 “true” and item 90 “false” or
consistent patterns of item endorsement. item 6 “false” and item 90 “true,” it is
scored as an inconsistent response. Slightly
more normal individuals and clients are likely
Variable Response Inconsistency
to endorse this pair of items inconsistently.
Scale (VRIN ) (MMPI-2)
VRIN actually consists of 49 pairs of unique
The Variable Response Inconsistency scale items, since two separate response patterns
(VRIN) consists of 67 pairs of items that are scored for 18 of these 67 item pairs, as
have similar or opposite item content. These with items 6 and 90.
pairs of items are scored if the client is incon¬ The first three columns of Table 3-19 il¬
sistent in his or her responses. Table 3-18 lustrate the distribution of scores on VRIN if
provides two examples of pairs of VRIN the client randomly “endorsed” the MMPI-2
items and the inconsistent response(s). For items. Since only one of the four possible
example, if a client endorses item 99 “false” combinations of “true” and “false” re¬
and 138 “true,” it is scored as an inconsistent sponse patterns are scored on each of the 67
response. pairs of items on VRIN, the average score in
Table 3-18 also shows that few normal such random sorts is 16.75 (67/4). Some 15.3
individuals or clients endorse this particular percent of these random sorts are at or below
pair of items inconsistently. If a client en¬ the recommended cutting score of 13 (T score
 Validity Indexes and Validity Configurations 69

TABLE 3-16 Frequency of Inconsistent Responses on the Sum of the Test-Retest (77?) Index
and the Carelessness (CLS) Scale for Normal Samples by Gender

Colligan et al. (1983) Greene (1986)

Adults Adults College Students

iMumDer ot
Inconsistent Male Female Male Female Male Female
Responses (N = 646) (N = 762) (N = 163) (N - 238) fN = 208) (M = 224)

0 11.5% 16.8% 13.5% 24.8% 9.1% 13.8%

1 20.6 21.5 28.2 24.8 18.3 25.0
2 22.8 22.4 15.3 19.7 18.3 21.0
3 14.4 14.4 1 5.3 16.0 18.3 17.4
4 10.5 8.3 12.3 6.7 16.8 1 2.5
5 9.4 6.2 6.1 5.0 10.1 7.1
6 4.2 4.3 1.8 0.4 2.4 0.9
7 2.2 2.4 3.1 0.8 2.4 0.9
8 1.1 1.0 1.8 0.8 1.4 0.4
Cumulative % (0 -8) 96.7% 97.3% 97.5% 99.2% 97.1% 99.1 %

Cumulative % (9 + ) 3.4 2.7 2.5 0.8 2.9 0.9

9 1.5 0.5 0.0 0.4 1.0 0.0
10 0.5 0.8 0.0 0.0 0.5 0.9
1 1 0.6 0.4 0.0 0.4 1.0 0.0
12 + 0.7 1.0 2.5 0.0 0.4 0.0

M 2.8 2.5 2.7 1.9 3.0 2.3

SD 2.4 2.4 2.6 1.8 2.2 1.8

of 80 on VRIN). However, it has been noted tently. In this latter case, the clinician is en¬
clinically that clients with scores on VRIN as couraged to examine the indexes described in
high as 13 or 14 appear to have endorsed the the next section. It also may be useful to ex¬
items consistently. As can be seen in Table amine these indexes even in cases where
3-19, 24.4 percent of random sorts are at or VRIN is 7 or lower to insure that the items
below a score of 14. have been endorsed consistently.
Research is needed to determine empiri¬
cally whether the suggested cutting score of
F-Fb| (MMPS-2)
13 or higher on VRIN to indicate an inconsis¬
tent pattern of item endorsement is appropri¬ The F scale and the Back F (FB) scale are
ate. Until such data are available, clinicians composed of items that were endorsed less
are urged to use the following guidelines: (1) than 10 percent of the time by the normative
when the scores on VRIN are 7 or lower, or sample on the MMPI and MMPI-2, respec¬
16 and higher, there is a high probability that tively. Each of these scales will be described
the client has endorsed the items consistently in more detail below when the traditional va¬
or inconsistently, respectively; and (2) when lidity scales of the MMPI-2 are discussed.
scores on VRIN axe in the intermediate range Since the items on both scales are infre¬
of 8 to 15, it is not clear whether the client has quently endorsed, clients would be expected
endorsed the items consistently or inconsis¬ to endorse approximately the same number
70 Chapter 3

TABLE 3-17 Frequency of Inconsistent Responses on the Sum of the Test-Retest (TR) Index
and the Carelessness (CLS) Scale for Psychiatric Patients and for Stimulus Avoidant Patterns

Stimulus
Psychiatric Patients (Hedlund & Won Cho, 1979) Avoidant Patterns

Adults Adolescents
Number of Stimulus
Inconsistent Male Female Male Female Random Avoidant
Responses (N = 8,646) tN - 3,743) tN - 693) tN = 290) (N = 100) (N = 436)

0 4.2% 2.4% 2.7% 4.1% 0.0% 0.0%

1 9.6 7.3 7.2 7.9 0.0 0.0
2 12.5 10.5 1 1.0 12.8 0.0 0.0
3 14.8 13.1 14.3 1 5.2 0.0 0.0
4 14.2 13.8 1 1.8 16.2 0.0 0.0
5 1 1.1 12.5 13.0 9.3 0.0 0.0
6 9.3 10.8 1 1.0 9.7 1.0 0.0
7 6.5 7.8 6.8 7.2 2.0 1.4
8 5.1 6.3 4.9 4.5 0.0 1.8
Cumulative % (0 -8) 87.3% 84.5% 82.7% 86.9% 3.0% 3.6%

Cumulative % (9 + ) 12.7 15.5 1 7.3 13.1 97.0 96.4

9 3.6 4.8 4.3 3.4 1.0 7.3

10 2.6 3.3 2.3 2.4 3.0 5.0

1 1 2.1 2.3 3.3 2.4 10.0 10.1
12 + 4.4 5.1 7.4 4.9 83.0 74.4

M 4.76 5.27 5.38 4.83 13.96 20.63

SD 3.29 3.28 3.58 3.30 2.68 2.79

Note: Stimulus avoidance profiles were defined as sequences of responses such as TFTF, TTFTTF,
FFTFFT, and so on. See Nichols, Greene, and Schmolck (1989) for a complete description.

of items on each scale. Consequently, the that the client has endorsed the items con¬
absolute value of the difference between the sistently if this index is 6 or lower when
number of items that the client has en¬ VRIN is in the intermediate range of 8 to
dorsed on each scale can be used as a measure 15.
of the consistency of item endorsement. The clinician also can add the score on
Table 3-19 provides the distribution of VRIN to the absolute value of F — FB to
this measure of the consistency of item en¬ provide a second measure of consistency of
dorsement for randomly endorsed MMPI-2s. item endorsement for scores in the intermedi¬
Exactly 75 percent of these random sorts ate range of 8 to 15. Less than 15 percent of
have a score of 7 or higher on this index, the random sorts had scores of 20 or less on
and the mean is approximately 10. Less this index (see Table 3-19). Only 31.1 per¬
than 26 percent of these random sorts with cent of the scores in the intermediate range
VRIN scores in the intermediate range (8 to on VRIN had scores of 20 or lower. Conse¬
15) had scores of 6 or lower on this index. quently, the clinician can use scores of 20 or
Thus, the clinician can be fairly confident lower on this index (L/?//Vplus the absolute
 Validity Indexes and Validity Configurations 71

TABLE 3-18 Examples of Scored Responses on VRIN with Frequency of Response

by Sample

1 38. I believe that I am being plotted against.

Normal Individuals Clients
99. Someone has it
in for me. True False True False
True 2.0% 3.7% 14.8% 9.7%
False 0.0 94.3 4.8 70.7
Deviant Response: 99F-138T.

90. I love my father or (if your father is dead) I loved my father.

Normal Individuals Clients
6. My father is a ——
good man. True False True False
True 91.2% 1.3% 85.3% 3.3%
False 5.5 2.0 5.5 5.9
Deviant Responses: 6T-90F; 6F-90T.

Note: N = 401 normal individuals; N — 1,500 clients.

value of F — FB) to indicate that the client has clearly inconsistent. Conversely, almost two-
endorsed the items consistently. thirds of these clients with scores in the inter¬
Finally, the clinician can add the raw mediate ranges on VRIN, and most of the
scores on the F and FB scales to the absolute cases (#25, 26, 28, 29, 30) with the highest
value of F — Fb to provide a third measure scores within this range, appear to have en¬
of consistency of item endorsement for scores dorsed the items consistently. Hopefully, the
in the intermediate range on VRIN. None of clinician realizes the necessity of routinely
these random sorts had scores lower than 36 checking the consistency of item endorsement
on this index (see Table 3-19), which the cli¬ with these additional measures for clients
nician could use as another means of assess¬ who have intermediate scores on VRIN.
ing consistency of item endorsement for these In discussing the consistency of item en¬
intermediate scores on VRIN. dorsement, it has been assumed that the client
Table 3-20 provides data for 31 clients has followed the same pattern of item en¬
who obtained intermediate scores (8 to 15) on dorsement for all 567 items. This assumption
VRIN. These 31 cases were identified in a may not always be appropriate, since clients’
sample of 90 MMPI-2s collected on an inpa¬ motivation may change as they go through
tient alcohol treatment program and a state the test. For example, a client could endorse
psychiatric hospital. Over one-third (34.4 the first 400 items consistently and then en¬
percent) of these clients scored in the interme¬ dorse the remainder of the items inconsis¬
diate range on VRIN, although approxi¬ tently. Since all of the items on the standard
mately 14 percent would be expected to have validity and clinical scales on the MMPI-2
scored in this range based on a T score of 61 occur in the first 370 items, if the client
for a raw score of 8 on VRIN in men. Five of started endorsing the items randomly after
these 31 cases, marked with asterisks in Table item 370, the clinician could still score and in¬
3-20, have very questionable consistency of terpret these scales. If it were possible to as¬
item endorsement, and one case (#27) is sess the clients’ patterns of item endorsement
 <D LO CO
> 4- CD LO
4—' OCsILOT-OOCOr^CDCOI^OO^J-CDLOCDOOh'OOCNLnOOOOCDOO
TO Ca) r^LOLOLOCDCD'^-4-oocMLnr^cocncj)CT>oo d r-’
o o O 4- cn d LO
= B T-cNco^-LocDr^cnoocncocDCDCDcDa^oo
E 05
d Q-
o O
CO

+ ^r^LO'd-o^-r-^t-cNcocOLO^cocDOO P" ID 04 O CN
a> c
i- a> t— CM LO 00 t— LO (7) LO LO CO LO C7) 00 CO 4— co co cn 4-
u!1 U- O 4— r— t— CN CN1 CN CN r* 4— 4 r-
CT
+
u»

*■^* r\
£o CDCOOCNNCOOOOCMNCOCOOCMNCDOOOCM'tCDOOOCJCDCO
cc o
CD
cocO'^^t-'t^'d-LOLDLOLOtocDCDCDcDCDr^r^r^r^r^oocococo
CO
Assessing Random Responses on the MMPI-2 with VRIN, F, and Back F Scales

00 CO
0) d oo
> 4-C
4-4 r-r-cON(DOONO0O)4-CJCONCDCMOOOO^ ^t^tCOOWCNCOOCOlOCOCNO CN
CO CL)
dddddd'-cNcodoo 'st;o6codidcNddcNo6coMCNcodcNdcdMo6d n
E £
c 4-T-CNCNCO^^tlOCOCDr^MOOOOOOa>a><DO)COO
=S &- ^ Q
5 Co
a? u

>
+ CN CO LO CD CO LO MLOOOOO’-OCOM^J-L04-OOOOO^t^'ct-r-LOMCDlOCO
>- cd LOLOr^O)CnCOrJ-COMCOCOCD'rj-COO’-CnCOCOCO^}-Cs|4-^t
u- D
cr
5

CD
£ o +
CO
o ooa^Or-cNco^j-LocDh'OOcDO’-cNco^tLocor^oocnO'-cNco^tLOcoMoocDO
CC CO 4-4-4-4-r-4-v-^4-4-CNCNCNCNCNCNCNCNCNCNCOCOCOCOCOCOCOCOCOCO^t

LO CO
CD O 00
> 4-
■M
co a)
C ^■CDcocDc£)OcNLoa)4-^ta)4-oor^r^cDoo(N^cn^cDcocoa>o O d
coLodcoc6Lr)4-oocDco4-o6cDr^ddcoLodo6o6cr)cr)cr)da)d
E £
c 05
r-4-cNcocO'ctLncDC£>r^-cocDCDa)a)a)a)a)CDa)a)a)cno
ll =J CL
CJ Q
CO

>
r^cocgcDoo^tc7)t^4-ocDr^oocy)^j- r-(0\T^01- CO ID CO CN CN CN
CD
i_ C
(D cNiocDooocNLOLDcor-Lnocor^'^-
Note: N = 2,500 for each comparison.
CN O MO CO CO <-
U- 3 4-4-4-4— 4— 0\|4— C\|4— 4— 4-
cr

CD
> 4 -
4—' C OCMNCNCO0NCONCONCOOOOiOOOOOOCDa)O CO c\i
CO Q)
ddd<-c\idoc>iodLoo6dddo6cN(bo6dddd
c
e &£ 4-csico^j-Lor^cocx)cocj)a)cr)cr)CDO
=S CL
u Q
CO
§
>
CO CN 00 CO LO CO LO 00 CO 00 O CO CO M CN
0
3
7

o c
1

CN CN to CD CM 00 4— 00 LO 00 O M LO CN 4-
TABLE 3-19

it s 4— CN CN CO CN CN CN 4— 4—
cr

S g
TO o CNCO^LOCOMCOCDO CNCON-LOONCOCDO'-CNCONIOCDOO
CC o 4— 4— 4— 4— 4— 4— 4— 4— CNCNCNCNCNCNCNCN
CO

72
 Validity Indexes and Validity Configurations 73

TABLE 3-20 Assessing Consistency of Item Endorsement for Intermediate Scores (8 to 1 5)

on VRIN with F and Back F Scales

Raw Scores

Client

a.

4»
VRIN F Fb VRIN + | F - Fb | F + Fb + \F - Fe

1
1 8 10 1 3 1 1 20
2 8 6 8 2 10 16
3 8 3 4 1 9 8
4 8 1 1 1 2 1 9 24
5 8 14 13 1 9 28
6 8 10 9 1 9 20
7 8 7 1 1 4 12 22
8 8 7 4 3 1 1 14
9 8 10 7 3 1 1 20
10 8 8 10 2 10 20
1 1 8 1 7 19 2 10 38
1 2 8 24 22 2 10 48
1 3* * 8 21 1 2 9 17 42
14 9 4 2 2 1 1 8
1 5 9 13 7 6 1 5 26
16 9 5 2 3 12 10
17 9 1 5 9 6 1 5 30
18 9 1 1 10 1 10 22
19 9 14 16 2 1 1 32
20 9 6 7 1 10 14
21 9 1 2 5 7 16 24
22* * 9 10 19 9 18 38
22** 10 14 19 5 1 5 38
24* * 10 13 19 6 16 38
25 1 1 1 7 18 1 12 36
26 1 1 5 3 2 13 10
21** 1 1 25 7 18 29 50
28 1 2 8 8 0 12 16
29 13 6 4 2 1 5 1 2
30 13 1 1 13 2 1 5 26
31 ** 1 5 1 7 1 5 2 1 7 34

Note: These 31 cases were obtained from a sample of 90 alcoholic and psychiatric inpatients.
Those clients marked with double asterisks have questionable consistency of item endorsement
by one or more of the indexes discussed in the text.

in blocks or groups of items rather than for determine when clients start to make incon¬
the entire 567 items, it might be possible to sistent responses. This approach might be
score those items up to the point where the particularly appropriate for intermediate
client started responding randomly. scores on VRIN where it is more difficult to
As shown in Table 3-21, the VRIN, F, make an assessment of the consistency of
and Fb items are fairly evenly distributed item endorsement. Research that addresses
throughout the MMPI-2 so clinicians could the consistency of item endorsement as the
74 Chapter 3

TABLE 3-21 Distribution of F, FB, and VRIN Items by Blocks of 100 and an Example
of a Client Who Endorsed the Items Inconsistently

Number of Items on Scale Number of Items Endorsed by Client

Item Numbers F Fb VRIN F Fb VRIN Total

1-100 16 0 4 1 0 0 1
101-200 1 7 0 6 1 0 1 2
201-300 1 7 2 9 4 0 1 5
301-400 10 13 9 0 5 3 8
401-500 0 1 1 10 0 6 2 8
501-567 0 14 1 1 0 7 0 7
Total 60 40 49 6 18 7

Note: There are only 49 unique item pairs on VRIN, since two response patterns are scored on
1 8 item pairs.

client goes through the MMPI-2 clearly is dard validity and clinical scales still appeared
needed. to provide a good description of her since
Clinicians should realize that VRIN may they are based on the first 370 items. The
be insensitive to cases in which the client be¬ MMPI-2 content scales, which tend to be con¬
gins to endorse the items inconsistently part centrated in the last 150 items (see Chapter 5),
way through the MMPI-2 for the same rea¬ were elevated rather significantly and did not fit
sons that VRIN is insensitive to random re¬ her clinical picture at all.
sponse patterns. Since only one of the four
possible combinations of “true” and “false”
True Response Inconsistency Scale
response patterns are scored on each of the 67
[TRIN ) (MMPI-2)
pairs of items on VRIN, the probability of a
client endorsing any single pair of items is The True Response Inconsistency scale
.25, not .50. Consequently, a client may have (TRIN ) consists of 23 pairs of items. TRIN is
a low score on VRIN since the first item in a very similar to VRIN except that the scored
pair may have been endorsed appropriately. response is either “true” or “false” to both
It would seem that the relationship between items in each pair. Table 3-22 provides two
the number of Fand FB items endorsed may examples of pairs of TRIN items and the in¬
be more sensitive to this type of inconsistency consistent response(s). For example, if a cli¬
than VRIN. ent endorses both items 40 and 176 “true,” it is
Table 3-21 also shows the pattern of scored as an inconsistent response. Similarly, if
item endorsement on VRIN, F, and FB for a a client endorses both items 125 and 195 either
33-year-old alcoholic inpatient. It is apparent “true” or “false,” it is scored as an inconsis¬
that the patient started to endorse the items tent response. Table 3-22 also shows that few
inconsistently somewhere between items 300 normal individuals or clients endorse either of
and 400, since she endorsed almost one-half these pairs of items inconsistently.
of the Fb items in each block of 100 items TRIN has 14 pairs of items to which the
after endorsing very few of the Fitems. When inconsistent response is “true” and 9 pairs to
asked about her performance on the MMPI- which the inconsistent response is “false.”
2, she related that she became “tired” as she Scoring TRIN is somewhat complicated. One
went through the items and responded in a point is added to the client’s score for each of
“hurry” toward the end of the test. The stan¬ the 14 item pairs that are scored if endorsed
 Validity Indexes and Validity Configurations 75

TABLE 3-22 Examples of Scored Responses on TRIN

1 76. I have very few headaches.

Normal Individuals Clients
40. Much of the time -———----
my head seems True False True False
to hurt all over.
True .5% 7.2% 4.5% 11.4%
False 72.1 20.2 68.6 15.5
Deviant Response: 40T-176T.

1 95. There is very little love and companionship in my family as

compared to other homes.
Normal Individuals Clients
125. I believe that —-——-—
my home life is True False True False
as pleasant as
that of most
people I know.
True 2.7% 84.1% 11.3% 51.6%
False 3.7 9.5 20.4 16.7
Deviant Responses: 125T-195T; 125F-195F.

Note: N = 401 normal individuals; N = 1,500 clients.

“true,” whereas one point is subtracted for flect that clients tend to be “yea-sayers,” low
each of the 9 item pairs that are scored if en¬ scores are obtained by clients who are “nay¬
dorsed “false.” Then 9 points are added to sayers,” regardless of the item content. The
this score. (Nine points are added to the score empirical correlates of high and low scores on
so that it is not possible to obtain a negative TOW will need to be determined.
score on TRIN. If a client endorsed none of A final comment about TRIN needs to
the 14 “true” item pairs and all 9 of the be made for those clinicians who use the com¬
“false” item pairs, a score of -9 would be ob¬ puter scoring of the MMPI-2 provided by
tained. Adding 9 points avoids this problem.) National Computer Systems. The Extended
For example, if a client endorsed three Score Report for the MMPI-2 indicates a
of the “true” item pairs on TRIN, and six of score of 9 on TRIN by placing an asterisk at a
the “false” item pairs, the score would be 6 T score of 50. Otherwise, a “T” or “F” is
(3 — 6 -F 9). If a client endorsed eight of the plotted at the appropriate T score based on
“true” item pairs on TRIN, and two of the the client’s raw score. Since only T scores
“false” item pairs, the score would be 15 (8 above 50 are possible on TRIN, there needs
— 2 + 9). The former client with a relatively to be a mechanism whereby high and low
low score of 6 on TRIN has a propensity to scores are differentiated. Scores from 9 to 19
say “false” to the items regardless of their are indicated by placing a “T” at the appro¬
content, whereas the latter client with a score priate T score.
of 15 is tending to say “true” to the items. For example, if a woman’s raw score on
Scores on TRIN should not be used to TOW was 15, a “T” would be plotted at a T
determine whether a client has endorsed the score of 95. Scores from 0 to 8 are indicated
items consistently. High scores on TRIN re¬ by placing a “F” at the appropriate T score.
76 Chapter 3

Thus, if a man’s raw score was 3, a “F” TABLE 3-23 Potential Causes of and
would be plotted at a T score of 92. This pro¬ Solutions for Inconsistent Item Endorsement
cedure may be clearer if the standard profile
Cause Solution
form for the Supplementary Scales is exam¬
ined (see Profile 7-3, page 298). 1. Client has not 1. Explain why the
been told why MMPI-2 is being ad-
the MMPI-2 ministered and how
Summary is being ad- the data are to be
ministered. used.
Now that VRINand the relationship between 2. Inadequate 2. Present the MMPI-2
F and FB as measures of the consistency of reading abil- orally by tape ad-
item endorsement on the MM PI-2, and the ity or com- ministration (see
TR index and the CLS scale have been de¬ prehension; Chapter 2). Dahl-
inadequate strom and col-
scribed as measures of the consistency of item
educational leagues (1972) re-
endorsement on the MMPI, the next step in ported that tape
opportunity.
the implementation of the flowchart in Fig¬ administrations are
ure 3-1 can be described. The criteria, sum¬ effective with read¬
marized in Table 3-45 for the MMPI-2 and ing/education levels
as low as the third
Table 3-46 for the MMPI, provide the quan¬
grade.
titative data necessary to determine whether 3. Present the MMPI-2
3. Limited
the items have been endorsed in a consistent intellectual orally by tape ad-
manner. Once it has been determined that the ability. ministration (see
items have been endorsed consistently, the Chapter 2). Dahl-
strom and col¬
clinician then can proceed to the next step to
leagues (197 2) re¬
assess the accuracy of item endorsement. If ported that tape
the items have not been endorsed consistently, administrations are
the clinician will need to ascertain the reasons effective with IQs as
for the inconsistency in item endorsement. low as 65.
4. Too confused 4. Readminister the
A summary of the potential causes of in¬
psychiatrically MMPI-2 when the
consistency in item endorsement is presented or neuropsy- client is less con-
in Table 3-23 as well as some possible solu¬ chologicaily. fused.
tions for these problems. Most of these solu¬ 5. Still toxic 5. Readminister the
tions allow for the MMPI-2 to be readminis¬ from sub- MMPI-2 when the
stance client is detoxified.
tered so that a consistent pattern of item
abuse.
endorsement can be obtained and the clini¬ 6. Noncom- 6. Be sure client under-
cian then can proceed with the assessment of pliant or un- stands the impor-
the accuracy of item endorsement. Conse¬ cooperative. tance of the MMPI-2
quently, this step insures that the items have for treatment/inter¬
vention and read¬
been endorsed consistently once any prob¬
minister the MMPI-2.
lems have been identified and corrected. If the client is still
noncompliant, that
issue becomes the
ACCURACY OF ITEM focus of treatment.
ENDORSEMENT
After item omissions and consistency of item
endorsement have been checked, the next
step in the process of assessing the validity of
 Validity Indexes and Validity Configurations 77

the MMPI-2 is to verify the accuracy of item end of the dimension and underreporting at
endorsement (see Figure 3-1). Accuracy of the other (see Figure 3-2). Consequently, ac¬
item endorsement verifies whether the client curate patterns of item endorsement grad¬
has adopted a response set either to over¬ ually will shade into overreporting or under¬
report (“fake-bad,” malinger, make socially reporting of psychopathology as one moves
undesirable responses, etc.) or underreport up or down this dimension; there is no exact
(“fake-good,” defensiveness, make socially point at which the client’s performance sud¬
desirable responses, etc.) either the presence denly reflects either overreporting or under¬
or severity of psychopathology. reporting of psychopathology. Instead, a
As noted in Chapter 1, the terms over¬ probability statement can be made that this
reporting and underreporting of psychopa¬ client’s performance has a particular likeli¬
thology will be used throughout this book hood of reflecting either overreporting or un¬
rather than the terms indicated parentheti¬ derreporting of psychopathology.
cally since a client’s motivation for over¬ Second, it will be assumed that clients
reporting or underreporting may range from who are endorsing the items inaccurately will
being very conscious and intentional to being overreport or underreport psychopathology
out of awareness and unconscious. Since the in general rather than a specific mental disor¬
client’s test data reveal only that the items der or a set of symptoms. It is very difficult
have been endorsed inaccurately, it is neces¬ for clients to take the MMPI-2 in an accurate
sary to determine the client’s motivation for manner as if they have a specific mental dis¬
inaccurate item endorsement from a clinical order, that has been documented frequently
interview and a review of the client’s reasons (see Gough Dissimulation Scale and Simula¬
for taking the MMPI-2. tion as Role Playing below). The interested
Several issues about overreporting and clinician is encouraged to take the MMPI-2
underreporting of psychopathology must be with a specific mental disorder in mind and
made explicit before the scales and indexes see how well the scales and indexes designed
for assessing accuracy of item endorsement to assess accuracy of item endorsement detect
are discussed. First, it will be assumed that it.
overreporting and underreporting represent Third, the presence of overreporting or
an unitary dimension that is characterized by underreporting of psychopathology cannot
the overreporting of psychopathology at one be taken as evidence that the client does or

FIGURE 3-2

Psychopathology Endorsement Psychopathology

78 Chapter 3

does not have actual psychopathology, since scale, the K scale, and the F-K index. In addi¬
a client who actually has some specific mental tion to these measures, the Positive Malinger¬
disorder can overreport or underreport psy¬ ing Scale (Cofer et al., 1949) can be used to
chopathology. The scales and indexes to as¬ assess underreporting of psychopathology on
sess accuracy of item endorsement cannot de¬ the MMPI. Only the additional measures for
termine whether the client actually has assessing underreporting of psychopathology
psychopathology—only whether the client will be discussed here. The L scale and the K
has provided an accurate self-description. scale will be reviewed below in the section on
Finally, the scales used to assess the con¬ the standard validity scales.
sistency of item endorsement (TR index and The reader should realize that it is not
CLS on the MMPI; VRIN, F, and FB on the necessary to score all of these methods for as¬
MMPI-2) are not appropriate to assess the ac¬ sessing overreporting and underreporting of
curacy of item endorsement (Gallucci, 1984), psychopathology for every client. Several
and the scales/indexes used to assess the ac¬ methods will be illustrated within each re¬
curacy of item endorsement are not appropri¬ sponse set and the relative advantages and
ate to assess the consistency of item endorse¬ disadvantages of each will be provided. The
ment (Rogers, 1983). These two steps in reader will need to decide which method is
assessing the validity of an specific adminis¬ most appropriate for his or her specific treat¬
tration of the MMPI-2 need to be understood ment setting and clients. Overreporting and
as independent events using the scales/in¬ underreporting of psychopathology will be
dexes that are appropriate at each step. examined in turn.
In assessing the accuracy of item en¬
dorsement, it is more efficient to discuss pro¬
Overreporting of Psychopathology
cedures for assessing overreporting of psy¬
chopathology and then underreporting since
Wiener and Harmon Obvious and Subtle
the same scales and indexes do not always
Subscales (MMPI-2 and MMPI)
work for both response sets.
Overreporting of psychopathology on Examining endorsements to obvious versus
the MMPI-2 can be assessed by the Wiener subtle items has shown some promise in de¬
and Harmon (Wiener, 1948) Obvious and tecting overreporting and underreporting of
Subtle subscales, critical items (cf. Lachar & psychopathology. In the early research in this
Wrobel, 1979), the/7and FB scales, and the area, Wiener and Harmon (Wiener, 1948)
F-K Dissimulation index. In addition to these performed a rational inspection of MMPI
measures, the Gough Dissimulation Scale items, identifying obvious items as those that
(Gough, 1954, 1957) can be used to assess they thought were easy to detect as indicating
overreporting of psychopathology on the emotional disturbance, and subtle items as
MMPI. All of these methods for assessing those that were relatively difficult to detect as
overreporting of psychopathology except for reflecting emotional disturbance. This proce¬
the F and FB scale will be discussed within dure resulted in the identification of 146 obvi¬
this section. The F and FB scale will be re¬ ous and 110 subtle items.
viewed below in the section on the standard The empirically determined deviant re¬
validity scales. sponse for 65 (59 percent) of these subtle
Underreporting of psychopathology can items was in the opposite direction from what
be assessed by the Wiener and Harmon (Wie¬ would be expected by merely inspecting item
ner, 1948) Obvious and Subtle subscales, crit¬ content, whereas only 8 (5 percent) of these
ical items (cf. Lachar & Wrobel, 1979), the L obvious items were scored in the opposite di-
 Validity Indexes and Validity Configurations 79

rection. These findings substantiate the sub¬ and T scores near 50 on all five subtle scales is
tle and obvious nature of these two groups of trying to overreport. The converse relation¬
items, respectively. Although Wiener and ship between scores on the obvious and subtle
Harmon had intended to develop obvious subscales should arouse the suspicion of a un¬
and subtle subscales for each clinical scale, it derreporting response set. Since it will be as¬
was possible to do so for only five scales: sumed that overreporting and underreporting
Scales 2(D), 3(Hy), 4(Pd), 6(Pa), and 9(Ma). of psychopathology are a general process,
Thus, the total score on each of these five one method for creating a criterion to assess
scales can be divided into an obvious score these response sets would be to sum the dif¬
and a subtle score, which could be evaluated ferences between the obvious and subtle sub¬
as to their respective contributions to the scales. This procedure is illustrated in Table
total score. The items on the obvious and 3-24.
subtle subscales for each of these five scales The T scores for each of the obvious and
appear in Appendix A of this text. subtle subscales has been calculated, their
The other clinical scales were composed difference determined on each clinical scale,
primarily of obvious items so it was not pos¬ and these differences have been totaled into a
sible to develop obvious and subtle subscales. single overall measure. The first client has a
These clinical scales include the scales that re¬ total T score difference of +207, which
quire the most AT-correction (Scales l[Hs], strongly suggests that overreporting of psy¬
7[Pt], and <5[Sc]). Wiener and Harmon’s pre¬ chopathology has occurred. The second cli¬
liminary work also suggested that elevation ent has a total T score difference of -99,
of the obvious scales tended to predict failure which strongly suggests that underreporting
in school or vocational training, whereas the of psychopathology has occurred.
subtle scales were not significantly related to Before the use of this total T score dif¬
these criteria. ference to assess accuracy of item endorse¬
Research on the Wiener and Harmon ment is explored, several issues must be ad¬
Obvious and Subtle subscales (Wiener, 1948) dressed. First, there are a number of obvious
has demonstrated their usefulness in identify¬ and subtle subscales that could be used to as¬
ing both overreporting and underreporting of sess accuracy of item endorsement (see the
psychopathology (Anthony, 1971; Greene, section on Obvious and Subtle subscales later
1988b; Harvey & Sipprelle, 1976; Hyer et al., in this chapter). The Wiener and Harmon
1988; Walters, 1988a; Walters, White, & (Wiener, 1948) Obvious and Subtle subscales
Greene, 1988). A general pattern for the ob¬ were selected because they have the longest
vious scales to be more elevated than the sub¬ history of usage in the MMPI field. The high
tle scales when the clients were overreporting degree of item overlap among the various ob¬
psychopathology was apparent in these stud¬ vious and subtle subscales and their high cor¬
ies. It is difficult, however, to integrate the relations suggest that any of these obvious
results into any type of decision rule since no and subtle subscales would work equally
more than any two of these studies used the well.
same index of overreporting. Second, the question of whether large T
The available research does not suggest score differences on the individual clinical
explicit criteria for defining an overreporting scales have any significance has not been ex¬
response set based on the obvious and subtle plored (e.g., it is not clear whether a T score
subscales as was noted above. It probably is difference of +30 points on Scale 2 [Depres¬
safe to assume that a client who achieves T sion] has the same meaning as + 30 points on
scores of 80 or more on all five obvious scales Scale 9 [Hypomania].
80 Chapter 3

TABLE 3-24 Assessing Accuracy of Item Endorsement by the Total T Score Difference on
the Obvious and Subtle Subscales

Client 1 Client 2

Scale T Score Difference T Score Difference

2 (D)
Obvious 98 50
Subtle 28 + 70 74 -24
3 (Hy)
Obvious 95 41
Subtle 45 + 50 64 -23
4 (Pd)
Obvious 98 45
Subtle 55 + 43 66 -21
6 (Pa)
Obvious 83 48
Subtle ' 52 + 31 67 -19
9 (Ma)
Obvious 82 37
Subtle 69 + 13 49 -12
Total T Score Difference + 207 -99

Note: Wiener and Harmon's (Wiener, 1 948) Obvious and Subtle subscales were used in these
examples.
\

Finally, it must be explicit that these ob¬ mean total T score difference in normal sam¬
vious and subtle subscales are not being used ples is in the range of 0 to -30, whereas in psy¬
to predict specific external criteria since it is chiatric samples it is in the range of 50 to 60.
reasonably well-known that the obvious Psychiatric samples do score slightly higher
scales are better predictors of most criteria on this index than normals, which is to be ex¬
than subtle scales (see the section on Obvious pected since they should be acknowledging
and Subtle subscales later in this chapter). In¬ the presence of some form of psychopathol¬
stead the total T score difference between the ogy that will increase their score on the obvi¬
obvious and subtle subscales is being used as ous subscales. The psychiatric samples also
an index of the accuracy of item endorse¬ are more variable on this index with their
ment. This usage of the difference between standard deviations about 50 percent larger
the obvious and subtle subscales is in the same than the normal samples. The distributions
vein as the first approach to assess test-taking appear to be relatively normal in all of the
attitudes outlined by Meehl and Hathaway various samples. There appear to be few gen¬
(1946), which was described in Chapter 1. der or age (adolescent versus adult) differ¬
Tables 3-25 and 3-26 present the distri¬ ences in any of the samples.
bution of this total T score difference on the It must be understood that a total T
Wiener and Harmon (Wiener, 1948) Obvious score difference above whatever criterion is
and Subtle subscales for normal and psychi¬ used to identify overreporting of psychopa¬
atric samples by gender, respectively. The thology must be used as presumptive rather
 Validity Indexes and Validity Configurations 81

TABLE 3-25 Distribution of the Total T Score Difference between Obvious and Subtle
Subscales on the MMPI for Normal Samples by Gender

Greene (1986)

Adults College Students

Total T Score Male Female Male Female

Differences (N = 163) tN = 238) (N = 208) tN = 224)

276- 300 0.0% 0.0% 0.0% 0.0%

251- 275 0.0 0.0 0.0 0.0
226- 250 0.0 0.0 0.0 0.0
201- 225 0.0 0.0 0.5 0.0
1 76- 200 0.0 0.0 0.9 0.0
151- 175 0.0 0.0 1.0 0.9
126- 1 50 0.6 1.3 3.9 1.3
101- 1 25 2.5 1.6 3.8 3.6
76- 100 6.1 2.6 7.2 4.0
51- 75 8.6 3.3 12.0 9.4
26- 50 10.4 8.8 12.0 1 2.5
1- 25 10.5 6.8 16.9 17.0
-24- 0 12.8 1 7.6 1 7.8 13.8
-49- -25 14.8 16.0 12.5 18.8
-74- -50 22.0 18.2 7.2 1 1.6
-99- -75 6.8 14.7 2.9 3.5
- 1 24- -100 4.9 8.8 1.4 3.6
- 149- - 1 25 0.0 0.4 0.0 0.0

M -11.9 -30.5 19.1 0.2

SD 59.1 56.4 61.7 56.8

than definitive evidence of this overreporting The setting in which the client is taking
response set. It always is necessary to verify the MMPI will have a significant effect on the
that the client is overreporting psychopathol¬ range of scores that are seen on this index.
ogy rather than actually experiencing severe Table 3-27 summarizes the percentile equiva¬
psychopathology (cf. Schretlen, 1990). When lents for this index in several different popu¬
the total T score difference exceeds +200, it lations. It appears that normal adults obtain
should be readily apparent by an interview slightly higher scores on this index than per¬
whether the client is overreporting psychopa¬ sons taking the MMPI as part of a job appli¬
thology, since if the client’s responses were cation process, and neither group is likely to
accurate, there is severe psychopathology. report psychopathologic symptoms. In fact,
Occasionally in an inpatient setting, total T the job applicants are less likely to report any
score differences above + 200 will be seen in a form of psychopathology since it might re¬
client who is endorsing the items accurately. duce the probability of their obtaining a posi¬
In an outpatient setting, it is much less likely tion.
for difference scores in this range to reflect For example, a total T score difference
actual psychopathology since the client of +80 is at the 95th percentile in normal
should be so devastated that he or she would adults and the 99th percentile in the job ap¬
be unable to function. plicants, whereas it is between the 50th and
82 Chapter 3

TABLE 3-26 Distribution of the Total T Score Difference between Obvious and Subtle
Subscales on the MMPI for Psychiatric Samples by Gender

Psychiatric Patients (Hedlund & Won Cho, 1979)

Adults Adolescents

Total T Score Male Female Male Female

Differences (H = 8,646) (N = 3,743) (N = 693) (N = 290)

276- 300 0.3% 0.1 % 0.1 % 0.0%

251 - 275 0.7 0.2 0.9 0.0
226- 250 1.7 1.0 1.5 1.0
201- 225 2.8 1.9 2.7 2.4
1 76- 200 3.9 3.4 4.3 4.2
151- 175 5.3 5.2 4.8 6.9
126- 1 50 6.9 7.0 6.9 7.9
101- 125 8.9 9.1 10.1 13.8
76- 100 10.2 1 1.4 10.3 8.3
51- 75 10.9 10.9 8.6 9.6
26- 50 10.8 1 1.6 11.1 10.0
1- 25 9.7 10.7 12.6 4.2
-24- 0 8.8 9.2 11.1 10.0
-49- -25 7.8 7.1 7.2 8.9
-74- -50 5.7 5.7 3.9 6.2
-99- -75 3.4 3.6 2.3 3.2
-124- -100 1.8 1.3 1.5 2.7
-149- - 1 25 0.4 0.6 0.1 0.7

M 56.5 51.8 58.4 55.5

SD 85.1 79.7 81.8 85.7

75th percentile in psychiatric samples (see 1985), bipolar disorder, manic patients (Da¬
Table 3-27). Thus, higher scores on this vies et al., 1985), and a general sample of psy¬
index occur more frequently in psychiatric chiatric patients (Hedlund & Won Cho,
samples who both are more likely to have sig¬ 1979). It should be evident that these groups
nificant psychopathology and have some of patients have very similar distributions on
potential motivation to report it. The this index. Even though the first two groups
reader also should note the large number of of patients met all DSM-III criteria for
patients who achieve negative numbers on their respective diagnoses, a substantial
this index that suggest the underreporting number of them can be considered to have
of psychopathology. This issue will be ex¬ endorsed the items inaccurately either by
plored later. overreporting or underreporting psychopa¬
In order to emphasize the statement that thology.
accuracy of item endorsement can be inde¬ When a client’s responses have been
pendent of whether the client actually has identified as reflecting the overreporting of
psychopathology, Table 3-28 reports the psychopathology, the standard profile is no
scores on this index for a sample of schizo¬ longer interpretable since it reflects an over¬
phrenic patients (Davies, Nichols, & Greene, reporting response set. The clinician should
 Validity Indexes and Validity Configurations 83

TABLE 3-27 Percentile Equivalents for the Total T Score Difference between Obvious
and Subtle Subscales on the MMPI by Population

Nuclear Power
Plant Personnel Normal Adults Psychiatric Patients

(Lavin, 1984) (Greene, 1986) (Hedlund & Won Cho, 1979)

Male Female Male Female Male Female

Percentile (N = 1,031) fN = 146) fN - 163) (N - 238) (N = 8,646) (N = 3, 760)

99 80 80 128 133 250 229

95 42 28 80 84 203 184
90 25 12 70 47 172 1 58
75 -9 -20 33 -4 1 1 5 107
50 -45 -53 -22 -33 53 50
25 -80 -86 -58 -74 -8 -7
10 - 1 14 -117 -77 - 102 -55 -55
5 -132 -134 -100 -103 -79 -79
1 -140 - 140 -122 -116 - 1 1 1 - 1 1 5

describe the client’s style of overreporting vantages and disadvantages for using this
psychopathology, determine the potential index of the total T score difference between
causes for this response set, and assess the the obvious and subtle subscales as a means
implications for treatment/intervention; of assessing the accuracy of item endorse¬
the clinician should not attempt to inter¬ ment. The reader will need to compare the
pret the codetype or any of the individual advantages and disadvantages of this index
scales. with the other means of assessing accuracy of
Table 3-29 summarizes the relative ad¬ item endorsement described below to deter-

TABLE 3-28 Percentile Equivalents for the Total T Score Difference between Obvious and
Subtle Subscales on the MMPI by Diagnostic Group

Schizophrenic Bipolar Disorder,

Psychiatric Patients Patients Manic Type, Patients

(Hedlund & Won Cho, 1979) (Davies et al., 1985) (Davies et al., 1 985)

Male Female
CO
Ol

Percentile fN = 8,646) (N = 3,760) fN = 83)

II

99 250 229 239 204

95 203 184 195 1 57
90 1 72 1 58 185 144
75 1 1 5 107 133 96
50 53 50 81 41
25 -8 -7 1 1 -23
10 -55 -55 -51 -75
5 -79 -79 -96 -87
1 - 1 1 1 -115 - 1 22 -116
84 Chapter 3

TABLE 3-29 Relative Advantages and Disadvantages of the Total T Score Difference
between Obvious and Subtle Subscales to Assess Accuracy of Item Endorsement

Advantages Disadvantages

1. Sensitive to extent or degree of overre¬ 1. Too complex to score by hand because of

porting and underreporting of psycho¬ large number of scoring templates re¬
pathology so a single index can be used quired and transforming raw scores to
to assess the accuracy of item endorse¬ T scores and obtaining difference score.
ment.
2. Some research data that supports sensi¬ 2. Cannot check for inaccuracy past item
tivity of obvious and subtle subscales to 361 since the last 206 items are not
inaccuracy of item endorsement. scored on these scales.
3. Relatively easy to separate genuine psy¬
chopathology from overreporting.

mine the usefulness of this measure in his or any critical item set could be used to assess
her particular setting. accuracy of item endorsement, and the clini¬
cian may prefer to use another set of these
items that are described in Chapter 5. Re¬
Lachar and Wrobel (19 79) Critical Items gardless of the set of critical items that are
(MMPI-2 and MMPI) used, the rationale for assessing accuracy of
Despite the inherent difficulties in under¬ item endorsement will remain the same.
standing responses to individual MMPI items Lachar and Wrobel (1979) developed
(difficulties that provided the original impe¬ their critical items to be face-valid (obvious)
tus for the empirical selection of items on the descriptors of psychological concerns. They
MMPI), clinicians have been unwilling to ig¬ first identified 14 categories of symptoms
nore the information that might be contained that summarized problems that motivate peo¬
in those responses. The original set of “criti¬ ple to seek psychological treatment and that
cal” items, which were thought to require help the clinician make diagnostic decisions.
careful scrutiny if answered in the deviant di¬ Then 14 clinical psychologists read each
rection, was rationally or intuitively selected MMPI item and nominated items that would
by Grayson (1951). Grayson’s early work on be face-valid indicators of psychopathology
critical items has since been followed by the in one of these 14 categories. These items
development of other sets of critical items were empirically validated by contrasting
(see Chapter 5). Since these critical items item response frequencies for normals and
have obvious or face valid item content psychiatric samples matched for gender and
(Wrobel & Lachar, 1982), they provide an¬ race. Lachar and Wrobel were able to vali¬
other means of assessing the accuracy of item date 130 of the 177 items nominated.
endorsement. After eliminating 19 items that were
The Lachar and Wrobel (1979) critical highly duplicative of item content in other
items will be used to illustrate this procedure; items on the list, they arrived at a final list of
 Validity Indexes and Validity Configurations 85

111 (20.2 percent) critical items out of a pos¬ these critical items. The psychiatric patients
sible 550 MMPI items. endorse more of these items, as would be ex¬
The total number of Lachar and Wrobel pected. They also are more variable in en¬
critical items that are endorsed by the client dorsing these items, with standard deviations
can become another index of the accuracy of almost twice as large as the normal individu¬
item endorsement. A client who is trying to als. It is interesting to note that almost 20 per¬
overreport psychopathology would be ex¬ cent of these psychiatric patients endorse
pected to endorse a large number of these fewer total critical items than the average
items, whereas a client who is trying to under¬ normal individual. Again, it should go with¬
report psychopathology would be expected to out saying that overreporting and underre¬
endorse few of them. Tables 3-30 and 3-31 porting of psychopathology do not indicate
summarize the total number of Lachar and whether a person has psychopathology, only
Wrobel critical items that are endorsed by that the person does not endorse the items ac¬
normal individuals and psychiatric patients, curately.
respectively. Table 3-32 summarizes the relative ad¬
The normal individuals endorse 17 to 26 vantages and disadvantages of using the total
(16 to 23 percent) of these items on the aver¬ number of Lachar and Wrobel (1978) critical
age, whereas the psychiatric patients endorse items endorsed as a means of assessing the ac¬
36 to 40 (32 to 36 percent). There do not ap¬ curacy of item endorsement. Again, the
pear to be any gender or age (adolescent ver¬ reader will need to decide which of these var¬
sus adult) differences in the endorsement of ious indexes of the accuracy of item endorse-

TABLE 3-30 Distribution of Total Number of Lachar and Wrobel (1979) Critical Items
Endorsed in Normal Samples by Gender

Colligan et al. (1983) Greene (1986)

Adults Adults College Students

Total -- ---——-—— ——-——-
Critical Male Female Male Female Male Female
Items (N = 646) (N = 762) (N = 163) (N = 238) (N = 208) (N = 224)

91 + 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%

81- 90 0.0 0.0 0.0 0.0 0.0 0.0
71- 80 0.0 0.0 0.0 0.0 0.5 0.4
61- 70 0.3 0.2 0.0 0.0 0.9 0.0
51- 60 0.6 0.4 2.5 2.1 6.8 2.3
41 - 50 1.7 2.4 6.1 1.3 8.6 5.8
31- 40 6.5 6.7 8.0 13.4 1 7.8 1 7.4
21- 30 21.2 18.0 20.8 18.5 22.1 26.8
1 1- 20 40.7 40.5 33.8 40.8 35.1 37.0
0- 10 29.0 31.8 28.8 23.9 8.2 10.3

M 16.9 16.5 19.4 18.6 26.4 23.4

SD 9.9 10.1 12.5 10.9 14.1 1 1.9

Note: These are the total number of critical items endorsed, not T scores.
86 Chapter 3

TABLE 3-31 Distribution of Total Number of Lachar and Wrobel (1979) Critical Items
Endorsed in Psychiatric Samples by Gender

Psychiatric Patients (Hedlund & Won Cho, 1979)

Adults Adolescents
Total
Critical Male Female Male Female
Items (N = 8,646) fN = 3,743) (N = 693) fN = 290)

91 + 0.3% 0.3% 0.7% 0.0%

81-90 1.8 1.4 2.6 1.7
71-80 3.9 3.2 4.8 4.2
61-70 6.9 7.3 6.6 1 1.7
51-60 10.7 12.7 13.0 13.8
41-50 14.3 18.0 16.2 19.3
31-40 18.6 18.7 16.7 1 5.2
21-30 20.3 19.3 21.7 14.4
1 1-20 17.3 14.1 13.9 16.9
0-10 5.9 5.0 3.8 2.8

M 36.5 38.0 39.2 40.5

SD 19.3 18.3 19.5 19.3

Note: These are the total number of critical items endorsed, not T scores.

ment is most appropriate for his or her spe¬ oped by combining two of the three tradi¬
cific clients and treatment setting. tional validity scales. The reader is cautioned
not to confuse the F-K index (Gough Dissim¬
ulation index) with the Gough Dissimulation
scale (Ds: Gough, 1954), which will be de¬
F-K Index (Gough Dissimulation Index)
scribed in the next section. The Ds is a set of
(MMPI-2 and MMPI)
empirically derived items designed to assess
Another validity indicator, the F-K index overreporting of psychopathology, whereas
(Gough Dissimulation index), has been devel¬ the F-K index utilizes the relationship be-

TABLE 3-32 Relative Advantages and Disadvantages of the Lachar and Wrobel Critical Items
to Assess Accuracy of Item Endorsement

Advantages Disadvantages

1. Easy to score by hand. 1. Limited research to evaluate how clients

with actual psychopathology endorse
these items.
2. Assesses accuracy of item endorsement 2. Requires administration of all 567 items,
throughout the test (last item scored is
#471).
3. Can be used to assess overreporting and
underreporting of psychopathology.
 Validity Indexes and Validity Configurations 87

tween the standard validity scales of Fand K structed 40 male nonpsychotic clients to exag¬
to assess overreporting. gerate their presenting psychopathology
Gough (1947, 1950) suggested that this when they took the MMPI a second time.
index—the raw score of the F scale minus the Using the cutoff points on this index sug¬
raw score of the K scale—would be useful in gested by Gough (1947), Anthony reported
screening MMPI profiles for accuracy of that fewer than half (16) of the 40 exagger¬
item endorsement. If the F-K index was ated profiles were identified by the F-K
greater than +9, the profile was designated index. Almost one-half (17) of the exagger¬
as overreporting (i.e., the client was trying to ated profiles had a raw score on the Fscale of
feign the presence of psychopathology). If less than 15, which would directly account
the F-K index was less than 0, the profile was for the limited success of the F-K index in
classified as underreporting (i.e., the client identifying the exaggerated profiles since the
was trying to deny the existence of any form raw score on the F scale is close to 9 even be¬
of psychopathology). Intermediate scores on fore the raw score on the K scale is sub¬
the F-K index (0 to 9) indicated accurate item tracted.
endorsement (i.e., valid profiles). Gough Anthony (1971) found that even with
(1950) reported that the F-K index readily de¬ using different cutoff scores on the F-K
tected overreporting profiles; in one sample it index, he could only correctly classify 75 per¬
accurately classified 97 percent of the authen¬ cent of the profiles. Thus, the usefulness of
tic profiles and 75 percent of the overreport¬ the F-K index seems to be more limited with
ing profiles. clients who are exaggerating their psychopa¬
Most studies of the F-K index in identi¬ thology than with students. Since the former
fying overreporting profiles on the MMPI problem is the one faced in a clinical situa¬
have utilized normal persons who were in¬ tion, the F-K index at best may screen a few
structed to feign psychopathology. Numer¬ such clients.
ous investigators working with students Tables 3-33 and 3-34 summarize the dis¬
(Cofer et al., 1949; Exner, McDowell, Pabst, tribution of scores for the F-K index (Gough
Stackman, & Kirk, 1963; Hunt, 1948) have Dissimulation index) in normal individuals
confirmed the ability of the F-K index to and psychiatric patients, respectively. The
identify students who are instructed to over¬ normal individuals achieve mean scores of
report psychopathology; some of these inves¬ nearly -10 on this index, whereas the psychi¬
tigators (Exner et al., 1963; Hunt, 1948), atric patients achieve mean scores near 0. If
however, also noted that the F scale alone F-K scores greater than -I-10 are said to be
identified overreporting profiles even more overreporting psychopathology, almost 25
efficiently than the F-K index. percent of the adolescent psychiatric patients
Gallucci (1984) found that scores on the and 16 percent of the adult psychiatric pa¬
F-K index corresponded directly with the pre¬ tients would be so classified.
sumed motivation of veterans who were un¬ Since the F scale also may reflect the
dergoing psychiatric evaluations to determine presence of actual psychopathology, it would
their eligibility for disability benefits. These be expected that this index would have a high
veterans also seemed to dissimulate psycho¬ false positive rate (clients who are said to be
pathology in general rather than any specific overreporting psychopathology who are ac¬
psychiatric diagnosis. tually experiencing significant psychopathol¬
Working with clients referred for psy¬ ogy). A cutting score much higher than +9
chological evaluation, Anthony (1971) found would need to be used on the F-K index to de¬
the F-K index to have limited success. He in¬ crease the number of false positives; it is not
98 Chapter 3

ity scales. After reviewing the relative advan¬ scales/indexes with independent measures of
tages and disadvantages of each of these in¬ underreporting.
dexes, the reader should select the index Table 3-43 provides the percentile equiv¬
that is most appropriate for his or her spe¬ alents among these four scales/indexes of un¬
cific clients and treatment setting. The se¬ derreporting of psychopathology so that the
lection of one of these scales/indexes of reader can compare them directly. A total T
underreporting is more difficult than for score difference of -55 on the Wiener and
overreporting, since they appear to be mea¬ Harmon (Wiener, 1948) Obvious and Subtle
suring slightly different aspects of underre¬ subscales in female psychiatric patients is
porting as indicated by their relatively low equivalent to a T score of 65 on the Mp
intercorrelations. scale (Cofer et al., 1949), a total of 14
For example, the correlations between Lachar and Wrobel (1979) critical items en¬
the Wiener and Harmon (Wiener, 1948) dorsed, and a -15 on the F-K index (Gough,
Obvious and Subtle subscales and the 1947).
Lachar and Wrobel (1979) critical items When a client’s responses have been
and the Mp scale (Cofer et al., 1949) in identified as being endorsed inaccurately be¬
Hedlund and Won Cho’s (1979) psychiatric cause of underreporting of psychopathology,
samples ranged between —.30 and —.36. the standard profile is no longer interpretable
The correlations between the Mp scale and since it reflects an underreporting response
the F-K index (Gough, 1947) also were very set. The clinician will have little reason to try
low, ranging between —.18 and —.26 in to interpret such a profile, however, since ex¬
these same samples of patients. Research is treme underreporting results in no clinical
needed that examines which of these scales being elevated over a T score of 65 (T
scales/indexes of underreporting of psycho¬ score of 70 on the MMPI) and frequently no
pathology is most appropriate in a particular clinical scales are above a T score of 60. The
clinical setting and which validates these clinician should describe the client’s style of

TABLE 3-43 Percentile Equivalents for Four Scales/Indexes of Underreporting

Psychopathology in Adult Psychiatric Patients (Hedlund & Won Cho, 1979) by Gender

Total T Score Mp Scale Total Critical

Difference3 (T Score) Items Endorsed15 F-K Index

Male Female Male Female Male Female Male Female

(N = (N = (N = (N = (N = (N = (N = (N =
Percentile 8,646) 3,743) 8,646) 3,743) 8,646) 3,743) 8,646) 3,743)

99 -1 1 1 -1 1 5 77 81 5 5 -23 -22
95 -79 -79 69 70 9 10 -18 -18
90 -55 -55 64 65 13 14 -16 -15
75 -8 -7 59 59 21 23 - 10 - 10
50 53 50 48 48 33 36 -3 -3
25 1 1 5 107 41 40 49 50 5 5
10 1 72 1 58 36 34 64 63 1 5 14
5 203 184 33 32 72 70 21 19
1 250 229 28 26 85 84 32 30

aSee Table 3-24 for the procedure for calculating this total T score difference.
The Lachar and Wrobel (1979) critical items are used in this example.
 Validity Indexes and Validity Configurations 99

underreporting psychopathology, determine TABLE 3-44 Potential Causes of

the potential causes for this response set, and and Solutions for Underreporting
assess the implications for treatment/inter¬ of Psychopathology
vention. If the clinician can appreciate that
Cause Solution
an underreported profile is not interpretable
because of a response set since no clinical Client is not the 1. If the client can
scales are elevated, the parallel situation that identified pa- begin to trust
overreported profiles are equally uninterpret¬ tient (i.e., the cli- the clinician it
ent's spouse or may be possible
able because of a response set since most or
another family to readminister
all clinical scales are elevated may become member is the the MMPI-2.
more apparent. identified pa¬
Once these various scales and indexes to tient) and wants
identify underreporting of psychopathology to convince the
clinician that he
have been described, the next step in the im¬
or she does not
plementation of the flowchart in Figure 3-1 have any prob¬
can be made. The criteria, summarized in lems.
Table 3-45 for the MMPI-2 and in Table 3- Client believes 2. May be no solu-
that underre¬ tion.
46 for the MMPI, provide the quantitative
porting symp¬
data necessary to determine whether the toms is neces¬
items have been endorsed inaccurately (i.e., sary to obtain
to underrreport psychopathology). Once it some desired
has been determined that the client has un¬ outcome such as
derreported psychopathology, the clinician a personnel po¬
sition, transfer
will need to ascertain the reasons for the inac¬ to another
curate item endorsement by a clinical inter¬ agency or insti¬
view. tution, and so
A summary of the potential causes of on.
Client believes 3. Clients with Sub-
underreporting of psychopathology as well as
that he or she stance Depen-
some possible solutions for these problems does not have dence Disorders
are presented in Table 3-44. any problems, sometimes can
Once clinicians realize that underreport¬ which is encoun- be encouraged
ing of psychopathology is encountered fre¬ tered fre- to be more rigor-
quently in Anti- ously honest and
quently in a clinical setting, the empirical cor¬
social and have the MMPI-2
relates of such a response set can be studied. Narcissistic Per¬ readministered.
It may be that clients who underreport psy¬ sonality Disor¬
chopathology see their problems as less trou¬ ders, Manic
bling to themselves and hence are less moti¬ Mood Disorders,
vated to change. Their problems also may be and many Sub¬
stance Depen¬
more chronic in nature and consequently they dence Disorders.
may be more difficult to treat if they remain
in treatment. Duckworth and Barley (1988)
have provided a summary of the correlates
of clients who produce such underreported
profiles, which should be consulted by the
interested reader. The reader also should re¬
view the discussion of K+ profiles below,
90 Chapter 3

They presented 45 MMPI items to clinical scale had a higher hit rate in identifying these
psychologists, graduate students in psychol¬ clients who were overreporting psychopa¬
ogy, and undergraduates, and asked them to thology than did either the raw score on the F
indicate the clinical scale on which each item scale or the F-K index, although the differ¬
was found. There were no differences be¬ ences in hit rate were not statistically signifi¬
tween the groups in their ability to indicate cant. Thus, the Ds scale appears to deserve
the scale on which the item appeared, and all serious attention in its utility to identify over¬
groups could accurately identify only about reporting of psychopathology on the MMPI.
four or five items. Tables 3-36 and 3-37 summarize the
These results are not surprising, since range of scores that are seen on Ds-r for nor¬
MMPI scales were formed empirically, not mal individuals and psychiatric patients on
rationally. Because these persons could not the MMPI, respectively. T scores of 70 or
guess which scales contained various items, higher occur in about 10 percent of the nor¬
the implication is that they also would be un¬ mal individuals, 30 percent of the adult psy¬
able to distort their responses to simulate ac¬ chiatric patients, and 40 percent of the ado¬
curately those of a client who actually had a lescent psychiatric patients. If a T score of 90
specific form of psychopathology and or higher on the Ds-r scale were used to indi¬
achieve an elevated score on the appropriate cate overreporting of psychopathology, 6 to 9
scale. percent of the psychiatric patients would be
In investigating the Ds scale, Anthony classified as endorsing the items inaccurately
(1971) reported that it had an optimum hit with little difference between the adult and
rate of 86 percent in identifying male clients adolescent patients, whereas none of the nor¬
with nonpsychotic diagnoses referred for psy¬ mal adults and 0.5 percent of the college stu¬
chological testing who were instructed to ex¬ dents would be so classified. The reader can
aggerate their presenting symptoms. The Ds see what would happen if other cutting scores

TABLE 3-36 Distribution of T Scores for the MMPI Gough Dissimulation Scale —Revised
(Ds-r) in Normal Samples by Gender

Colligan et al. (1983)a Greene (1986)

Adults Adults College Students

T Score Male Female Male Female Male Female

Range (N = 646) (N - 762) (N = 163) (N = 238) tN - 208) (N = 224)

90 + 0.3% 0.3% 0.0% 0.0% 0.5% 0.4%

80-89 1.0 0.6 0.0 0.4 5.8 0.8
70-79 3.6 4.8 9.2 6.6 9.6 8.4
60-69 16.4 1 5.8 1 1.0 18.5 18.2 23.7
50-59 35.1 30.4 32.5 21.3 30.8 28.2
40-49 39.3 42.4 43.6 45.7 34.6 33.9
0-39 4.3 5.7 3.7 7.1 0.5 4.5

M 51.5 51.6 52.8 52.2 56.7 55.0

SD 9.7 9.8 10.3 10.2 1 1.4 10.4

8Colligan et al. (1983) scored the original Gough Dissimulation Scale.

 Validity Indexes and Validity Configurations 91

TABLE 3-37 Distribution of T Scores for the MMPI Gough Dissimulation Scale —Revised
(Ds-r) in Psychiatric Samples by Gender

Psychiatric Patients (Hedlund & Won Cho, 1979)

Adults Adolescents

T Score Male Female Male Female

Range (N = 8,646) (N = 3,743) (N - 693) (N = 290)

110 + 0.2% 0.1% 0.2% 0.0%

100- 109 2.1 1.6 3.5 0.6
90- 99 5.0 4.7 5.8 5.8
80- 89 7.7 8.4 1 1.6 12.1
70- 79 1 5.8 16.9 18.0 24.5
60- 69 17.9 22.6 20.2 22.1
50- 59 22.7 20.5 24.7 16.9
40- 49 26.1 21.9 1 5.3 16.2
0- 39 2.4 3.7 0.7 1.7

M 61.8 62.4 65.9 65.8

SD 16.3 1 5.4 16.1 14.8

were used to identify inaccurate patterns of and the Lachar and Wrobel (1979) critical
item endorsement. items range between .88 and .92 in Hedlund
At this time, there is not an empirically and Won Cho’s (1979) psychiatric samples,
derived scale on the MMPI-2 to detect over¬ whereas the correlations between the F scale
reporting of psychopathology. The develop¬ and the F-K index (Gough Dissimulation
ment of such a scale is strongly encouraged. Index: Gough, 1947) and these three
scales/indexes range from .70 to .86 in these
same samples.
Summary The lower correlations with the F scale
Four different indexes to assess overreporting and the F-K index may reflect that these two
of psychopathology have been described measures are influenced by the consistency of
above and two additional methods (the F and item endorsement and the presence of actual
Fb scales) will be described below. After re¬ psychopathology, which was conjectured
viewing the relative advantages and disad¬ above. These lower correlations would sug¬
vantages of each of these indexes, the reader gest that one of the two earlier mentioned
should select the index that is most appropri¬ measures (the total T score difference on the
ate for his or her specific clients and treat¬ Wiener and Harmon Obvious and Subtle
ment setting. It is not necessary to use several subscales or the Lachar and Wrobel critical
of these indexes simultaneously since they are items) may be a more “pure” estimate of
correlated highly and consequently are very overreporting of psychopathology. Research
redundant. For example, the correlations for that evaluates this hypothesis with indepen¬
the MMPI among the total T score difference dent estimates of overreporting of psychopa¬
on the Wiener and Harmon (Wiener, 1948) thology is clearly needed.
Obvious and Subtle subscales, the Ds-r scale Table 3-38 provides the percentile equiv¬
(Gough Dissimulation Scale: Gough, 1957), alents for these five scales/indexes on the
92 Chapter 3

TABLE 3-38 Percentile Equivalents for Five Scales/Indexes of Overreporting Psychopathology

in Adult Psychiatric Patients (Hedlund & Won Cho, 1979) by Gender

Total T Score Ds-r Scale Total Critical F Scale

Difference3 (T Score) Items Endorsed6 (Raw Score) F-K Index

Male Female Male Female Male Female Male Female Male Female
(N = (N = (N = (N = (N = (N = (N = (N = (N = (N =
Percentile 8,646) 3,743) 8,646) 3,743) 8,646) 3,743) 8,646) 3,743) 8,646) 3,743)

99 250 229 104 100 85 84 39 37 32 30

95 203 184 93 90 72 70 30 28 21 19
90 1 72 1 58 86 83 64 63 23 22 15 14
75 1 1 5 107 72 73 49 50 14 15 5 5
50 53 50 58 60 33 36 8 8 -3 -3
25 -8 -7 47 49 21 23 4 4 - 10 -10
10 -55 -55 42 43 13 14 3 2 - 16 -15
5 -79 -79 40 39 9 10 2 1 -18 -18
1 - 1 1 1 -1 1 5 35 37 5 5 0 0 -23 -22

aSee Table 3-24 for the procedure for calculating this total T score difference.
bThe Lachar and Wrobel (1 979) critical items are used in this example.

MMPI to assess the overreporting of psycho¬ though such a procedure rarely results in a
pathology so that they can be compared di¬ valid profile. It seems that once a client is mo¬
rectly. A total T score difference of 172 (the tivated for whatever reason to overreport
90th percentile) on the Wiener and Harmon psychopathology, it is very difficult for him
(Wiener, 1948) Obvious and Subtle subscales or her to endorse the items accurately in this
in these adult male psychiatric patients is administration. It is not known whether over¬
equivalent to a T score of 86 on the Ds-r scale reporting of psychopathology would persist
(Gough, 1957), a total of 64 of the Lachar across treatment settings for a particular cli¬
and Wrobel (1979) critical items endorsed, a ent.
raw score of 23 on the F scale, and a F-K Although the codetype from an over¬
index (Gough, 1947) of +15. reported profile cannot be interpreted, it is
It is important that the reader under¬ possible to ascertain whether there are empir¬
stand that once a profile has been defined as ical correlates of such profiles in a similar
reflecting the overreporting of psychopathol¬ manner to Marks, Seeman, and Haller’s
ogy, it cannot be interpreted as a valid pro¬ (1974) description of a K+ profile, which is
file. The client’s specific reasons for over¬ described below. Only a few studies have re¬
reporting psychopathology should be ported explicitly the correlates of over¬
ascertained by a clinical interview, and the reported profiles. Both Greene (1988b) and
profile can be described as reflecting such a Hale, Zimostrad, Duckworth, and Nicholas
process; however, neither the codetype nor (1986) found that clients who overreported
the individual scales can be interpreted. psychopathology were very likely to termi¬
The reader also should note that in Fig¬ nate treatment within the first few sessions;
ure 3-1 once an MMPI-2 is said to be charac¬ frequently they did not return after the initial
terized by overreporting of psychopathology session. This finding that these clients termi¬
the interpretive process stops. The client nate treatment quickly is almost exactly the
could have the MMPI-2 readministered, al¬ opposite of what might be anticipated, since
 Validity Indexes and Validity Configurations 93

these clients are sometimes described as “cry¬ TABLE 3-39 Potential Causes of
ing for help” and they would be expected to and Solutions for Overreporting
remain in treatment longer than most clients. of Psychopathology
Additional research is needed to determine
Cause Solution
whether there are other correlates of over¬
reported profiles. 1. Client is making 1. Explain to the cli-
Once these various scales and indexes to a "plea for ent that treat-
identify overreporting of psychopathology help." ment/services
have been described, the next step in the im¬ will be provided.
2. Client has a phe- 2. May not be any
plementation of the flowchart in Figure 3-1 nomenologic solution short of
can be made. The criteria, summarized in style to overre- long-term treat-
Table 3-45 for the MMPI-2 and in Table act and to be ment.
3-46 for the MMPI, give the quantitative traumatized; fre¬
quently seen in
data necessary to determine whether the
Dependent and
items have been endorsed inaccurately (i.e., Histrionic Per¬
to overreport psychopathology). Once it has sonality Disor¬
been determined that the client has over¬ ders and De¬
reported psychopathology, the clinician will pressive Mood
Disorders.
need to ascertain the reasons for the inaccu¬
3. Client is trying to 3. May not be any
rate item endorsement by a clinical interview. look psycho- solution.
A summary of the potential causes of pathologic be¬
overreporting of psychopathology are pre¬ cause litigation,
sented in Table 3-39, as well as some possible compensation,
etc., are in¬
solutions for these problems. It remains to be
volved.
determined whether it is possible to readmin¬
ister the MMPI-2 to a client who has over¬
reported psychopathology and obtain an ac¬
curate pattern of item endorsement. In any
event the clinician must remember that nei¬
ther the codetype nor the individual scales in index of overreporting and underreporting of
an overreported profile can be interpreted. psychopathology was outlined in Table 3-24.
In the situation where the client is trying to
underreport psychopathology, the T scores
Underreporting of Psychopathology
on the subtle scales will be larger than on the
obvious scales. Consequently, the total T
Wiener and Harmon Obvious and Subtle
score difference will be negative, reflecting
Subscales (MMPI-2 and MMPI)
that the client has endorsed more subtle than
The use of the total T score difference on the obvious items.
Wiener and Harmon Obvious and Subtle Tables 3-25 and 3-26 presented the dis¬
subscales to assess overreporting of psycho¬ tribution of this total T score difference on
pathology was described earlier. These scales the Wiener and Harmon (Wiener, 1948) Ob¬
also can be used to assess underreporting of vious and Subtle subscales by gender for nor¬
psychopathology. The procedure for calcu¬ mal and psychiatric samples, respectively.
lating the differences between the T scores Clearly, total T score differences in the range
for the obvious and subtle subscales and sum¬ of -75 to -150 are strongly suggestive of un¬
ming these difference scores to create an derreporting. If total T score differences of
94 Chapter 3

-75 or lower are deemed to be suggestive of below, the reader will see that a similar per¬
underreporting, 5.6 percent of the adult male centage of clients are identified.
patients (3.4 percent + 1.8 + 0.4) and 5.5 The reader is cautioned to remember
percent of the adult female patients (3.6 per¬ that the setting in which the client is taking
cent + 1.3 + 0.6) exceed this criterion. Some the MMPI-2 will have a significant effect on
3.9 percent of the adolescent male patients this index. As was shown in Table 3-27, indi¬
and 6.6 percent of the adolescent female pa¬ viduals who are taking the MMPI in a per¬
tients exceed this same criterion which sug¬ sonnel setting will have large negative num¬
gests that age per se has little effect on bers on this index since they realize that
whether a client will underreport psychopa¬ underreporting of any form of psychological
thology. problem may facilitate their selection.
The clinician may decide that a different Since the relative advantages and disad¬
criterion should be used to identify underre¬ vantages of using this index of the total T
porting of psychopathology in a clinical set¬ score difference between the Obvious and
ting. Unless there is some reason to believe Subtle subscales as an means of assessing the
that “normal” individuals may be evaluated accuracy of item endorsement were presented
in this specific setting, clients that underre¬ in Table 3-29, they will not be repeated here.
port psychopathology should be relatively
unusual since they would be expected to have
Lachar and Wrobel (1979) Critical Items
some form of psychological problem if they
(MMPI 2 and MMPI)
are in the clinical setting.
The fact that a sizable proportion of cli¬ The total number of the Lachar and Wrobel
ents achieve negative numbers on this index (1979) critical items that are endorsed can be
(see Table 3-26) suggests that underreporting used as another index of the accuracy of item
is not very unusual. Since normal adults’ endorsement. A client who is trying to under¬
mean score on this index is about -25 (see report psychopathology would be expected to
Table 3-25), a clinician could decide that any endorse few of these items since their item
client who scores less than -25 is underre¬ content is obvious (face valid) and reflective
porting psychopathology. By this criterion, of psychopathology. Tables 3-30 and 3-31
19.1 percent of the adult male patients (7.8 summarized the total number of critical items
percent + 5.7 + 3.4 + 1.8 + 0.4) and 18.3 that were endorsed by normal individuals and
percent of the adult female patients (7.1 per¬ psychiatric patients, respectively. Almost 20
cent + 5.7 + 3.6 -I- 1.3 + 0.6) would be percent of the psychiatric patients endorsed
classified as underreporting. Almost 20 per¬ fewer total critical items than the normal in¬
cent of these psychiatric patients would be dividuals who endorsed 17 to 26 (16 to 23 per¬
defined as underreporting psychopathol¬ cent) of these items on the average. That is,
ogy! nearly 20 percent of these psychiatric patients
The reader should not conclude that this endorsed fewer total critical items than nor¬
high percentage of psychiatric patients who mal individuals despite their presence in a
underreport psychopathology necessarily re¬ clinical setting. Again, the reader should note
flects some inherent flaw in this index; in¬ the similar percentage of patients who are
stead it simply reflects the large number of identified as underreporting by this index.
clients who are evaluated in a clinical setting Since Table 3-32 summarized the rela¬
and yet underreport psychopathology. As the tive advantages and disadvantages for using
other indexes and scales to assess underre¬ the total number of Lachar and Wrobel
porting of psychopathology are described (1979) critical items as a means of assessing
 Validity Indexes and Validity Configurations 95

the accuracy of item endorsement, they will psychiatric patients instructed to underreport
not be repeated here. The reader will need to psychopathology could not be detected by the
determine which of these indexes/scales to F-K index. Thus, the F-K index appears to be
assess the accuracy of item endorsement is even more limited in detecting underreport¬
most appropriate for his or her specific cli¬ ing profiles among psychiatric clients than in
ents and clinical setting. identifying overreporting profiles in such cli¬
ents; in either case its utility is questionable.
Several investigators have noted that
F-K Index (Gough Dissimulation Index)
some psychiatric clients have a difficult time
(MMPI-2 and MMPI)
in trying to underreport psychopathology.
Gough’s initial reservations about the effi¬ For example, Grayson and Olinger (1957)
ciency of the F-K index in detecting underre¬ discovered that rather than producing a nor¬
porting of psychopathology have been mal or a underreported profile with underre¬
corroborated by numerous investigators. porting instructions, their patients merely
Most studies (Cofer et al., 1949; Exner et al., changed the degree of severity or the nature
1963; Hunt, 1948; McAnulty, Rappaport, & of their behavior disorder.
McAnulty, 1985) have found extensive over¬ Similarly, Lawton and Kleban (1965)
lap in the distributions of the F-K index in found that their prison sample could not pro¬
students who took the MMPI normally and duce a normal profile when instructed to take
then retook the MMPI under underreporting the MMPI as a person not in trouble with the
instructions. Consequently, it has been diffi¬ law. Although their prisoners were able to
cult to find any specific score on the F-K lower significantly seven of the clinical scales,
index that reliably distinguishes normal stu¬ they could not alter their high-point scales.
dent profiles from their underreporting pro¬ Grayson and Olinger (1957) did find that
files. those patients in their sample who were able
Another problem with the F-K index in to simulate a normal profile with underre¬
identifying underreporting profiles is that porting instructions were more likely to re¬
anyone who is acknowledging the capability ceive an early discharge from the hospital.
to handle his or her own problems, who is Thus, additional research is needed to investi¬
well-adjusted (high raw score on K), and who gate whether normal profiles produced by cli¬
is not experiencing stress or conflict simulta¬ ents given underreporting instructions are a
neously (low raw score on F) will most likely favorable prognostic sign in other settings.
be defined as underreporting rather than nor¬ If scores of -10 or lower on the F-K
mal by this index. Thus, normal persons tak¬ index are used a criterion of underreporting
ing the MMPI-2 often will be inappropriately of psychopathology, nearly 25 percent of
classified as producing underreporting pro¬ Hedlund and Won Cho’s (1979) adult psychi¬
files. atric patients and 20 percent of the adolescent
Several studies have examined the ability psychiatric patients would be so classified
of the F-K index to detect underreporting (see Table 3-34). Again, there is the sizable
profiles in pathologic populations. Hunt (1948) percentage of psychiatric patients who are
reported that a score lower than -11 on the evaluated in a clinical setting that underre¬
F-K index correctly classified 62 percent of port psychopathology. Since normal individ¬
his prison sample who were instructed to un¬ uals routinely achieve negative scores on this
derreport psychopathology. Grayson and index (see Table 3-33), the F-K index will not
Olinger (1957) and Johnson, Klingler, and distinguish between normal individuals who
Williams (1977), however, found that their should score in this range and psychiatric pa-
96 Chapter 3

tients who are underreporting psychopathol¬ tive malingering). They found that a cutting
ogy. However, if it is known that this client score of 20 or higher correctly identified 96
should be reporting psychopathology be¬ percent of the accurate MMPIs and 86 per¬
cause of his or her presence in a treatment set¬ cent of the underreported MMPIs. A cutting
ting, the F-K index can alert the clinician to score of 20 is equivalent to a T score of 69 in
the possibility of underreporting of psycho¬ males and 73 in females in the original Min¬
pathology. nesota normative group. They also noted that
Since the relative advantages and disad¬ scores on the Mp scale tended to be related
vantages of using the F-K index to assess positively to scores on the Wiener (1948) Sub¬
accuracy of item endorsement were summa¬ tle subscales. Six of the 34 items on the Mp
rized in Table 3-35, they will not be repeated scale overlap with the L scale.
here. There has been little research reported
on the Mp scale (Cofer et al., 1949). Otto,
Lang, Megargee, and Rosenblatt (1988) re¬
ported that the Mp scale was able to identify
Positive Malingering Scale (Mp) (MMPI)
nearly 80 percent of alcoholics who were in¬
The Positive Malingering Scale (Mp: Cofer et structed to hide any problems or shortcom¬
al., 1949) was developed to identify underre¬ ings.
porting of psychopathology. Cofer and asso¬ Tables 3-40 and 3-41 summarize the
ciates asked groups of college students to range of scores that were found on the Mp
endorse the MMPI items like an emotionally scale in normal individuals and psychiatric
disturbed person (overreport) or as to make patients, respectively. T scores of 70 or higher
the best possible impression (underreport). were found in 2 to 6 percent of the normal in¬
They then identified 34 items that were insen¬ dividuals and 5 to 8 percent of the psychiatric
sitive to overreporting (negative malingering) patients. If a T score of 60 or higher was used
and yet susceptible to underreporting (posi¬ to identify underreporting of psychopathol-

TABLE 3-40 Distribution of T Scores for the MMPI Positive Malingering (Mp) Scale in Normal
Samples by Gender

Colligan et al. (1983) Greene (1986)

Adults Adults College Students

T Score Male Female Male Female Male Female

Range (N = 305) (N = 335) (N = 163) (N = 238) (N = 208) (N = 224)

80 + 0.0% 0.3% 0.0% 0.0% 0.0% 0.4%

70-79 4.6 5.7 1.8 2.5 4.3 1.8
60-69 1 5.5 21.8 1 2.9 9.7 1 2.5 7.6
50-59 36.6 23.5 46.6 24.4 33.7 32.6
40-49 34.4 37.0 31.3 41.6 38.0 37.1
30-39 8.2 10.2 7.4 1 7.2 1 1.0 16.9
0-29 0.7 1.5 0.0 4.6 0.5 3.6

M 52.2 51.5 51.7 46.8 50.6 47.6

SD 9.9 10.7 8.3 10.8 9.5 10.1
 Validity Indexes and Validity Configurations 97

TABLE 3-41 Distribution of T Scores for the MMPI Positive Malingering (Mp) Scale
in Psychiatric Samples by Gender

Psychiatric Patients (Hediund & Won Cho, 1979)

Adults Adolescents

T Score Male Female Male Female

Range (N - 8,646) (N = 3,743) (N = 693) (N = 290)

80-89 0.6% 1.4% 0.3% 1.0%

70-79 4.4 6.4 5.9 5.9
60-69 16.8 1 5.2 21.8 9.7
50-59 31.7 30.0 38.5 33.1
40-49 31.8 31.5 27.6 37.2
30-39 13.6 12.9 5.8 1 1.7
0-29 1.1 2.6 0.1 1.4

M 51.3 51.2 54.3 50.4

SD 11.1 1 2.1 9.8 1 1.0

ogy, 17 to 27 percent of the psychiatric pa¬ these advantages and disadvantages with the
tients would be classified as endorsing the other indexes and scales described in this sec¬
items inaccurately. The reader should recall tion to assess underreporting of psychopa¬
how a similar percentage of patients were clas¬ thology.
sified as underreporting by the total T score
difference on between the Wiener and Harmon
Summary
(Wiener, 1948) Obvious and Subtle subscales.
Table 3-42 summarizes the relative ad¬ Four different means of assessing the under¬
vantages and disadvantages of using the Mp reporting of psychopathology have been de¬
scale (Cofer et al., 1949) to assess the accu¬ scribed above, and two additional methods
racy of item endorsement on the MMPI. The (the L scale and the K scale) will be described
reader will need to compare and contrast below in the section on the traditional valid-

TABLE 3-42 Relative Advantages and Disadvantages of the MMPI Positive Malingering (Mp)
Scale to Detect Underreporting of Psychopathology

Advantages Disadvantages

1. Easy to score by hand. 1. Requires use of a separate scale/index to

identify overreporting of psychopath¬
ology.
2. Empirically derived with some research to 2. Requires administration of all 566 items,
document its use.
3. Uses all of the items (#556 is last item
scored).
98 Chapter 3

ity scales. After reviewing the relative advan¬ scales/indexes with independent measures of
tages and disadvantages of each of these in¬ underreporting.
dexes, the reader should select the index Table 3-43 provides the percentile equiv¬
that is most appropriate for his or her spe¬ alents among these four scales/indexes of un¬
cific clients and treatment setting. The se¬ derreporting of psychopathology so that the
lection of one of these scales/indexes of reader can compare them directly. A total T
underreporting is more difficult than for score difference of -55 on the Wiener and
overreporting, since they appear to be mea¬ Harmon (Wiener, 1948) Obvious and Subtle
suring slightly different aspects of underre¬ subscales in female psychiatric patients is
porting as indicated by their relatively low equivalent to a T score of 65 on the Mp
intercorrelations. scale (Cofer et al., 1949), a total of 14
For example, the correlations between Lachar and Wrobel (1979) critical items en¬
the Wiener and Harmon (Wiener, 1948) dorsed, and a -15 on the F-K index (Gough,
Obvious and Subtle subscales and the 1947).
Lachar and Wrobel (1979) critical items When a client’s responses have been
and the Mp scale (Cofer et al., 1949) in identified as being endorsed inaccurately be¬
Hedlund and Won Cho’s (1979) psychiatric cause of underreporting of psychopathology,
samples ranged between —.30 and —.36. the standard profile is no longer interpretable
The correlations between the Mp scale and since it reflects an underreporting response
the F-K index (Gough, 1947) also were very set. The clinician will have little reason to try
low, ranging between —.18 and —.26 in to interpret such a profile, however, since ex¬
these same samples of patients. Research is treme underreporting results in no clinical
needed that examines which of these scales being elevated over a T score of 65 (T
scales/indexes of underreporting of psycho¬ score of 70 on the MMPI) and frequently no
pathology is most appropriate in a particular clinical scales are above a T score of 60. The
clinical setting and which validates these clinician should describe the client’s style of

TABLE 3-43 Percentile Equivalents for Four Scales/Indexes of Underreporting

Psychopathology in Adult Psychiatric Patients (Hedlund & Won Cho, 1979) by Gender

Total T Score Mp Scale Total Critical

Difference3 (T Score) Items Endorsed6 F-K Index

Male Female Male Female Male Female Male Female

(N = (N = (N = (N = (N = (N = (N = (N =
Percentile 8,646) 3,743) 8,646) 3,743) 8,646) 3,743) 8,646) 3,743)

99 -1 1 1 -1 1 5 77 81 5 5 -23 -22
95 -79 -79 69 70 9 10 -18 -18
90 -55 -55 64 65 1 3 14 - 16 -1 5
75 -8 -7 59 59 21 23 - 10 - 10
50 53 50 48 48 33 36 -3 -3
25 1 1 5 107 41 40 49 50 5 5
10 1 72 1 58 36 34 64 63 1 5 14
5 203 184 33 32 72 70 21 19
1 250 229 28 26 85 84 32 30

aSee Table 3-24 for the procedure for calculating this total T score difference.
bThe Lachar and Wrobel (1 979) critical items are used in this example.
 Validity Indexes and Validity Configurations 99

underreporting psychopathology, determine TABLE 3-44 Potential Causes of

the potential causes for this response set, and and Solutions for Underreporting
assess the implications for treatment/inter¬ of Psychopathology
vention. If the clinician can appreciate that
Cause Solution
an underreported profile is not interpretable
because of a response set since no clinical 1. Client is not the 1. If the client can
scales are elevated, the parallel situation that identified pa¬ begin to trust
overreported profiles are equally uninterpret¬ tient (i.e., the cli¬ the clinician it
ent's spouse or may be possible
able because of a response set since most or
another family to readminister
all clinical scales are elevated may become member is the the MMPI-2.
more apparent. identified pa¬
Once these various scales and indexes to tient) and wants
identify underreporting of psychopathology to convince the
clinician that he
have been described, the next step in the im¬
or she does not
plementation of the flowchart in Figure 3-1 have any prob¬
can be made. The criteria, summarized in lems.
Table 3-45 for the MMPI-2 and in Table 3- 2. Client believes 2. May be no solu¬
that underre¬ tion.
46 for the MMPI, provide the quantitative
porting symp¬
data necessary to determine whether the toms is neces¬
items have been endorsed inaccurately (i.e., sary to obtain
to underrreport psychopathology). Once it some desired
has been determined that the client has un¬ outcome such as
derreported psychopathology, the clinician a personnel po¬
sition, transfer
will need to ascertain the reasons for the inac¬ to another
curate item endorsement by a clinical inter¬ agency or insti¬
view. tution, and so
A summary of the potential causes of on.
3. Client believes 3. Clients with Sub¬
underreporting of psychopathology as well as
that he or she stance Depen¬
some possible solutions for these problems does not have dence Disorders
are presented in Table 3-44. any problems, sometimes can
Once clinicians realize that underreport¬ which is encoun¬ be encouraged
ing of psychopathology is encountered fre¬ tered fre¬ to be more rigor¬
quently in Anti¬ ously honest and
quently in a clinical setting, the empirical cor¬
social and have the MMPI-2
relates of such a response set can be studied. Narcissistic Per¬ readministered.
It may be that clients who underreport psy¬ sonality Disor¬
chopathology see their problems as less trou¬ ders, Manic
bling to themselves and hence are less moti¬ Mood Disorders,
and many Sub¬
vated to change. Their problems also may be
stance Depen¬
more chronic in nature and consequently they dence Disorders.
may be more difficult to treat if they remain
in treatment. Duckworth and Barley (1988)
have provided a summary of the correlates
of clients who produce such underreported
profiles, which should be consulted by the
interested reader. The reader also should re¬
view the discussion of K+ profiles below,
100 Chapter 3

which is one form of underreporting psycho¬ underreporting of psychopathology in psy¬

pathology. chiatric settings on the MMPI-2, and Table
3-46 provides similar information for the
MMPI. The discussion in this section will be
CUTTING SCORES
limited to the MMPI-2, since the only differ¬
FOR ASSESSING VALIDITY
ences between the two tables are in the spe¬
IN PSYCHIATRIC SETTINGS
cific cutting scores used on a scale and less
Table 3-45 summarizes the cutting scores frequently the scale that is used. For example,
that can be used to assess overreporting and total T score differences on the Wiener and

TABLE 3-45 Cutting Scores for Assessing MMPI-2 Validity

Acceptable Marginal Unacceptable

1. Item Omissions
Cannot Say (?) 0- 10 11- 30 31 +
II. Consistency of Item Endorsement
Variable Response 0- 7 8- 1 5 16 +
Inconsistency Scale (VRIN)
\F — Fb\ (raw scores) 0- 6 7- 10 1 1 +
VRIN + \F - Fb| 0- 16 1 7- 20 21 +
F + Fe + \F - Fe| 0- 36 37- 49 50 +
III. Accuracy of Item Endorsement
A. Overreporting of Psychopathology
Total T Score Difference on 4- 130 131- 230 231 +
Wiener and Harmon
Obvious and Subtle
subscales
Total Lachar and Wrobel critical 21- 48 49- 68 69 +
items endorsed
B. Underreporting of Psychopathology
Total T Score Difference on + 130- -4 -5- -65 < -65
Wiener and Harmon
Obvious and Subtle
subscales
Total Lachar and Wrobel critical 48- 21 20- 9 <9
items endorsed

Note: These cutting scores are set so that approximately the 7 5th percentile separates the accept¬
able and marginal categories, and approximately the 95th percentile separates the marginal and
unacceptable categories in psychiatric patients. Clinicians may consider adjusting these cutting
scores based on the specific base rates in their clients and the relative cost of identifying a certain
percentage of clients' MMPI-2s as unacceptable. In other settings where the MMPI-2 is adminis¬
tered, such as personnel selection, appropriate cutting scores will need to be derived.
It is not intended for the clinician to score all of the scales/indexes within a given section of
this table. The clinician should select the scale/index that is most appropriate for his or her clinical
setting.
These MMPI-2 cutting scores were developed using Hedlund and Won Cho's (1 979) MMPI
data. These MMPI data were rescored after eliminating the 1 3 items that were dropped from the
standard validity and clinical scales on the MMPI-2 and these raw scores were converted to MMPI-
2 T scores. Thus, the cutting scores in this table need to be understood as being only close ap¬
proximations to actual MMPI-2 data, although there is little reason to expect for them to be signifi¬
cantly different.
 Validity Indexes and Validity Configurations 101

TABLE 3-46 Cutting Scores for Assessing MMPI Validity

Acceptable Marginal Unacceptable

1. Item Omissions
Cannot Say (?) 0- 10 11-30 31 +
II. Consistency of Item Endorsement
Test-Retest (77?) Index 0- 4 5 6+
Carelessness (CLS) Scale 0- 4 - 5 6+
Sum of 77? + CLS 0- 8 9 10 +
III. Accuracy of Item Endorsement
A. Overreporting of Psychopathology
Total T Score Difference on -7-110 111-190 191 +
Wiener and Harmon
Obvious and Subtle
subscales
Ds-r Scale (T score) <70 70- 90 91 +
Total Lachar and Wrobel critical 23- 50 51- 70 71 +
items endorsed
F Scale (T score) <78 78-104 105 +
B. Underreporting of Psychopathology

CD

■xj
CO
Total T Score Difference on + 110-- 7 < -79

1
1
Wiener and Harmon
Obvious and Subtle
subscales
Mp Scale (T score) <59 59- 70 71 +
Total Lachar and Wrobel critical 23- 50 22- 1 1 < 1 1
items endorsed

Note: These cutting scores are set so that approximately the 7 5th percentile separates the accept¬
able and marginal categories, and approximately the 95th percentile separates the marginal and
unacceptable categories in psychiatric patients. Clinicians may consider adjusting these cutting
scores based on the specific base rates in their clients and the relative cost of identifying a certain
percentage of clients' MMPIs as unacceptable. In other settings where the MMPI is administered,
such as personnel selection, appropriate cutting scores will need to be derived.
It is not intended for the clinician to score all of the scales/indexes within a given section of
this table. The clinician should select the scale/index that is most appropriate for his or her clinical
setting.

Harmon (Wiener, 1948) Obvious and Subtle 1979). These cutting scores have been estab¬
subscales between 131 and 230 are in the mar¬ lished very conservatively based on clinical
ginal range, and total T score differences of experience since limited empirical data are
231 and higher are in the unacceptable available at this time. Consequently, scores in
range for the overreporting of psychopa¬ the unacceptable range should be considered
thology. as presumptive evidence of the overreporting
The cutting scores for all indexes (Wie¬ of psychopathology. It should be readily ap¬
ner and Harmon Obvious and Subtle sub¬ parent in a clinical interview if a patient’s re¬
scales, total number of Lachar and Wrobel sponses were accurate since pervasive and se¬
critical items endorsed, and F scale) have vere psychopathology should be present.
been set at the 75th and 95th percentiles for Empirical research is needed to determine
psychiatric patients (Hedlund & Won Cho, further correlates of the overreporting of psy-
102 Chapter 3

chopathology and whether these cutting which inconsistent, overreported, and under¬
scores should be raised or lowered. reported profiles are encountered by code-
Similarly, Tables 3-45 and 3-46 summa¬ type on the MMPI. A client’s profile was de¬
rize the cutting scores that can be used to as¬ fined as being inconsistent in Table 3-47 if
sess underreporting of psychopathology on the sum of the TR index and the CLS scale
the MMPI-2 and MMP1, respectively. For exceeded nine.
example, total T score differences on the Some MMPI codetypes could be very
Wiener and Harmon (Wiener, 1948) Obvious likely to reflect inconsistent patterns of item
and Subtle subscales between -8 and -79 are endorsement. For example, 43.0 percent of
in the marginal range, and total T score dif¬ 6- 8/8-6 codetypes could result from inconsis¬
ferences of -80 and lower are in the unaccept¬ tent patterns of item endorsement! In fact,
able range. The cutting scores for all indexes most codetypes that include Scale 8 (Schizo¬
(Wiener and Harmon Obvious and Subtle phrenia) frequently could result from incon¬
subscales and total number of Lachar and sistency. Some codetypes rarely result from
Wrobel critical items endorsed) have been set inconsistent patterns of item endorsement
at the 25th and 5th percentiles for psychiatric (i.e., most Spike codetypes and codetypes
patients. Since normal individuals average that include Scale 3 [Hysteria]). Table 3-47
around -25 on the total T score difference on shows that 16 of the 55 codetypes (29.1 per¬
the Wiener and Harmon Obvious and Subtle cent) could result from inconsistency 10 per¬
subscales, it seems plausible to assume that cent or more of the time.
psychiatric patients who score below -25 are Table 3-47 also indicates the frequency
underreporting psychopathology. This exam¬ with which overreported and underreported
ple illustrates how conservatively the present profiles occur by MMPI codetype. A client’s
cutting scores have been set. Research is profile was defined as reflecting overreport¬
needed to establish more precisely the cutting ing of psychopathology in Table 3- 47 if the
scores that should be used to indicate the un¬ total T score difference on the Wiener and
derreporting of psychopathology. Harmon (Wiener, 1948) Obvious and Subtle
Research is also needed to determine subscales exceeded 110 (75th percentile), and
whether a specific scale or index of over¬ as underreporting if this same index was less
reporting and underreporting is the most ap¬ than -7 (25th percentile). Both of these cut¬
propriate for a given clinical setting, and ting scores reflect the boundaries between ac¬
whether raising or lowering these proposed ceptable and marginal accuracy of item en¬
cutting scores would facilitate the identifica¬ dorsement in Table 3-46. Although these
tion of overreporting or underreporting. The cutting scores have been selected somewhat
establishment of the base rates with which arbitrarily, they do allow comparisons
overreporting and underreporting are en¬ among the codetypes as to the frequency with
countered in a specific clinical setting is man¬ which overreporting and underreporting psy¬
datory in any research that examines these chopathology could occur.
cutting scores. Clinicians need to be aware of Some MMPI codetypes could be very
the frequency with which overreporting and likely to reflect overreporting of psychopa¬
underreporting occur in their clinical setting thology with four codetypes (1-8/8-1, 2-8/
and begin to establish the empirical correlates 8-2, 6-8/8-6, and 7-8/8-7) exceeding the cutting
of these response sets so that better assess¬ score over 40 percent of the time. An addi¬
ments, treatments, and interventions can be tional six codetypes (1-6/6-1, 1-7/7-1, 2-7/
made. 7- 2, 4-878-4, 6-777-6, and 8-0/0-8) exceed the
Table 3-47 provides the frequency with cutting score over 20 percent of the time. Un-
TABLE 3-47 Frequency of Inconsistent and Inaccurate Profiles on the MMPI by Codetype

Inaccurate

Inconsistent3 Underreporting13 Overreportingc Total

Codetype Number Percent Number Percent Number Percent Number Percent

6-818-6 535 43.0 14 1.1 51 1 41.1 1 243 85.3

1-8/8- 1 1 14 36.8 16 5.2 1 28 41.3 310 83.2
7-818-7 1 78 27.5 12 1.9 288 44.5 647 73.9
Spike 4 0 0.0 537 71.4 1 0.1 752 71.5
2-818-2 142 22.2 21 3.3 289 45.2 639 70.7
Spike 3 0 0.0 31 70.5 0 0.0 44 70.5
3-515-3 1 1.4 46 64.8 0 0.0 71 66.2
Spike 5 0 0.0 74 65.5 0 0.0 1 13 65.5
3-919-3 0 0.0 44 62.0 0 0.0 71 62.0
4-818-4 198 19.3 145 14.1 266 25.9 1026 59.4
8-919-8 191 32.8 37 6.3 109 18.7 582 57.9
3-616-3 4 6.3 26 41.3 4 6.3 63 54.0
3-414-3 1 3 2.7 242 50.2 3 0.6 482 53.5
1-7/7- 1 7 1 7.1 3 7.3 1 1 26.8 41 51.2
4-515-4 4 1.3 146 48.5 2 0.7 301 50.5
3-818-3 1 1 1 5.7 1 7 24.3 7 10.0 70 50.0
2-313-2 20 7.3 98 35.6 16 5.8 275 48.7
2-616-2 18 10.5 27 1 5.8 33 1 9.3 171 45.6
3-0/0-3 0 0.0 5 45.5 0 0.0 1 1 45.5
6- 7/ 7- 6 7 1 2.5 5 8.9 13 23.2 56 44.6
1-2/2- 1 66 1 2.2 46 8.5 96 1 7.7 485 42.9
1-4/4- 1 27 10.0 81 30.1 7 2.6 269 42.8
1-3/3- 1 22 6.5 1 14 33.7 8 2.4 338 42.6
5-616-5 5 8.2 19 31.1 2 3.3 61 42.6
Spike 7 0 0.0 12 36.4 2 6.1 33 42.4
1-6/6- 1 6 14.6 2 4.9 9 22.0 41 41.5
Spike 6 0 0.0 39 38.2 2 2.0 102 40.2
Spike 9 0 0.0 1 35 38.9 3 0.9 347 39.8
8-010-8 3 6.3 4 8.3 12 25.0 48 39.6
Spike 8 0 0.0 18 33.3 3 5.6 54 38.9
6-919-6 37 1 3.9 32 1 2.0 34 1 2.7 267 38.6
3- 7/ 7- 3 1 3.8 9 34.6 0 0.0 26 38.5
4-616-4 52 8.8 1 25 21.2 50 8.5 590 38.5
5-919-5 3 1.7 61 35.3 1 0.6 1 73 37.6
5-818-5 1 1 16.2 5 7.4 9 1 3.2 68 36.8
2- 7/7-2 49 10.1 23 4.7 104 21.4 487 36.1
Spike 1 0 0.0 26 34.2 1 1.3 76 35.5
2-919-2 8 8.8 22 24.2 2 2.2 91 35.2
2-414-2 68 6.3 245 22.8 63 5.9 1074 35.0
4-919-4 33 3.0 307 28.3 29 2.7 1083 34.1
2-515-2 1 1.1 24 25.5 7 7.4 94 34.0
5-010-5 0 0.0 12 27.9 1 2.3 43 30.2
4-7/ 7-4 20 7.4 44 16.3 1 7 6.3 270 30.0
1-919- 1 4 4.2 12 1 2.6 12 1 2.6 95 29.5
Spike 2 0 0.0 61 26.8 4 1.8 228 28.5
7-0/0-7 0 0.0 4 10.5 5 1 3.2 38 23.7
7-919-7 7 8.8 6 7.5 5 6.3 80 22.5
Spike 0 0 0.0 1 1 20.4 1 1.9 54 22.2
6-010-6 1 1.8 10 1 7.5 1 1.8 57 21.1
9-0/0-9 2 5.3 6 1 5.8 0 0.0 38 21.1
5- 7/ 7- 5 0 0.0 4 16.0 1 4.0 25 20.0
4-0/0-4 1 0.8 21 16.5 2 1.6 1 27 1 8.9
1-010- 1 0 0.0 2 10.5 1 5.3 19 1 5.8
2-0/0-2 4 1.4 23 8.3 14 5.1 276 14.9
1-5/5- 1 0 0.0 3 1 2.5 0 0.0 24 1 2.5
Total 1874 1 3.2 31 14 22.0 2189 1 5.5 14,149

alnconsistent was defined as the sum of 77? and CLS greater than 9.
bUnderreporting was defined as the total T score difference on the Wiener and Harmon (Wiener, 1 948) Obvious
and Subtle subscales less than —7 (25th percentile).
cOverreporting was defined as the total T score difference on the Wiener and Harmon (Wiener, 1 948) Obvious
and Subtle subscales greater than +110 (75th percentile).

103
104 Chapter 3

derreporting of psychopathology could even totype scores for each codetype are dis¬
be more frequent with five codetypes (Spike cussed.
3, 3-5Z5-3, 3-9/9-3, Spike 4, and Spike 5) ex¬
ceeding the cutting score over 60 percent of
IMPRESSION MANAGEMENT
the time. An additional 15 codetypes exceed
AND SELF-DECEPTION
the cutting score over 30 percent of the time.
Finally, Table 3-47 summarizes the fre¬ Paulhus (1984, 1986) has proposed a two-fac¬
quency with which invalid profiles (i.e., both tor model of socially desirable responding in
inconsistent and inaccurate patterns of item en¬ self-reports of personality that provides an¬
dorsement) occur by MMPI codetype. Since other means of looking at how clients may
the codetypes are ranked by the total percent¬ produce inaccurate patterns of item endorse¬
age of invalid profiles, it is easily seen that ment. His model distinguishes between self-
16 codetypes are invalid by these criteria deception (where clients believe their positive
over 50 percent of the time and 6 codetypes self-reports) and impression management
are invalid over 70 percent of the time. The (where clients consciously dissemble to create
lowest ranked codetype (7-5/5-/) is still in¬ a favorable impression in others).
valid 12.5 percent of the time. The necessity Paulhus (1986) hypothesized that self-
for checking the consistency and accuracy of deception refers to a motivated unawareness
item endorsement for all clients should be of one of two conflicting cognitions, whereas
readily apparent. impression management can be conceptual¬
Table 3-48 provides the relative ranking ized as a strategic simulation, a motive, or as
of all codetypes by the total T score differ¬ a skill. Paulhus (1986) suggested that
ence on the Wiener and Harmon (Wiener, Edwards’ Social Desirability Scale (ESD: Ed¬
1948) Obvious and Subtle subscales for the wards & Diers, 1962) assessed self-deception
MMPI-2 and the MMPI. It is apparent that on the MMPI, whereas the L (Lie) Scale, the
the rankings of the codetypes are very similar Positive Malingering Scale (Mp\ Cofer et al.,
on the MMPI-2 and the MMPI. If over¬ 1949), and Wiggins’ Social Desirability Scale
reporting and underreporting of psychopa¬ {Wsd: Wiggins, 1959) assessed impression
thology are an unitary dimension, as was as¬ management.
sumed earlier (see page 77), Table 3-48 Only one study has examined Paulhus’
illustrates how the various codetypes distrib¬ model to date (Greene, Davis, & Welch,
ute themselves along this dimension. 1988). Greene and associates generated means
Clinicians probably will not be surprised and standard deviations on ESD (M = 25.1;
to see that clients with Spike 4 codetypes are SD = 7.4) and Wsd (M = 17.4; SD = 5.3) on
among the farthest toward underreporting of a large sample (TV = 958) of psychiatric inpa¬
psychopathology and 6-8/8-6 codetypes are tients. Patients who scored one standard de¬
the farthest toward overreporting. It is inter¬ viation above their respective means were de¬
esting, however, to see how the other fined as high self-deception or impression
codetypes rank relative to each other. For ex¬ management, and patients who scored one
ample, it seems somewhat unexpected for standard deviation below their means were
2-4/4-2 codetypes to be ranked so low and defined as low self-deception or impression
for 1-6/6-1, 7-7/7-/, and 1-8/8-1 codetypes management.
to be ranked so high. This information on Greene and colleagues (1988) reported
the relative elevation of a number of validity that there was a trend for more male (70 per¬
and supplementary scales within each codetype cent) patients to be included in the high self-
will be explored further in Chapter 6 where pro¬ deception groups and for more female pa-
TABLE 3-48 Ranking by Codetype for the Total T Scale Difference for Obvious
and Subtle Subscales

MMPI MIVfPI-2

Codetype M SD Codetype M SD

6-818-6 153.7 73.3 6-818-6 202.0 71.0

7-818-7 131.0 67.7 7- 8/8- 7 164.4 63.6
2-818-2 128.1 74.8 1-8/8- 1 1 55.1 73.2
1-8/8- 1 127.2 78.4 2-818-2 1 55.0 73.0
1-7/ 7-1 80.6 66.9 8-010-8 140.4 46.6
1-6/6- 1 79.3 74.5 6-717-6 1 39.0 61.7
8-919-8 76.5 67.1 1-6/6- 1 1 28.3 63.1
2-7/7-2 75.2 54.0 1-7/ 7-1 1 26.9 62.6
8-0/0-8 72.5 59.5 6-010-6 1 24.3 49.4
1-2/2- 1 67.7 62.5 3-818-3 1 23.5 86.8
6- 7/ 7- 6 62.6 65.7 8-919-8 1 21.9 63.1
4-818-4 58.2 90.8 2-616-2 1 1 9.8 63.2
2-616-2 57.2 70.6 6-919-6 1 1 5.0 62.5
7-919-7 52.6 51.0 2-7/ 7-2 111.2 55.0
5-818-5 47.4 58.7 7-0/0-7 106.3 49.6
6-919-6 46.5 61.2 4-818-4 104.7 77.8
7-0/0-7 44.3 52.3 9-0/0-9 104.7 48.3
1-0/0- 1 44.0 47.0 1-2/2- 1 98.1 64.8
1-9/9- 1 43.0 57.0 7-919-7 94.6 54.4
2-0/0-2 42.9 44.6 2-0/0-2 93.1 46.1
5- 7/ 7- 5 34.0 50.7 1-9/9- 1 88.0 61.0
1-5/5- 1 33.0 54.0 2-919-2 86.1 49.7
Spike 0 30.3 43.1 3-616-3 85.9 74.2
9-010-9 29.6 44.3 1-010- 1 85.7 41.9
6-0/0-6 28.8 51.6 3- 7/ 7-3 85.1 57.8
Spike 7 28.5 52.7 5-818-5 83.4 44.2
4-7/7-4 26.5 61.1 5- 7/ 7- 5 83.1 59.4
3-818-3 26.3 70.8 4-616-4 82.6 68.7
2-919-2 25.5 53.1 4-0/0-4 77.8 46.5
Spike 2 23.0 47.5 4-7/ 7-4 77.3 60.8
4-616-4 22.9 68.7 2-515-2 71.3 55.2
4-010-4 21.4 47.8 5-616-5 70.5 70.9
5-0/0-5 21.4 44.8 2-313-2 68.4 67.9
2-515-2 18.9 58.7 2-414-2 68.3 62.3
2-414-2 16.5 62.2 Spike 0 62.7 44.0
3-717-3 16.0 60.2 1-5/5- 1 58.3 60.7
Spike 8 1 5.9 62.1 1-3/3- 1 55.4 71.0
Spike 1 12.9 54.1 Spike 7 50.2 57.2
Spike 6 1 2.8 52.7 3-010-3 49.5 109.6
Spike 9 12.6 48.2 5-010-5 47.1 40.0
2-313-2 7.6 61.2 4-919-4 46.6 60.0
1-3/3- 1 7.1 59.8 3-919-3 42.7 46.0
4-919-4 1 .7 55.7 1-4/4- 1 40.4 67.9
5-616-5 1.1 53.6 Spike 2 36.6 50.8
1-4/4- 1 0.2 58.4 Spike 9 35.4 50.6
3-010-3 0.1 65.0 Spike 6 34.7 55.4
5-919-5 - 14.4 47.6 Spike 8 33.4 62.2
3-616-3 - 1 5.6 68.1 5-919-5 30.1 58.5
Spike 5 -23.9 53.6 Spike 1 24.6 53.3
3-919-3 -28.4 52.6 3-414-3 1 8.7 66.0
4-515-4 -30.5 52.8 Spike 5 0.8 54.3
Spike 4 -36.0 52.2 4-515-4 -2.0 53.8
Spike 3 -39.4 41.4 3-515-3 -7.6 49.9
3-515-3 -44.4 39.6 Spike 4 - 1 5.0 54.4
3-414-3 -51.3 54.5 Spike 3 - 1 8.3 51.4

105
106 Chapter 3

tients (62 percent) in the low self-deception dishonesties and denial of aggression, bad
groups. MMP1 codetypes including Scale 4 thoughts, and weakness of character. Exam¬
(Psychopathic Deviant) were particularly ples of L scale items with the deviant answer
prominent in the high self-deception groups, indicated in parentheses are:
and Scale 8 (Schizophrenia) was more likely
to be elevated in the low self-deception “At times I feel I like swearing.” (false)
groups. They noted that 4-6codetypes, which “I get angry sometimes.” (false)
are a fairly infrequent codetype, were the
“Sometimes when I am not feeling well I
most frequent (38.5 percent) in the High Self-
am irritable.” (false)2
Deception and Low Impression Management
Group.
Patients in the High Impression Man¬ The original Minnesota normative group
agement Groups scored lower consistently on answered most of the L scale items in the
group measures of intelligence. Patients in nondeviant direction; only three items—15,
the Low Impression Management Groups 135, and 165—were answered in the deviant
were much more likely to receive some form direction by a majority of this sample. In
of depressive diagnosis and patients in the Gravitz’s (1970) sample of job applicants,
High Impression Management Groups were these three items plus two more—45 and
more likely to receive a schizophrenic or 255—were endorsed in the deviant direction
manic diagnosis. There were fewer differ¬ by nearly a majority or more of the respon¬
ences seen among these groups on Axis II dents. A majority of the MMPI-2 normative
(Personality Disorder) diagnoses. Patients in group also endorsed these same three items
the Low Self-Deception and Low Impression (16[15], 123[135], and 153[ 165]) in the devi¬
Management Group were much more likely ant direction.
to receive a diagnosis in the Avoidant, De¬ Gravitz (1970) also noted small but con¬
pendent, Compulsive, or Passive-Aggressive sistent gender differences in responding to
cluster than any of the other three groups. the L scale items; this pattern would suggest
Paulhus’ model seems to produce inter¬ the need for separate T scores by gender,
esting correlates in an inpatient psychiatric which are not available on the MMPI. The T
setting that need to be replicated and ex¬ scores for the L scale as well as the other three
tended to other clinical settings. validity scales (?, F, and K) provided by
Hathaway and McKinley (1967) for the
MMPI are identical for males and females.
TRADITIONAL VALIDITY SCALES Separate T scores by gender are available for
all of the validity scales on the MMPI-2
(Butcher et al., 1989). T scores on the L scale
Lie (L) Scale tend to vary by only one to three T points be¬
The L scale includes 15 items that were se¬ tween men and women on the MMPI-2, sim¬
lected on a rational basis to identify persons ilar to those reported by Gravitz (1970) for
who are deliberately trying to avoid answer¬ the MMPI.
ing the MMPI frankly and honestly Most adults respond to all L scale items
(Dahlstrom et a!., 1972). The scale assesses (Gravitz, 1971); if any items are omitted, they
attitudes and practices that are culturally are most likely to be 255 or 285 (Gravitz,
laudable but actually found only in the most 1967). Dahlstrom and associates (1972) have
conscientious persons. Content areas within suggested that college students and adoles¬
the L scale include denial of minor, personal cents are likely to omit a few of the L scale
 Validity Indexes and Validity Configurations 107

items (e.g., items 15, 135, 165, and 255) since paranoid trends. Fjordbak (1985) reported
they may not have had experience with the that hospitalized male forensic patients who
item content. As yet, no reported research had raw scores greater than 6 on the L scale
has studied this issue, and similar informa¬ and no clinical scales elevated had some type
tion is not available for the MMPI-2. of psychotic disorder with prominent para¬
Since “false” is the deviant answer to all noid features.
L scale items, the L scale is extremely suscep¬ The higher the elevation on the L scale,
tible to unsophisticated deviant test-taking the lower will be the elevation of most clinical
sets, such as the set to answer all items scales. Denial is characteristic of high scores
“false.” Unusually high (7 or above) raw on L; denial also results in refusing to ac¬
scores on the L scale, particularly in persons knowledge the presence of any form of psy¬
for whom such a score is unexpected, should chopathology, thus lowering the elevation of
at least raise the suspicion of a deviant test¬ the clinical scales. Elevations on the L scale
taking set. The effects of such test-taking at¬ are occasionally accompanied by elevations
titudes on the various validity indicators and on Scale 1 (Hypochondriasis) and Scale 3
methods of detection of these attitudes will be (Hysteria), which appear to tap the similar
examined later in this chapter. personality dynamics of denial and a lack of
More sophisticated deviant response sets psychological mindedness. Elevations on the
may go undetected by the L scale. Inspection L scale sometimes also are associated with an
of the L scale items reveals that it is readily elevation on Scale 9 (Hypomania) in which
apparent which responses are the deviant the client displays a grandiose self-concept
ones. Numerous studies have shown that the centered around a pervasive denial of psycho¬
L scale does not detect sophisticated persons pathology.
who were given instructions to falsify their Although few researchers have directly
answers to the MMPI (Dahlstrom et al., investigated the L scale, Burish and Houston
1972; Vincent, Linsz, & Greene, 1966). These (1976) have provided some validational evi¬
persons apparently realized that it would be dence for it. In 66 male college students, they
unconvincing to give distorted responses to L found that L correlated significantly (+ .55)
scale items. with the Denial (Dn) scale (see Chapter 4,
Thus, the L scale can be construed as a page 145) and was unrelated to Scale 1 (Hy¬
measure of psychological sophistication with pochondriasis) and Scale 8 (Schizophrenia).
high scores indicating a lack of such sophisti¬ Hence, the L scale was significantly related to
cation. College-educated persons and per¬ a construct (denial) that the L scale is in¬
sons of higher socioeconomic classes rarely tended to measure (convergent validity) and
score above a raw score of 4. Conversely, per¬ unrelated to two constructs (hypochondriasis
sons who are not psychologically minded (in¬ and schizophrenia) from which the scale is in¬
cluding some persons from minority groups tended to differ (discriminant validity).
and lower socioeconomic classes) tend to Burish and Houston (1976) also found
score higher. Thus, a person’s education level that students with high L scale scores per¬
and socioeconomic class must be kept in formed better in a stressful situation than
mind when interpreting the L scale. Persons those with low scores. The high L scale stu¬
with a college education who score high on dents appeared to use their defensive strate¬
the L scale are likely to display deficiencies in gies across different kinds of potentially
judgment and lack of insight into their own threatening situations. Similarly, Matarazzo
behavior. Coyle and Heap (1965) have ques¬ (1955) found that male medical students who
tioned whether such persons might have scored high on the L scale were less anxious
108 Chapter 3

(as measured by the Taylor Manifest Anxiety The F scale is sometimes called the fre¬
scale) than those who scored low. quency or infrequency scale, but the exact
Test-retest reliability coefficients for the derivation of the label “F” is unknown. The
L scale on the MMPI tend to be slightly lower scale taps a wide variety of obvious and un¬
than those reported for Scales F and K ambiguous content areas, including bizarre
(Dahlstrom et ah, 1975). Reliability coeffi¬ sensations, strange thoughts, peculiar experi¬
cients for intervals up to one week range from ences, feelings of isolation and alienation,
.70 to .85, and for intervals of one year or and a number of unlikely or contradictory be¬
more range from .35 to .60. Little informa¬ liefs, expectations, and self-descriptions
tion is available for test-retest reliability on (Dahlstrom et al., 1972). Examples of F scale
the MMPI-2. Test-retest reliability coeffi¬ items with the deviant answer indicated in pa¬
cients for the L scale on the MMPI-2 are .77 rentheses are:
for men and .81 for women over a one-week
interval (Butcher et al., 1989) “When I am with people, I am bothered
Hathaway and McKinley (1951) originally by hearing very strange things.” (true)
arbitrarily assigned T scores to the raw scores
“No one cares much what happens to
for the L scale on the MMPI. Inspection of
you.” (true)
Profile 1-1 will show that they assigned a raw
score of 10 on the L scale to a T score of 70 “I believe in law enforcement.” (false)3
and a raw score of 7 to a T score of 60. On the
basis of their clinical experience of the fre¬ Most of the Fscale items (35 of 60) on the
quency with which the various raw scores on MMPI-2 are scored only on the F scale; 15
the L scale occur, Hathaway and McKinley items overlap with Scale 8 (Schizophrenia)
(1967) have suggested that a raw score of 7 on and 9 items with Scale 6 (Paranoia). Eight of
the L scale should equal a T score of 70, and the Fscale items (12, 48, 120, 132, 204, 222,
Rosen (1952) has indicated that a raw score of 264, and 288) do not meet the 10 percent or
10 should equal a T score of 80. T scores for the below criterion for either males or females in
L scale on the MMPI-2 were computed in the the MMPI-2 normative sample, and an addi¬
standard manner; they correspond reasonably tional 4 items for males do not meet this crite¬
well with those suggested by Hathaway and rion (Butcher et al., 1989). There also are 69
McKinley (1967) for the MMPI L scale. more items on the MMPI-2 that meet the 10
The levels of L scale elevation and their percent or below criterion; 37 of these items
interpretation are summarized in Table 3-49. are found on the FB scale (see below).
Five of theFscale items (20, 54, 112, 115,
and 185) on the MMPI do not meet the 10
FScale
percent or below criterion for either males or
The F scale consists of 60 items (64 items on females in the original Minnesota normative
the MMPI) that were selected to detect un¬ sample, and an additional 11 items for
usual or atypical ways of answering the test males and 3 for females do not meet this
items. Unlike most of the other scales, the F criterion (Dahlstrom et al., 1975). Most of
scale was not derived by comparing item en¬ these items, however, exceeded the criterion
dorsements between criterion and normal by only a few percentage points. There also
groups; it is made up of items that no more are 38 more items that meet the 10 percent or
than 10 percent of an early subsample of the below criterion that could have been included
Minnesota normative sample answered in the in the F scale but were not for unknown rea¬
deviant direction. sons.
 Validity Indexes and Validity Configurations 109

TABLE 3-49 Interpretations of Lie (L) Scale Elevation

MMPI-2 MMPI
T Score Raw Score3 Interpretation

44 and below 0-2 1.Low. A set to endorse all items as "true" is possible.
The other validity indicators should be evaluated. Cli¬
ents may be attempting to create an extremely patho¬
logic picture of themselves. Normal persons who are
relatively independent or self-reliant are generally willing
to admit these minor social faults.
45-55 3-5 2. Normal. Scores in this range indicate clients who are
able to achieve an appropriate balance between admit¬
ting and denying minor social faults. These clients may
be sophisticated persons who are attempting to create
a favorable self-image.
56-64 6-7 3. Moderate. A random sort may have occurred. The
other validity indicators should be evaluated. Scores in
this range may indicate normal persons who are slightly
more conforming than usual or clients who have a tend¬
ency to resort to denial mechanisms.
65 and 8-15 4. Marked. An error in scoring may have occurred on the
above MMPI when this scale is scored without a template (i.e.
"true" responses were counted instead of the "false"
responses). Scores in this range may indicate: normal
persons who are very self-controlled and who lack in¬
sight into their own behavior, persons with religious
and moralistic training or occupations that deny even the
most common human faults, unsophisticated persons
who are trying to create an unusually favorable impres¬
sion of themselves as in personnel selection, clients
whose dynamics revolve around denial (frequently en¬
countered in histrionic and somatization disorders), or
psychiatric inpatients who may be overtly psychotic
when all the clinical scales are not elevated (T scores
<65 on the MMPI-2; T scores <70 on the MMPI).

aSince the raw scores arbitrarily assigned by Hathaway and McKinley (1 967) to the T scores on
the L scale on the MMPI are too low, in this table the raw scores have been adjusted based on
clinical judgment to reflect more accurately the frequency with which specific raw scores are
obtained.

Gravitz (1987) reported that normal job concluded that these changes in the frequency
applicants endorsed a number of the MMPI of item endorsement indicate that it may be
Fscale items more frequently than the 10 per¬ necessary to renorm the MMPI. These same
cent criterion of the original normative sam¬ items also were endorsed more than 10 per¬
ple. More than 10 percent of the men and cent of the time by the MMPI-2 normative
women endorsed MMPI Group Booklet sample except for item 206, which was
items 112, 115, 199, and 206 in the deviant di¬ dropped.
rection. The men also endorsed item 215 Researchers have investigated the rela¬
more than 10 percent of the time. Gravitz tionship between F scale elevations, validity
110 Chapter 3

of the profile, and extent of psychopathol¬ able to understand proverbs, were monosyl¬
ogy. Hathaway and McKinley (1951) origi¬ labic, had delusions of reference, had audi¬
nally recommended that MMPI profiles tory hallucinations, were disoriented for
should be ruled invalid if the T score on the F place, had short attention spans, and did not
scale exceeds 70 (raw score >12). Research¬ know why they were hospitalized.
ers quickly showed, however, that all profiles Essentially the white patients with a T
with a T score greater than 70 on the F scale score greater than 98 on the F scale could be
were not invalid. Kazan and Sheinberg (1945) described as confused psychotics, and their
reported that 35 of 37 male mental hygiene extreme profile elevations reflected the sever¬
clinic patients with a T score greater than 70 ity of their psychopathology. No descriptors
on the F scale were providing valid self¬ could be cross-validated among the black pa¬
descriptions on the MMPI. Similarly, tients (i.e., there were no replicable differ¬
Schneck (1948) believed that 10 of 17 male ences between black patients with a T score
prisoners with a T score on the/7scale greater greater than 98 versus those with a T score
than 70 were providing accurate responses, equal to or less than 98). Obviously, the F
and their elevation on the F scale appeared to scale was tapping different dimensions in
reflect the severity of their personality distur¬ black as compared to white patients. More
bance. information on the performance of blacks
The next flurry of research investigated and other minority groups on the MMPI will
whether profiles with a T score on the F scale be provided in Chapter 8.
greater than 80 (raw score > 16) should be Gynther, Lachar, and Dahlstrom (1978)
considered invalid. Preliminary research in have developed a new F scale for blacks that
this area indicated that persons with a T score is designed to serve the same validity function
greater than 80 were likely to be diagnosed as as the standard F scale (i.e., to identify per¬
having behavior disorders (Gynther, 1961; sons who endorse items that members of a
Gynther & Shimkunas, 1965a) in court-re¬ normal population infrequently endorse).
ferred cases and as being psychotic in psychi¬ This new F scale for blacks also will be de¬
atric samples (Blumberg, 1967; Gauron, scribed in Chapter 8.
Severson, & Englehart, 1962; Gynther & Other studies have investigated clients’
Shimkunas, 1965b). responses to individual F scale items.
Again, it appears that a specific raw Gynther and Petzel (1967) observed that psy¬
score on the F scale cannot be used to con¬ chotics and persons with behavior disorders
sider profiles routinely as invalid since a ma¬ were not differentiated by their raw scores on
jority of the persons in these studies were ac¬ the F scale; they hypothesized that this oc¬
curately classified. Instead, significant curred because these persons endorsed differ¬
psychopathology tends to be correlated with ent subsets of items within the scale. They
T scores greater than 80 on the F scale, al¬ found, however, only one item that discrimi¬
though the exact correlates differ as a func¬ nated the two groups; therefore, their hy¬
tion of the setting in which the MMPI is pothesis was not supported.
given. McKegney (1965) observed that delin¬
Finally, Gynther, Altman, and Warbin quent adolescents have elevated/7scale scores
(1973b) examined the correlates of F scale and hypothesized that this occurred because
scores greater than a T score of 98 (raw scores some of the F scale items are accurate re¬
>25) in white and black hospital patients. sponses for them. Hence, McKegney thought
They identified and cross-validated seven de¬ that delinquent adolescents were consistently
scriptors of the white patients: they were un¬ endorsing only certain F scale items that are
 Validity Indexes and Validity Configurations 111

meaningful for them. Three professionals in mind, however, that the content of the F
and three nonprofessionals familiar with ad¬ scale items is obvious, and clients may lower
olescents identified 21 F scale items that they or raise their F scale scores virtually as de¬
felt could be answered truthfully by delin¬ sired. Such overreporting of psychopathol¬
quent adolescents. ogy, however, can be detected easily by the
McKegney (1965) also found that 21 F indexes and scales described above.
scale items were answered more frequently by Hathaway and McKinley (1951) also ar¬
delinquent adolescents than by normal ado¬ bitrarily assigned the T scores to the raw
lescents. These items dealt with such content scores on the F scale as they did on the ? and
as stealing, misbehaving in school, and injur¬ L scales. They assigned a T score of 70 to a
ing others; the items directly tap the behav¬ raw score of 12 and a T-score of 80 to a raw
iors and attitudes that caused the adolescents score of 16 (see Profile 1-1, page 3). Again,
to be labeled delinquent. McKegney reported clinical experience has shown that these T
that these two sets of 21 items were positively scores do not properly reflect the frequency
correlated, but he did not report how many with which the specific raw scores occur on
or which items overlapped between the two the F scale. Based on clinical experience,
sets. Hathaway and McKinley (1967) later sug¬
Thus, it appears that high F scale scores gested that a raw score of 16 should equal a T
among delinquent adolescents may result at score of 70.
least partially from the fact that some items T scores for the F scale on the MMPI-2
are genuinely more applicable to juveniles were computed in the standard manner; sur¬
than to the original normative group. Archer prisingly, the raw scores do not appear to cor¬
(1984, 1987) has noted that most adolescents respond to the scores that would be expected
have higher scores on the F scale items than based on clinical experience. For example, a
adults, which also suggests that a number of raw score of 16 on the MMPI-2 F scale is a T
these items may not be infrequent items for score of 85 in men and 92 in women rather
adolescents. than approximately a T score of 70, as sug¬
Test-retest reliability coefficients for the gested by Hathaway and McKinley (1967) on
MMPI F scale range from .80 to .97 for an the MMPI.
interval up to two weeks and range from .45 The levels of F scale elevation and their
to .76 for intervals from eight months up to interpretation are summarized in Table 3-50.
three years (Dahlstrom et al., 1975). Test-re¬ Since it is not clear whether the T scores on
test reliability coefficients for the F scale on the MMPI-2 accurately reflect the frequency
the MMPI-2 are .78 for men and .69 for with which raw scores will be seen in clinical
women over a one-week interval (Butcher et settings, the levels of elevation in Table 3-50
al., 1989). have been based on raw scores rather than T
Elevation of the F scale is positively cor¬ scores.
related with the overall elevation of the entire
clinical portion of the profile and particularly
Back F (Fb ) Scale
Scales 6 (Paranoia) and 8 (Schizophrenia),
both in adult psychiatric patients and in ado¬ The Fb scale consists of 40 items on the
lescents (Dahlstrom et al., 1972). Elevation MMPI-2 that no more than 10 percent of the
of the F scale also is a rough index of the se¬ MMPI-2 normative sample answered in the de¬
verity of the psychological distress that the viant direction. This scale is analogous to the
client is experiencing; higher scores indicate standard F scale except that the items are
more severe distress. Clinicians should keep placed in the last half of the test. Item 281 is
112 Chapter 3

TABLE 3-50 Interpretations of F and Back F (FB) Scale Elevations

MMPI-2 MMPI
Raw Score8 Raw Score3 Interpretation

0- 2 0- 2 1. Low. These clients have systematically avoided ac¬

knowledging the socially unacceptable or disturbing
content represented in the scale. They may be trying to
deny serious psychopathology (underreporting). Or
they may be normal persons who are very conventional,
unassuming, and unpretentious.
3- 7 3- 7 2. Normal. These clients are willing to acknowledge a
typical number of unusual experiences.
8-15 8-15 3. Moderate. These clients are acknowledging the un¬
usual experiences represented in this scale more than
the typical person. The elevation reflects the extent and
severity of the client's psychopathology, and how the
client has adjusted to his or her psychopathology (i.e.,
an intact psychotic patient frequently falls in the middle
of this range).
16-22 16-22 4. Marked. The profile may be invalid; other validity indi¬
cators should be checked. The elevation reflects the se¬
verity of distress and extent of psychopathology that the
client is experiencing. Patients are likely to be diag¬
nosed as having behavior disorders or being psychotics,
depending on their age and type of treatment facility.
Adolescents may be honestly acknowledging the devi¬
ant behaviors that caused them to be labeled delin¬
quent.
23 and 23 and 5. Extreme. THE PROFILE IS PROBABLY INVALID. Other
above above validity indicators should be checked. These clients are
likely to be severely disorganized and psychotic; these
characteristics are readily apparent in an interview.
Scores in this range may indicate adolescents going
through an identity crisis.

aSince the raw scores arbitrarily assigned by Hathaway and McKinley (1 967) to the T scores on
the F scale on the MMPI are known to be too high, and the T scores seem too low on the F scale
and too high on the FB scale on the MMPI-2, in this table the raw scores have been adjusted based
on clinical judgment to reflect more accurately the frequency with which specific raw scores are
obtained.

the first item on the FB scale, whereas item Fscale, whereas for women a raw score of 12
361 is the last item on the standard F scale is a T score of 89 on the FB scale and a T
(see Appendix A). score of 79 on the Fscale. These T scores also
There are large differences between raw are significantly higher than clinical experi¬
scores and their corresponding T scores on ence suggested was appropriate on the F scale
the Fb and F scales on the MMPI-2 (compare on the MMPI (Hathaway & McKinley, 1967).
Profile 7-1, page 296, and Profile 7-3, page Since items on both the FB and F scales
298). For men a raw score of 16 is a T score of were endorsed by no more than 10 percent of
108 on the FB scale and a T score of 85 on the the normative sample, it is surprising that
 Validity Indexes and Validity Configurations 113

there should be such large differences in T scale is heterogeneous and covers self-control
scores between these two scales. If clients and family and interpersonal relationships.
were more likely to endorse the items ran¬ Examples of K scale items with the deviant
domly on the last half of the MMPI-2, higher answer indicated in parentheses are:
raw scores, and hence lower T scores, would
be expected on FB rather than F; however, “I like to let people know where I stand
the opposite pattern was found in the MMPI- on things.” (false)
2 normative sample. Clearly, research is
“I have very few quarrels with members
needed that examines the relationship be¬
of my family.” (true)
tween raw scores on FB and Fin clinical sam¬
ples. “People often disappoint me.” (false)4
The clinician should recall the use of the
relationship between the raw scores on the F A F-corrected profile is automatically
and Fb scales as a measure of consistency of plotted if the standard profile sheet is used
item endorsement (see page 69). (see Profile 2-1). The F-corrected profile was
Test-retest reliability coefficients for the developed by determining the proportion of
Fb scale on the MMPI-2 are .86 for men and K that, when added to the raw score on the
.71 for women over a one-week interval clinical scale, would maximize the discrimi¬
(Butcher et al., 1989). These coefficients are nation between the normative groups and the
slightly higher than those reported for the F criterion group. McKinley, Hathaway, and
scale. Meehl (1948) determined that the discrimina¬
The levels of FB scale elevation and their tions could be improved on five of the clinical
interpretation should be very similar to the scales by the addition of a proportion of K.
standard F scale since the items were selected Thus, Scales 7(Psychasthenia) and 8 (Schizo¬
by the same criterion. Consequently, the phrenia) are corrected by the addition of the
same interpretive statements are suggested whole raw score of K, whereas Scales 1 (Hy¬
for both Fb and Fscales (see Table 3-50). pochondriasis), 4 (Psychopathic Deviate),
and 9 (Hypomania) are corrected by the addi¬
tion of a fractional value of K (.5, .4, and .2,
K Scale
respectively). McKinley and associates found
The K scale consists of 30 items that were em¬ that the addition of K to the other clinical
pirically selected to assist in identifying persons scales (Scales 2 [Depression], 3 [Hysteria], 5
who displayed significant psychopathology [Masculinity-Femininity], and 6 [Paranoia])
yet had profiles within the normal range. actually reduced their discriminability; so
Since the derivation of the K scale was de¬ these scales were not F-corrected.
scribed in Chapter 1, it will not be repeated Adolescent profiles on the MMPI are
here. not K-corrected (see Chapter 8; Archer, 1987)
Most of the K scale items also are scored so the standard profile sheet cannot be used
on other clinical scales and are fairly equally directly. Special adolescent profile sheets are
dispersed across these scales; only five items available for the MMPI so that clinicians can
are unique to the K scale. The items usually plot the profile without looking up the non-
are scored in the same direction when they F-corrected T scores in a table.
appear on another clinical scale except that The F-correction procedure was not ex¬
six of the seven items also found on Scale 0 amined in the restandardization of the
(Social Introversion) are scored in the op¬ MMPI, and the same F-weights are added to
posite direction. Item content on the K the same scales on the MMPI-2. Conse-
114 Chapter 3

quently, any critiques of the K scale on the ples. Jenkins found virtually no change in the
MMPI should apply directly to the MMPI-2. accuracy of identifying pain patients with
Despite urgings for investigators to Scale 1 (Hypochondriasis) or schizophrenic
cross-validate these ^-corrections (Dahl- patients with Scale 8 (Schizophrenia) using K-
strom et al., 1972; McKinley et al., 1948; weights that ranged from -1.0 to + 1.5. Es¬
Meehl & Hathaway, 1946), almost no re¬ sentially, the AT-correction procedure had no
search has been published on the issue. The effect on the accuracy of classifying these
few published studies on /f-correction weights groups of patients. However, Wooten found
(Heilbrun, 1963; Tyler & Michaelis, 1953; a slight improvement in hit rate using the K-
Yonge, 1966) found discouraging results. correction procedure in Air Force trainees.
Heilbrun investigated the ^-corrections Wooten (1984) also noted that there
that would maximize the discrimination be¬ were frequent changes in codetype when non-
tween normal college students and malad¬ AT-corrected profiles were compared with the
justed students who either sought treatment standard AT-corrected profile. He concluded
at a university counseling center or were hos¬ that his data favored the use of the ^-correc¬
pitalized. Heilbrun determined that only tion procedure. More than 40 years of using
three scales separated normal from malad¬ the MMPI without investigation of the ap¬
justed college students better when Ar-cor- propriateness of the AT-correction procedure
rected than when not AT-corrected. He found defies explanation. Hopefully, researchers
that the following weights, different in some will begin to investigate the AT-correction pro¬
cases for males than for females, worked cedure on the MMPI-2.
best: Scale 3 (Hysteria), -.IK males, ~.5K fe¬ Researchers have examined other as¬
males; Scale 7 (Psychasthenia), \.0K males, pects of the K scale than the /^-corrections
.SK females; and Scale 8 (Schizophrenia), .7 themselves. Several investigators (Heilbrun,
K males and females. Heilbrun did cross-val¬ 1961; Smith, 1959; Sweetland & Quay, 1953)
idate these weights in his student sample; examined the appropriateness of the scale as
since no additional research has been con¬ a measure of defensiveness and reported that
ducted in other student samples, the the K scale in a normal population is a mea¬
generalizability of his results remains open to sure not of defensiveness but of personality
question. integration and healthy adjustment, with
Both Tyler and Michaelis (1953) and high scores reflecting healthy adjustment.
Yonge (1966) reported that adding K to the Similarly, both Tyler and Michaelis
five A^-corrected scales in their college student (1953) and Yonge (1966) reported that in a
samples actually reduced the reliability and normal college student sample the K scale
validity of these scales. Clearly, any clinician was significantly negatively related to the five
using the MMPI-2 or MMPI in a college setting /^-corrected scales on which high scores do
needs to examine the usage of the traditional K- indicate psychopathology. Working with
correction procedures. Using Heilbrun’s pro¬ maladjusted college students, Heilbrun
posed AT-corrections, determining another more (1961) and Nakamura (1960) found, how¬
appropriate set of /^-corrections, or avoiding the ever, that the K scale was measure of defen¬
use of AT-corrections altogether might be prefer¬ siveness. Consequently, it appears that the
able. The need for additional research in other appropriateness of interpreting AT as a mea¬
settings cannot be overemphasized. sure of defensiveness varies according to the
Two recent studies (Jenkins, 1984; type of client. In a normal population high
Wooten, 1984) have evaluated the effective scores on the K scale do not indicate defen¬
of the /^-correction procedure in clinical sam¬ siveness; in a maladjusted population, how-
 Validity Indexes and Validity Configurations 115

ever, high K scores do suggest defensive¬ need to keep in mind is that clients can
ness. achieve a high score on a clinical scale that is
Research on the K scale in other settings ^-corrected in different ways. They can ei¬
also has yielded dismal results. Hunt, Carp, ther endorse a large number of items in the
Cass, Winder, and Kantor (1948), Schmidt deviant direction on the clinical scale or have
(1948), and Wooten (1984) reported that the a large AT-correction added.
/^-correction contributed little to diagnostic For example, a female client can achieve
efficiency for patients in a military setting. a total AT-corrected raw score of 40 on Scale 8
Silver and Sines (1962) also found that the K- (Schizophrenia) of the MMPI-2 in a variety
correction did not increase the accuracy of of ways, including endorsing 15 Scale 8 items
predicting diagnostic classification in state and 25 K scale items or 35 Scale 8 items and 5
hospital patients; the K scale was essentially K scale items. The non-/f-corrected T scores
unrelated to diagnostic classifications. Ruch for these two raw scores on Scale 8 are 56 and
and Ruch (1967) found that non-AT-corrected 78, respectively (see Appendix B). It should
clinical scales discriminated better between go without saying that in the latter case she is
good and poor salesmen; the AT-corrections more likely to be overtly schizophrenic than
actually decreased the discriminability of the in the former. This is not to suggest that ei¬
two groups. ther the A'-corrected or non-ATcorrected
The one positive outcome in this area score is more accurate. The point is that cli¬
was provided by Ries (1966), who compared ents achieving such scores will be very dif¬
state hospital patients who scored higher than ferent both behaviorally and clinically, and
a raw score of 15 (T score greater than 55) on clinicians need to be aware of these differ¬
the K scale with patients who scored 15 or ences.
lower. For patients with raw scores of 16 or Low scores on the K scale (the client con¬
higher, 19 of 22 were rated as being unim¬ sistently admits problems) are accompanied
proved after 60 days of hospitalization and 7 by more frequent elevations of the clinical
of 19 were rehospitalized within 12 months. profile, especially in the psychotic tetrad
For patients scoring 15 or lower, 25 of 31 (Scales 6, 7, 8, and 9). High K scale scores are
were rated as being improved after the same associated with lower profile elevations and
time interval, and only 2 of 25 were rehospi¬ peaks on the neurotic triad (Scales /, 2, and
talized within 12 months. 3) both in adult psychiatric and in adolescent
Clearly, in Ries’ study the K scale was populations (Dahlstrom et al., 1972).
significantly related to an external criterion. The K scale is the only validity scale on
In general, however, little research justifies the MMPI for which the T scores were de¬
the continued widespread use of the Af-cor- rived in a standard manner. The K scale
rection of the clinical scales. Hopefully, fu¬ also is the only validity scale for which
ture research will investigate this area more there is no specific score that indicates that
thoroughly. Meanwhile, clinicians probably the profile is invalid. The clinician, how¬
need to avoid using AT-corrections in settings ever, should be sensitive to K scores that are
in which normal persons are being evaluated atypical of clients taking the MMPI-2 or
with the MMPI-2, but they should use the K- the MMPI in a specific setting.
corrections in settings in which psychopathol¬ For example, in personnel selection very
ogy is suspected, keeping in mind the poten¬ low K scores would be unusual, and clinicians
tial inaccuracies that /^-corrections may should review such scores closely. Con¬
introduce. versely, in psychiatric settings very high K
Another consideration that clinicians scores would be unusual. When a client with
116 Chapter 3

known or suspected psychopathology has a over a one-week interval (Butcher et al.,

highly elevated score on K, the client is likely 1989).
being defensive about some form of psycho¬
logical distress; the cause of the distress may
VALIDITY SCALE CONFIGURATIONS
not be discernible from the profile. In this sit¬
uation it is usually recommended that the cli¬ Four validity scale configurations occur fre¬
ent be evaluated carefully for an underlying quently in most clinical and normal popula¬
psychotic process, particularly if the clinical tions. This limited number of common
scales are within the normal range (see Nor¬ validity scale configurations is surprising
mal K+ profiles, page 121). since 27 possible configurations exist if each
The interpretation of the AT scale changes of the three validity scales is classified as av¬
dramatically depending on the socioeco¬ erage, above average, or below average.
nomic class and education level of the client The most frequently encountered valid¬
and the setting in which the MMPI-2 is ad¬ ity scale configuration in most clinical set¬
ministered (e.g., personnel selection, state tings is illustrated in Figure 3-3 (solid line).
hospital, university). The potential impact The frequency with which this configuration
of these factors on the K scale becomes occurred in four samples of clients is pre¬
even more noteworthy on the MMPI-2 be¬ sented in Table 3-53.
cause of the relatively high socioeconomic The essential characteristics of this con¬
class and years of education that character¬ figuration are that the L scale and the K scale
ized the MMPI-2 normative group. Conse¬ are below a T score of 50 and the F scale is
quently, clinicians should be aware that above a T score of 60. The client is admitting
the T scores on the K scale on the MMPI- to personal and emotional difficulties, is re¬
2 may have been affected by these factors questing assistance with these problems, and
(cf. Caldwell, 1990). These factors appear is unsure of her own capabilities for dealing
to have minimal effect on the standard with these problems. As the F scale score in¬
clinical scales that are AT-corrected creases, the client is either experiencing more
(Butcher, 1990). problems and hence feeling worse, exaggerat¬
Clinicians should be cautious in interpre¬ ing the symptomatology in order to get help
ting scores on the K scale in clients whose oc¬ sooner, or simulating psychopathology. The
cupation or education level is subtantially clinician can use the other validity indexes de¬
lower than the MMPI-2 normative group scribed earlier in this chapter to distinguish
until further research has addressed the po¬ which of these various causes of the F scale
tential impact of these factors. Levels of K elevation is appropriate.
scale elevation and their interpretation for Under most conditions, this validity
persons with known or suspected psychopa¬ scale configuration is most desirable for any
thology are summarized in Table 3-51. Sim¬ form of psychological intervention or treat¬
ilar information for “normal” persons is ment. There are two reservations, though:
summarized in Table 3-52. Test-retest reli¬ The F scale should not be above a T score of
ability coefficients for the K scale on the 90-100, and the AT scale should not be below a
MMPI range from .78 to .92 for an interval T score of 35. In the former case the client
up to two weeks and range from .52 to .67 may be experiencing so much stress and con¬
for intervals from eight months to three flict that a psychotherapeutic intervention
years (Dahlstrom et al., 1975). Test-retest re¬ should not be initiated until some of the stress
liability coefficients for the K scale on the and conflict can be alleviated. In the latter
MMPI-2 are .84 for men and .81 for women case the client may not have the necessary
 Validity Indexes and Validity Configurations 117

1 ABLE 3-51 Interpretations of K Scale Elevations for Clients with Suspected

or Known Psychopathology

MMPI-2 MMPI
T Score T Score Interpretation

30-34 27-35 1. Markedly Low. These clients probably have either fab¬
ricated or greatly exaggerated their problems to create
the impression of a severe emotional disturbance (over¬
reporting). They may be experiencing acute psychotic
distress, which may require hospitalization. The clinician
should consider plotting a non-K-corrected profile be¬
cause of the lack of contribution of the K score to those
scales that are /(-corrected. The prognosis for a psy¬
chological intervention is guarded.
35-40 36-45 2. Low. In this range are clients with limited personal re¬
sources who are experiencing severe distress that is
being openly acknowledged. Such clients have poor self-
concepts and are strongly self-dissatisfied but lack
either the interpersonal skills or techniques necessary to
alter the situation. These scores also may indicate per¬
sons who tend to be excessively open and revealing and
who may be masochistic confessors. In lower-class cli¬
ents this elevation reflects a moderate disturbance,
whereas in higher-class clients it reflects low ego
strength and more serious distress. The prognosis for a
psychological intervention is guarded.
41-55 46-55 3. Normal. Scores in this range indicate clients who have
a proper balance between self-disclosure and self-pro¬
tection. Such persons have sufficient personal resources
to desire and tolerate a psychological intervention. In
higher-class clients a moderate level of personal distress
would be expected. The prognosis for a psychological
intervention is good.
56-64 56-69 4. Moderate. Scores in this range indicate clients who are
being defensive and unwilling to acknowledge psycho¬
logical distress. This defensiveness may be character¬
ized by denial and hysteroid defenses, particularly in
lower-class clients. The clinician should consider plot¬
ting a non-K-corrected profile because of the excessive
contribution of the high K score to those scales that are
/(-corrected. The prognosis is guarded.
65 and 70 and 5. Marked. Scores in this range indicate clients who are
above above consistently trying to maintain a facade of adequacy
and control and are admitting no problems or weak¬
nesses despite their presence in a mental health set¬
ting. Such persons have a serious lack of insight into and
understanding of their own behavior. These clients are
being extremely defensive about some kind of inade¬
quacy, which may not be directly discernible from the
profile. The clinician should examine the reasons that the
client might be denying psychopathology and should
plot a non-K-corrected profile. The prognosis for any
form of psychological intervention is very poor because
of the client's lack of insight into his or her own be¬
havior.
118 Chapter 3

TABLE 3-52 Interpretations of K Scale Elevations for Normal Persons

MMPI-2 MMPI
T Score T Score Interpretation

30-40 27-45 1. Low. These clients are acknowledging limited re¬

sources for dealing with problems. The clinician should
investigate whether the clients actually have some form
of psychopathology that they are willing to report.
41-55 46-55 2. Normal. Scores in this range indicate clients from lower
socioeconomic classes or with limited education levels
who have an appropriate balance between self-disclo¬
sure and self-protection.
56-64 56-69 3. Moderate. Scores in this range indicate college-edu¬
cated and upper-class persons who are well adjusted,
insightful, self-reliant, and easily capable of dealing with
their everyday problems. When under stress, such per¬
sons may be unwilling to seek help with their problems.
65 and 70 and 4. Marked. These clients' professed adjustment and self-
above above adequacy are likely to reflect a defensive facade. These
persons probably have little interest in examining the ap¬
propriateness of this facade.

personal resources for working on her prob¬ justed clients (Gross, 1959). Such a client is
lems. hardly an optimal candidate for most forms
In an inpatient setting clients with this of psychological intervention, and little
validity scale configuration are likely to evi¬ change in this configuration occurs over
dence poorer impulse control and a greater time.
frequency of inappropriate and destructive In the second variant of this validity
behavior than clients with other types of va¬ scale configuration the L scale is below a T
lidity scale configurations (Post & Gas- score of 50, the F scale is equal to or greater
parikova-Krasnec, 1979). As the client begins than the K scale, and the K scale is above a T
to improve, the F scale elevation should de¬ score of 55 (see Figure 3-3, dotted line). The
crease and the K scale elevation should in¬ F scale in this configuration will not typically
crease. exceed a T score of 75.
There are two variants of this validity This client has long-standing problems
scale configuration. In the first variant the L to which she has become so well adjusted that
scale and the K scale are between a T score of she can feel good about herself while at the
50 and 60, and the F scale is elevated above a same time admitting problems. As the Fscale
T score of 65 (see Figure 3-3, dashed line). increases in elevation, the client still feels se¬
This client is admitting problems, which are cure about herself despite the number and se¬
of increasing severity as the F scale increases verity of problems. This client simply wants
in elevation and simultaneously is trying to help in dealing with her current problems and
defend himself against these problems. will be satisfied when these current stresses
This pattern of ineffective defenses with are alleviated. The only change in this config¬
the simultaneous admission of fairly severe uration with the alleviation of stress will
problems is typical of chronically malad¬ likely be a decrease in elevation of the Fscale.
 Validity Indexes and Validity Configurations 119

The second of the four commonly en¬

countered validity scale configurations is il¬
lustrated in Figure 3-4. The frequency with
which this configuration occurred in the four
samples is presented in Table 3-53.
In this configuration the L scale and the
K scale are elevated above a T score of at least
60, and may approach a T score of 65, and
the Fscale is near to or below a T score of 50.
This client is attempting to avoid or deny un¬
acceptable feelings, impulses, and problems.
That is, the client is trying to present himself
in the best possible light. This client tends to
be simplistic and views the world in terms of
the extremes of good and bad. The client will
have an adequate social adjustment or, at
worst, a mild behavioral disturbance (Gross,
1959).
In inpatient settings, clients with this va¬
lidity scale configuration are likely to be psy¬
chotic, particularly if the clinical scales also
suggest the presence of a psychotic condition
(Sines, Baucom, & Gruba, 1979). This type
of validity scale configuration occurs most
L F K frequently among defensive normals (e.g.,
unsophisticated job applicants), hysterics,
Scale and hypochondriacs.

TABLE 3-53 Frequency of Validity Scale Configurations by Sample

Sample

Validity Scale Clinic Medical Prison University

Configuration Clients Patients Inmates Students

Figure 3-3
(solid line) 46.8% 19.5% 25.5% 38.8%
Figure 3-3
(dashed line) 1 7.4 1 1.2 21.0 10.5
Figure 3-3
(dotted line) 3.8 0.1 1.5 1.4
Figure 3-4 10.0 35.3 21.5 7.2
Figure 3-5 14.0 19.9 24.5 27.8
Figure 3-6 7.2 4.1 1.5 0.5
None of these3 0.8 9.9 4.5 1 3.9

aMost of these configurations were characterized by the L, F, and K scales being approximately
equal to a T score of 50. See Greene (1980) for a more complete description of these samples
of clients.
120 Chapter 3

FIGURE 3-4 Validity Scale Configuration: FIGURE 3-5 Validity Scale Configuration:
Inverted Caret

©
O
(/)

*“ 50

L F K L F K
Scale Scale

This configuration is often accompanied sential features of this configuration are that
by elevations on Scales 1 (Hypochondriasis) the three validity scales have a positive slope
and 3 (Hysteria) and average scores on the in which the L scale is less than the F scale
rest of the profile. Deliberate defensiveness and the F scale is less than the K scale. Gen¬
and denial of psychopathology may be sus¬ erally, the L scale is about a T score of 40,
pected, if there is a legitimate reason for eval¬ the F scale is about a T score of 50 to 55,
uating the client. Typically these clients nei¬ and the K scale is in the T score range of 60
ther are referred for nor seek treatment. to 70.
College students instructed to endorse the This configuration is typical of a normal
MMPI items in terms of their “ideal self” individual who has the appropriate resources
produce this validity scale configuration for dealing with problems and who is not ex¬
(Hiner, Ogren, & Baxter, 1969). periencing any stress or conflict at the present
The third of the four validity scale con¬ time. The K scale elevation in this configura¬
figurations is illustrated in Figure 3-5, and tion will move up or down depending on the
the frequency of its occurrence in the four person’s reference group. For example, a
samples is presented in Table 3-53. The es- normal college student will score at the upper
 Validity Indexes and Validity Configurations 121

end of this range, whereas the normal lower- age range) except for Scale 5 (Masculinity-
class individual will score at the lower end of Femininity) among males.
this range. A job applicant or a prison inmate The last commonly encountered validity
who is trying to look “good” may have this scale configuration is illustrated in Figure
validity scale configuration. 3-6, and the frequency of occurrence of this
Lanyon and Lutz (1984) found that this configuration in the four samples is presented
validity scale configuration was characteristic in Table 3-53. In this configuration the three
of felony sex offenders who denied any sex¬ validity scales have a negative slope in which
ual deviant behavior. It is unusual for a self- the L scale is greater than the Fscale, which is
referred individual in a mental health setting greater than the K scale. The L scale is ele¬
to have this configuration; however, it can vated to a T score of 65, the Fscale is about a
occur among (1) “normal” persons involved T score of 50, and the K scale is equal to a T
in marital conflict or (2) upper-class or col¬ score of 40 to 45.
lege-educated persons who show sophisti¬ This client is naive and unsophisticated
cated defensiveness. In these cases the clinical but is trying to look “good.” Such clients
scales all will be submerged (within the aver- usually have little education and come from
the lower socioeconomic classes. Their at¬
tempt to look good usually is ineffective, and
FIGURE 3-6 Validity Scale Configuration:
the neurotic triad (Scales 1, 2, and 5) gener¬
Descending Slope
ally is elevated. For males, Scale 5 will be
low. Even when the Fscale is elevated, it usu¬
ally will not exceed a T score of 65; the client
is still trying to maintain a facade of looking
good despite the admission of some prob¬
lems.
These clients are poor candidates for any
form of psychological intervention. They are
unlikely to admit their problems, and when
they do, they lack both the resources for
and the interest in psychological interven¬
tions. This configuration changes little over
time, except that the F scale will likely de¬
crease if it is elevated higher than a T score of
50.

NORMAL PROFILES
Occasionally a profile will be encountered in
which the only significant validity or clinical
scale elevation is on the K scale. Marks and
Seeman (1963) identified such a profile as a
Normal K+ profile. This profile is basically
an elaboration of the second commonly en¬
countered validity scale configuration al¬
L F K ready described (see Figure 3-4). A Normal
Scale K+ profile (Marks & Seeman, 1963; Marks,
122 Chapter 3

Seeman, & Haller, 1974) has the following Two recent studies (Barley, Sabo, &
specific characteristics: Greene, 1986; Winters, Newmark, Lumry,
Leach, & Weintraub, 1985) reported similar
1. Psychiatric inpatients only frequency of occurrence of the Normal K+
2. All clinical scales below a T score of 70 profiles. However, Craig (1984) found no
Normal K+ profiles among his male heroin
3. Six or more clinical scales less than or
addicts. It appears that the Normal K+ pro¬
equal to a T score of 60
file occurs in 3 to 5 percent of clients in most
4. Scales L and K higher than the F scale clinical settings other than possibly substance
with the Fscale below a T score of 60 abuse/dependence.
5. The K scale 5 or more T score points There has been limited research on the
higher than the F scale actual correlates of the Normal K+ profile. It
should be remembered that the original cor¬
(Clinicians should note that no specific eleva¬ relates of this profile were derived on female
tion on the K scale is specified by these cri¬ psychiatric inpatients (Marks et al., 1974).
teria, only the relationship between the K Gynther and Brilliant (1968) reported no reli¬
scale and Scales L and F.) Marks and col¬ able psychological or behavioral differences be¬
leagues described these clients as shy, anx¬ tween the clients with a Normal K+ profile and
ious, inhibited, and defensive about other clients. They did find that 45 percent of
admitting that their problems could be psy¬ their patients had psychotic diagnoses.
chological. They avoided close interpersonal Barley and colleagues (1986) found that
relationships and were passively resistant. their patients with a Normal K+ profile were
Their personality structure had a schizoid less likely to have had chronic illnesses, to
component, their stream of thought was have attempted suicide prior to hospitaliza¬
often incoherent, and they frequently ap¬ tion, and to have somatic complaints. Their
peared perplexed. They also displayed para¬ behavior and emotional state was more likely
noid features: they were suspicious, fearful, to be overactive, and their current hospital¬
and sensitive to anything that might be con¬ ization was shorter in length. They found that
strued as a demand. 47 percent of these patients had a psychotic
Almost half of the patients with this pro¬ diagnosis, but none had a brain disorder di¬
file were diagnosed as psychotic, and almost agnosis.
a quarter of them were diagnosed as having a These studies suggest that clients with a
chronic brain syndrome. As a group, the cli¬ Normal K+ profile are likely to have a psy¬
ents were significantly above average in intel¬ chotic diagnosis about 50 percent of the time.
ligence, and more than 60 percent were edu¬ It is much less clear whether they will have a
cated beyond the high school level. neuropsychological disorder and what spe¬
A number of investigators have reported cific behavioral and psychological correlates
the frequency with which the Normal K+ pro¬ will be found. Duckworth and Barley (1988)
file is encountered. Gynther and Brilliant have a comprehensive review of this profile
(1968) found that 3.6 percent of the clients re¬ that should be consulted by the interested
ferred for testing at a mental health center reader.
met all the preceding criteria for a Normal K+
profile. Newmark, Gentry, Simpson, and
SIMULATION AS ROLE PLAYING
Jones (1978) found that 4.7 percent of an in¬
patient population who were diagnosed schiz¬ The ability to detect clients who are simulat¬
ophrenic produced a Normal K+ profile. ing psychopathology has been mentioned as a
 Validity Indexes and Validity Configurations 123

benefit of several of the validity indicators. profile, Wilcox and Dawson noted that both
Most investigations of the ability to simulate groups significantly elevated Scale 6 (Para¬
psychopathology have not provided the indi¬ noia), which suggests that their manipulation
vidual with an explicit description of the ac¬ was effective. Their results indicate that some
tual behavior to be simulated. The typical students can simulate deviant profiles with¬
instructions vaguely request that the individ¬ out being detected by the validity scales.
ual answer the items as a “neurotic” or “psy¬ Anthony (1976) asked inpatients, outpa¬
chotic” person would. Under these tients, and nonprofessional employees of a
conditions the traditional validity indicators state hospital to simulate various codetypes
generally are able to detect at least some cases on the MMPI when they were given a person¬
of simulation. When a specific description of ality description and case study of that profile
the role to be simulated is given to the indi¬ type. He rated the degree of similarity between
vidual, two consistent results have occurred: the simulated and criteria profiles. His results,
(1) the persons are better able to simulate a which are difficult to interpret, suggest that
role than when less specific instructions are ability to simulate a psychopathologic role re¬
provided and (2) the traditional validity indi¬ flects a complex relationship among the client’s
cators are unlikely to detect this form of sim¬ actual psychopathology, race, gender, and
ulation. the role to be simulated. He did not find a
Lanyon (1967) asked well-adjusted male general negative relationship between psy¬
college students (no clinical scales greater chopathology and the ability to simulate psy¬
than a T score of 69) and maladjusted stu¬ chopathologic roles as he had anticipated.
dents (at least three clinical scales greater Thus, the hypothesis that psychopathologic
than a T score of 69 that included at least two individuals are generally deficient in role-
scales other than Scales 5 or 9) to simulate playing skills could not be supported.
“very good adjustment” and “psychopathic Kroger and Turnbull (1975) asked male
personality.” The students were given a de¬ college students to take the MMPI either as
scription of a psychopathic personality. Both an Air Force officer or as a creative artist,
groups of students could simulate good ad¬ without providing a more explicit description
justment, but the well-adjusted students sim¬ of these two roles. They found that the stu¬
ulated psychopathic personality better than dents could simulate the profile of Air Force
the maladjusted students. When simulating officers, and the traditional validity indica¬
good adjustment, approximately one-third of tors, including the F-K index, were unable to
the students in both groups were able to es¬ detect this simulation.
cape detection by the traditional validity indi¬ The students produced a simulated pro¬
cators. file for the artists that was similar in configu¬
In a similar study Wilcox and Dawson ration but more elevated or deviant than the
(1977) found that college students who were actual artist profile. To investigate this latter
given a description of a paranoid individual result further, they asked two more groups of
while under hypnosis could simulate a para¬ students to take the MMPI as a creative artist
noid profile without being detected by the va¬ and provided an extensive role description of
lidity scales. Students who were given the de¬ artists. One group received an accurate role
scription without being hypnotized, however, description and the other received an inaccu¬
produced deviant profiles that were easily de¬ rate one. The students with the accurate role
tected by the validity scales. Rather than description produced a simulated profile that
using an external criterion to assess whether was nearly identical to the profile of actual
their students actually simulated a paranoid artists except for Scale 5; the traditional va-
124 Chapter 3

lidity indicators did not detect this simula¬ various validity indicators can help detect
tion. these inappropriate response sets when either
Thus, students apparently are capable of the profile configuration or other circum¬
simulating profiles of both deviant and non¬ stances suggest their possible influence.
deviant roles without being detected by the The two most blatant of these response
traditional validity indicators. Deviant roles sets are “all true” (see Profile 3-1) and “all
appear to be more difficult to simulate. false” (see Profile 3-2). Such sets are easy to
Gough’s (1954, 1957) findings that both stu¬ detect by examining either the answer sheet
dents and professionals were unable to simu¬ or the profile sheet. The validity scale config¬
late neurosis also suggest that deviant roles uration is highly suspect in both instances. In
are more difficult to simulate, particularly if the “all true” response set, the T scores of
a specific role description is not provided. the L and K scales are 35 and 30, respectively,
Additional research seems necessary to and the F scale is greater than a T score of
furnish a clearer understanding of role simu¬ 120. The psychotic tetrad (Scales 6, 7, 8, and
lation and its detection with the MMPI-2. In 9) is extremely elevated, and the neurotic
addition, research on the ability of patho¬ triad (Scales 7, 2, and 3) is around a T score
logic groups to simulate specific deviant and of 50.
nondeviant roles also is needed to clarify In the “all false” response set, all three
whether role simulation is likely to be a prob¬ validity scales are elevated between T scores
lem when clients take the MMPI-2. of 80 and 100. The elevation of the L and K
scales within this range never occurs in a valid
profile. The neurotic triad is extremely ele¬
DETECTION OF RESPONSE SETS
vated along with a moderate elevation on the
Another potential problem that the validity psychotic tetrad. Most persons using an “all
indicators attempt to address is detecting any true” or “all false” response set will include
inappropriate response sets that might be uti¬ a few answers in the other category, and this
lized by the client. Clients sometimes take the alters the profile somewhat. Even in these
MMPI-2 by using some response set other somewhat more sophisticated attempts at
than accurately endorsing the items. For ex¬ simulation, however, a “mostly true” or
ample, a client with limited but adequate “mostly false” response set still is readily ap¬
reading ability and intelligence may complete parent merely with a casual examination of
the MMPI-2 in a half hour or less, which the validity scales and indexes.
would arouse legitimate concern about the A response set that is slightly more diffi¬
authenticity of the item endorsements. Or a cult to detect is a random response set (Pro¬
client may be unable or unwilling to take the file 3-3). Frequently, this response set is iden¬
MMPI-2, but instead of directly refusing to tified as a random sort, a term that reflects
complete the task, he may endorse all items the early history of the MMPI where the cli¬
“true” or “false” or alternate “true” and ent sorted the items on the card form into
“false” responses. Or a client may try to “true” and “false” categories.
make herself look better or worse by over¬ In random sorts or random response
reporting or underreporting psychopathol¬ sets, the client endorses the items by ran¬
ogy; both of these response sets have been domly marking each item “true” or “false.”
discussed previously. This set may be suspected when a client com¬
Clinicians should try to prevent these sit¬ pletes the MMPI-2 much too quickly. It is im¬
uations by enlisting the client’s full coopera¬ portant in these cases to confirm that the cli¬
tion before starting the test. As mentioned, ent has the appropriate intellectual level and
 Validity Indexes and Validity Configurations 125

PROFILE 3-1
Name_

MMPI-2
s |\ I l.ltll.lW.IN . 11 H I I ( Ml Klllli A

ntnnt’Mtti Hu/U^ur.u'c Address _ __ _

ferjivutUty /mt'fiArtf -J"
Occupation_Date Tested

Profile for Basic Scales

Education__Age_Marital Status
Minnesota MiiltipJjTisic Pcrsi'iiaIit\ Inventors-2
C opv ni! hi - h\ INI RK.IMSOI Mil TMYIRSin Oh MINNESOTA
hMT hMT i renewed ll)7(h. IRNT III is Profile I orm I9NT Referred B>_
Ml rights reserved Distributed e\clusi\el> h\ NATION A l COMPl'TL.R SYSTEMS. IN(
under license from I he L niversiiv ol Minnesota.
MMPI-2 Code__
"MMPI-Y and "Minnesota Mult i phasic Persona I itv Inventors-2" are trademarks owned h\
Ihe I imei'si'IN ol Minnesota. Primed in the l lined Stales id' America.
Scorer's Initials
Hs- 5K D Hy Pd-4K Ml Pa PI-IK Sc* IK Ma-2K
TorTcL F K 1 2 3 4 5 6 7 8 9 0 T or Tc

factions ol k
• 4
■
i5 12
u 11 ■
14 11 9
13 Ml

,
13 ■
• ip in •
. 12 9 ■
!1 9
-

. 10 3 4
19 10 8 .
■- 9 4

8 6 3

"All True" Response Set

lb 3 6 3
14 h 3
13 7 5 3
12 6 9 2
11 6 • 2

10 5 4 2
• 5 4 2
- 4 3 2
■ 4 3 l
6 3 2 1

9 3 2 '
: 2 2 i
3 2 :

0
T 0 M "(1

Raw Score 0 41 1 11 20 13 24 23 25 39 59 35 34
NATIONAL
? Raw Score COMPUTER
^ K to be Added _1 Q 1 1 .0 SYSTEMS

Raw Score with k12 24 40 00 35 24001

 126 Chapter 3

PROFILE 3-2
Name_

MMPI-2
s l\ I I.HI>.iw.i\ .itkI It \K knilev

Address_
ferjc/ur/tty /mt'tii-ory
Occupation Date tested

Profile for Basic Scales Educalion Age_Marital Status

Minnesota Millliphasic Personal'it\ lmentor\-2
t op>rii!lii * h\ INI RKilVIs'oi INI I'NIYERSITY Oh MINNESOTA
I'M'. 194 a i renewed 1970). 19X9. Phis Profile form 19X9.
Referred By_
Ml ritillin reserved I)isiiihilled e\cltisi\cl\ h\ NATIONAL (OMPI TI R SYSTEMS. INC
under license from I lie l imciMU of Minnesota.
MMPI-2 Code
"MMPI-M iul 'Minnesota MnIiiphasic Personalitv ln\enior\-2" are trademarks owned h\
Ihc l ni\cfsiis ol Minnesoia Primed in the l nited States of America Scorer's Initials
Hs~ 5K Pd- -IK Pa Pi-IK Sc-IK Ma-2K Si
1 4 6 7 8 9 0 T or Tc

"All False" Response Set

Raw Score 151929 2137472631 15 9 19 11 35 NATIONAL

? Raw Score 0 15 29 29 6
COMPUTER
K to be Added 12 SYSTEMS

Raw Score with k36 38 38 48 17 24001

 Validity Indexes and Validity Configurations 127

PROFILE 3-3
Name

MMPI-2
s K I l.iih.i\\,i\ .11 ul I ( \l> KiiiUa

HtntH’-wfti Ha/tiff/ur.n'c Address

/m t’ftAvy -J
Occupation _ Date Tested

Profile for Basic Scales Education _ Age_Marital Status

Minnesota Mtiliipluisic Pcrsonalnv lmenior\-2
( op> fij2l.1t ' h\ III! RU.IMNOI INI IMMRSin ()l MINNESOTA
IlM_\ Iw4.' i renewed 1970). |9N9 This Profile Form 19X9 Referred By_
Ml riiihis reserved. Disirihmed e\eliisi\el\ h\ NATION A l (OMP1TI R SYSTEMS. INC
under license from I he l mversiiv of Minnesoui.
MMPI-2 Code.
MMPI-T and Minnesota Muliiphasic Personality Inventory-2" are irademarks owned by
llic l imersiiv ol Minnesoia. Primed in the l niied Slates of \merica.
Scorer's Initials
1
Hs- SK P0-4K
4
Pi- ik
7
Sc-IK
8
Ma- 2K
9
Si
0 T or Tc

15 12 6
1J 11 6
- 5
-11 13 ' 5

2; 13 -0 5
- >2 ’0
12 9 -
11 0 :
i; -

20 10 3
19 10 3 4
13 9
9
‘
'
4
3
"Random" Response Set
3 6 3
'

3 6
in Adults
'
3
13 ' - 3
12 6 5 ?
11 6 4 2

10 5 4 2
9 5 : 2
- - 3 2

6 3 2

5 3 2 ■
- 2 2 1
3 2 i l
2 ■ ' 0
1 l 0 0
0 0 0 0

Raw Score_7 30 15 16 28 30 25 28 20 24 39 23 34 NATIONAL

Raw Score 0 K to be Added _6 15 15 _3
COMPUTER
SYSTEMS

Raw Score with k24 31 39 54 26 24001

128 Chapter 3

reading ability to complete the MMPI-2. Fre¬ and occasionally with opposite responses as
quently, clients who are deficient in either deviant. Although these two profiles are the¬
area are reluctant to acknowledge their defi¬ oretical, they do illustrate the range in which
ciency and instead respond randomly. The each scale on the MMPI-2 can vary.
most apparent aspects of a random sort are
the F scale elevation that approaches a T
OBVIOUS AND SUBTLE ITEMS
score of 120 and the Scale 8 (Schizophrenia)
elevation that approaches a T score of 90. Numerous investigators have examined obvi¬
The L scale also is elevated higher than would ous and subtle items on the MMPI (see review
be anticipated with this clinical scale configu¬ by Dubinsky, Gamble, & Rogers, 1985), and
ration. there is a tendency for each investigator to de¬
Measures of the consistency of item en¬ fine obvious and subtle somewhat differently
dorsement can be used to assess whether the (cf. Hryckowian & Gynther, 1988; Ward,
client has responded randomly. In addition, a 1986). Since the Wiener and Harmon (Wie¬
clinical interview will readily identify clients ner, 1948) Obvious and Subtle subscales were
with genuine psychopathology who have pro¬ reviewed earlier, they will not be covered in
files that appear similar to a random response this section.
set. Such clients will be experiencing an acute Seeman and associates (Vesprani & See-
personality disorganization if they are not man, 1974; Wales & Seeman, 1968, 1972)
overtly psychotic. Frequently, they are so dis¬ noted that many of the subtle items are also
tressed that they are unable to complete the Zero items, items that are scored for abnor¬
MMPI-2. mality even though a majority of the norma¬
Archer, Gordon, and Kirchner (1987) tive group endorsed the item in the deviant di¬
examined the first three MMPI response sets rection. For example, 57.9 percent of the
in adolescents and concluded that the “all original normative group endorsed the fol¬
true” and “all false” responses were very lowing item as “true”: “At times my
similar to those produced by adults. How¬ thoughts have raced ahead faster than I could
ever, the random response set differed in speak them.” Yet a “true” response to this
both shape and elevation (see Profile 3-4). item is scored both on Scales 5 and 9. Thus,
Consequently, it will be even more important on these scales the deviant response for this
to check measures of the consistency of item item is “true” despite the fact that a majority
endorsement with adolescents since the pro¬ of the normative group endorsed this item as
file produced by a random reponse set is not “true,” since a higher proportion of the crite¬
as clearly deviant as with adults. rion groups for these scales answered it in the
Two other response sets are the “all devi¬ same direction.
ant” and “all nondeviant” response sets (see There are 84 Zero items that meet this
Profile 3-5). These response sets would require criterion (see Appendix B in Dahlstrom et al.,
the client to answer either all or none of the 1972). Seeman and colleagues further noted
items like the criterion group for each scale. that obvious items tended to be X items,
Both produce theoretical profiles since each which were endorsed by a minority of the
procedure requires that items be answered original normative group. Consequently, the
both “true” and “false” when the item is Zero and X items are roughly comparable the
scored “true” on one scale and “false” on to Wiener and Harmon (Wiener, 1948) Subtle
another. This double scoring of items occurs and Obvious items, respectively.
more often than might be expected since nu¬ Wales and Seeman (1968) found that
merous items are scored on more than one scale subtracting the total number of deviant re-
 Validity Indexes and Validity Configurations 129

PROFILE 3-4
NAME_
MINNESOTA MULTI PHASIC
PERSONALITY INVENTORY
S.R. Hathaway and J.C. McKinley ADDRESS_

PROFILE
OCCUPATION DATE TESTED

MlNNtSO! A Ml'L I [PHASIC PERSONALITY INVENTORY

Copv I HE UNIVFRSIl Y OF MINNESOTA Minnesota Mulliphasic Personality Inventory and MMPI' are
EDUCATION_ AGE.
RfiU-Wfi! 19/0 This Profile Fniirt 1948 1 976 1982 ATI lights reserved trademarks owned by The University ol Minnesota
DislnMnl I xi lusively hv NAII0NAI COMPU1ER SYSTEMS INC
Until" Licmsi tioni The University of Minnesota MARITAL STATUS__ REFERRED BY
Pi mtet! m the United Slates of'America . . _
1 23456789 o
Tor Tc 7 L F K Hs+5K 0 Hy Pd+4K Mf Pa Pt+IK Sc+1K Ma+ Si A R Es MAC*

"Random" Response Set

in Adolescents

(Adolescent Norms)

Raw Score Q TZ 30 IS 28 TO 25 28 20 2419 23 3_4_ NATIONAL

COMPUTER
K to be added _ _ _ _ _ SYSTEMS
Raw Score with K _ _ _ _ _ *49 Item version 27309
 130 Chapter 3

PROFILE 3-5
Name _

MMPI-2
s K 11.HI\a\\a\ .nul I t \K kmle\

.K'ttt Hn/Wp/nuk' Address_

ft’fMvut/Sfif fmenAvy - "
Occupation _Date Tested

Profile for Basic Scales

Education_ Age_Marital Status
Minnesota Multiplutsic Personality Inventory-2
( op\riuhi » hv INI RI CH MS Oh till l M\I RSH \ Oh MINM.SOTA
PM.: PM> i renewcd PMO). IPN9. This Profile honn PTX9.
Referred B\
Ml rights reserved. Distributed evcltisivclv hv NATIONAL (OMP( MR SYSTI MS. INC
tinder license from The l niversiiv ol Minnesota.
VI M PI-2 ( ode
"MMPI-2" and " Minnesota Mnlii phasic Personality lnventor\-2" are trademarks owned bv
live l niversiiv ol Minnesota, Primed in the I'nitcd States of America.
Scorer's Initials
Hs- 5K D Hy Pd-4K Ml Pa Pi-IK Sc-IK Ma+ 2K
TorTcL F K 1 2 3 4 5 6 7 8 9 0 T or Tc

- 5 •

b
15 i: 6
\4 ■'
' 14 ■ i •
13 in

13 10
"All Deviant" and "All Nondeviant"
i 12 10 5
12
11
9
9 . Response Sets
?l 11 3 4

10 3 4
is 10 8 .
13
17
9
9 '
4
3
"All Deviant"
16 8 1 3

15 8 6 3
14 7 6 3
13 7 5 3
1?
11
6
b
5
:
?
2
"All Nondeviant"
10 5 4 2
9 5 4 2
8 4 3 2
7 4 3
6 3 2 i

5 3 2 l
2 2 2 i
3 2 i 1
2 1 ' 0
1 l 0 0
0 0 0 0

Raw Score
NATIONAL
? Raw Score_ COMPUTER
K to be Added_ SYSTEMS

Raw Score with K_ 24001

 Validity Indexes and Validity Configurations 131

sponses to the X items from the total number 1 (Hypochondriasis) 3.13

of deviant responses to the Zero items pro¬ 4 (Psychopathic Deviate) 3.13
vides a useful index for identifying underre¬ 7 (Psychasthenia) 3.47
porting of psychopathology. Using 84 Zero 6 (Paranoia) 3.52
and 315 X items, they found that college stu¬ 8 (Schizophrenia) 3.64
dents instructed to underreport could suc¬ F 3.70
cessfully avoid endorsing the X (obvious)
items, but they increased their score on the The two surprising ratings are the some¬
Zero (subtle) items in the direction of in¬ what subtle mean rating for the L scale,
creased psychopathology. Wales and Seeman which is routinely described as an obvious
determined that a cutting score of - 4 for the scale that does not detect students instructed
Zero minus X difference score correctly identi¬ to underreport psychopathology (Dahlstrom
fied 100 percent of the underreported profiles et al., 1972), and the relatively neutral mean
and 70 percent of the real student profiles. rating of Scale l (Hypochondriasis), which
In extending the investigation of Zero has been described as a marker variable for
and X items to a psychiatric outpatient sam¬ obviousness (cf. Wiener, 1948). It is possible
ple, Vesprani and Seeman (1974) found a that the students insightfully interpreted the
similar pattern. When instructed to endorse L scale items as representing minor social foi¬
the items as their “ideal self,” the outpatients bles and the Hypochondriasis scale items as
decreased their scores on the X (obvious) representing physical/medical problems rather
items and increased their scores on the Zero than interpreting any of these items as repre¬
(subtle) items. senting psychological problems.
A third method of defining obvious and The mean ratings for the Zero and X
subtle items on the MMPI was devised by items were 2.04 and 3.16, respectively, and
Christian, Burkhart, and Gynther (1978). the mean ratings for Wiener’s (1948) obvious
They asked college students to rate on a 5- and subtle items were 2.44 and 3.45, respec¬
point scale from very obvious (5) to very sub¬ tively. Christian and associates (1978) noted
tle (1) how clearly each item indicated a psy¬ that the mean ratings for both the X items
chological problem. Mean ratings for each and Wiener’s obvious items fall in their neu¬
item were used to compute obviousness tral category of neither suble nor obvious.
scores for the standard MMPI validity and Researchers who are interested in using the
clinical scales, the Zero and X items,5 and Christian and colleagues’ ratings of the obvi¬
Wiener and Harmon’s Obvious and Subtle ousness of the MMPI items need to remem¬
items. A score of 3 indicated a neutral (nei¬ ber that their data are reported by Form R
ther subtle nor obvious) rating. The mean ob¬ item numbers. A table for converting Form R
viousness ratings for the validity and clinical item numbers to Group Booklet item num¬
scales, in order from most subtle to most ob¬ bers can be found in Dahlstrom and associ¬
vious, were: ates (1972).
Using Christian and associates’ (1978)
5 (Masculinity-Femininity) 2.21 five categories of subtlety (very subtle, sub¬
K 2.28 tle, neutral, obvious, very obvious), Burk¬
L (Lie) 2.41 hart, Christian, and Gynther (1978) found
0 (Social Introversion) 2.64 that when instructed to overreport psychopa¬
3 (Hysteria) 2.81 thology, students endorsed more neutral, ob¬
9 (Hypomania) 2.82 vious, and very obvious items and fewer very
2 (Depression) 2.94 subtle items. When instructed to underreport
132 Chapter 3

psychopathology, students endorsed fewer found that the Wiener and Harmon (Wiener,
neutral, obvious, and very obvious items and 1948) Subtle subscales attenuated validity to
more items in the two subtle categories. the same degree as the addition of a random
These results suggest that obvious and variable.
subtle items can help detect overreporting as Snyter and Graham (1984) using a dif¬
well as underreporting response sets, as was ferent definition of item subtlety also found
outlined above. The pattern of students en¬ that subtle items may have some utility on
dorsing more subtle items in the pathologic Scale 9. Similar to Gynther and associates
direction when instructed to underreport con¬ (1979), both Snyter and Graham (1984) and
firms previous findings by Vesprani and See- Worthington and Schlottmann (1986) found
man (1974). Although the cause of this para¬ that the subtle items on Scale 4 did not con¬
dox is unclear, Burkhart and colleagues tribute significantly to the prediction of other
(1978) conjectured that subtle items may be test scores.
endorsed more frequently when students are Grossman and Wasyliw (1988; Wasyliw,
instructed to underreport because such re¬ Grossman, Haywood, & Cavanaugh, 1988)
sponses are socially desirable. and Posey and Hess (1984, 1985) have inves¬
Using these same five categories of item tigated the use of the obvious and subtle sub¬
subtlety (Christian et al., 1978), Gynther scales in forensic settings to assess malinger¬
Burkhart, and Hovanitz (1979) examined the ing. Since Grossman and Wasyliw used the
relationship between subtle, neutral, and ob¬ Wiener and Harmon (Wiener, 1948) Obvious
vious subscales of Scale 4 (Psychopathic De¬ and Subtle subscales, and Posey and Hess
viate) and scores on a nonconformity ques¬ used the Christian and associates’ (1978) rat¬
tionnaire. On the nonconformity scale the ings of obvious and subtle items, their results
obvious and subtle subscales were directly re¬ are not easily integrated. Their research did
lated to the scale score for males; for females, generally support the use of obvious and sub¬
only the obvious and neutral subscales were tle subscales, to assess malingering and can
directly related to nonconformity. They con¬ provide direction for future work in this area.
cluded that the obvious subscale predicted Research on the obvious and subtle
nonconformity scores better than the subtle items again has raised the issue of the relative
or neutral subscales and that the obvious sub¬ merits of empirical versus content strategies
scale in conjunction with the subtle subscale of developing personality tests. Are obvious
was a better predictor than either subscale items or subtle items more directly related to
alone. In this situation the subtle subscale ap¬ the criterion being assessed? And do obvious
parently enhanced the predictive power of the items or subtle items provide the most effi¬
obvious subscale. cient means of measuring the criterion?
Gynther and colleagues have extended Jackson (1971), Hathaway (1972), Holden
this same methodology to Scales 2 (Burkhart, and Jackson (1979), and Gynther and Burk¬
Gynther, & Fromuth, 1980), 3 (Gynther & hart (1983) have presented overviews of this
Burkhart, 1983; Wilson, 1980), and 9 (Hov¬ issue.
anitz & Gynther, 1980). They found that sub¬
tle items supplied little additional informa¬
tion on Scales 2 and 3, and they probably
could be omitted on these scales. Subtle items ENDNOTES
seemed to provide information that was not 1. A condensed version of this section can
available from the obvious items on Scale 9. be found in Greene (1989).
Weed, Ben-Porath, and Butcher (1990) 2. Reproduced from the MMPI-2 by per-
 Validity Indexes and Validity Configurations 133

mission. Copyright © 1943, (renewed 1970), 1989 5. The Zero items used by Christian and
by the University of Minnesota. Published by the colleagues (1978), which are based on the Zero
University of Minnesota Press. All rights reserved. items found in Dahlstrom and colleagues (1972),
3. Ibid. are not the same as the Zero items used by Seeman
4. Ibid. and colleagues (cf. Wales & Seeman, 1968).
 CHAPTER 4

Clinical Scales

The clinician will need to become thoroughly Although a few comments have been
familiar with the clinical scales reviewed in made about the behavioral correlates of some
this chapter, for the information presented low points in the MMPI profile, little system¬
forms the basis for understanding combina¬ atic research has been done. The question of
tions of scales to be described in Chapter 6. whether the specific elevation of the low
For each scale the following areas will be sur¬ point is critical has scarcely been examined.
veyed: This neglect of investigation of low points on
the MMPI in part reflects the tradition that
1. The content areas tapped by the scale
low points represent adjustment, not psycho¬
2. How the scale was developed and the cri¬ pathology (Carson, 1969); there is some de¬
terion group bate, however, regarding the interpretation
3. General psychometric and clinical infor¬ of low points (Listiak & Stone, 1971).
mation While interpretation of high-point pairs
.
4 Interpretation and behavioral correlates or codetypes is the primary focus on the
of high scores in psychiatric and normal MMPI, low points on several scales deserve
populations careful attention regardless of the codetype
or overall elevation of the profile. Specific¬
5. Interpretations of low scores, usually in
ally, those MMPI scales for which low points
a normal population
need particular attention are Scales 3 (Hyste¬
6. The effects on scale scores in adults of ria), 4 (Psychopathic Deviate), 5 (Masculinity-
moderator variables such as gender, age, Femininity), 6 (Paranoia), and 9 (Hypomania).
education, and social class (ethnic group This chapter will discuss, therefore, low
membership will be discussed in Chapter points on each of these scales, although it is
8) not clear that these scales will be low points
7. A summary table of interpretations for on the MMPI-2 because of the use of uniform
the levels of elevation of the scale T scores. All MMPI-2 scales have been trun-

135
136 Chapter 4

cated at a T score of 30, which also limits how abnormal, psychoneurotic concern over bod¬
low scores can go. Clinicians should remem¬ ily functioning (McKinley & Hathaway,
ber that uniform T scores are used for all of 1940). (The development of Scale 1 was de¬
the clinical scales except for Scales 5 and 0 scribed in detail in Chapter 1; the reader
(Social Introversion), which still use linear T should review that material if necessary. The
scores. reader also should note that the current
Clinicians generally emphasize interpre¬ DSM-III-R [American Psychiatric Associa¬
tation of the codetype (the one or two highest tion, 1987] definition of hypochondriasis cm-
clinical scales elevated at or above a T score phasizes the fear or belief of the existence of
of 65) of the MMPI-2. Most of them, how¬ a serious disease rather than abnormal con¬
ever, also rely on individual scales to modify cern over bodily functioning.)
and supplement their interpretations. Con¬ Examples of Scale 1 items with the devi¬
trary to widespread assumptions, little actu¬ ant answer indicated in parentheses are:
arial research has been done on the behav¬
ioral correlates of individual MMPI scales. “I hardly ever feel pain in the back of
Some investigators (Boerger, Graham, & my neck.” (false)
Lilly, 1974; Hedlund, 1977; Hovey & Lewis,
“I have a great deal of stomach trou¬
1967; Zelin, 1971) have begun to examine the
ble.” (true)
correlates of individual MMPI scales, and
their results will be reported throughout this “The top of my head sometimes feels
chapter. Obviously, there has been little re¬ tender.” (true)1
search on the behavioral correlates of indi¬
vidual MMPI-2 scales. Most of the items on Scale 1 also are
The reader should note that comments in scored on the other clinical scales; only 8
the text to specific T scores reflecting high items are unique to the scale. A majority of
scores, moderate scores, and low scores on a the items (20) overlap with Scale 3 (Hysteria)
specific scale refer to the MMPI-2. Rather and are scored in the same direction. Only 5
than indicate parenthetically each time the of the items overlap with scales from the psy¬
comparable score on the MMPI, these scores chotic tetrad (the Paranoia, Psychasthenia,
will only be noted in the tables describing the Schizophrenia, and Hypomania scales), 4
interpretation of each scale below. items with Scale 8 (Schizophrenia), and 1
item with Scale 6 (Paranoia).
The deviant response for two-thirds of
SCALE 7; HYPOCHONDRIASIS (Hs)
the items on Scale 1 is “false”; hence, a ten¬
A wide variety of vague and nonspecific com¬ dency toward a “false” response set will ele¬
plaints about bodily functioning are tapped vate scores on this scale. The scale items ap¬
by the 32 items (33 items on the MMPI) of pear to be obvious in content (Dahlstrom,
Scale 7. These complaints tend to focus on Welsh, & Dahlstrom, 1972), although Chris¬
the abdomen and back, and they persist de¬ tian, Burkhart, and Gynther (1978) found
spite all reassurances and negative medical that students rated Scale 1 items as neutral
tests to the contrary. Scale / is designed to as¬ (neither obvious nor subtle) when asked how
sess a neurotic concern over bodily function¬ clearly these items were indicative of a psy¬
ing (i.e., psychotic concerns about bodily chological problem.
functioning are not found on this scale). Factor analyses of the items in several
The criterion group used in developing different populations have consistently iden¬
the scale was a group of hypochondriacs with tified one common factor, which has been la-
 Clinical Scales 137

beled poor physical health (Comrey, 1957a; cerns, the clinician should not ignore the ele¬
Eichman, 1962; O’Connor & Stefic, 1959; vation on Scale 1, The hypochondriacal fea¬
Stein, 1968) and a second factor, labeled gas¬ tures in these individuals usually are evident
trointestinal difficulties (Comrey, 1957a; despite their protests to the contrary.
O’Connor & Stefic, 1959). Scale 7 is a crude index of psychological
The 33-item Tryon, Stein, and Chu mindedness or sophistication, with high scor¬
(Stein, 1968) Bodily Symptoms scale (see ers lacking these attributes. It also is nega¬
Chapter 5), which was developed by a cluster tively correlated with intelligence (Brower,
analysis of all 550 items on the MMPI, con¬ 1947). This would substantiate the lack of
tains 23 items from Scale 7. Wiggins’ (1966) psychological mindedness in high scorers.
content scales of Organic Symptoms and Such persons are uninterested in exploring
Poor Health from the MMPI, which were de¬ any psychological reasons for their bodily
veloped on a rational or intuitive basis (see complaints. In fact, pity the clinician who di¬
Chapter 5), also overlap substantially with rectly suggests such a relationship; the client
Scale 7. The 36-item Organic Symptoms scale concludes that the clinician is poorly trained
has 13 items in common with Scale 7 and the since he or she does not recognize the symp¬
28-item Poor Health scale also contains 13 toms as genuine. With disparaging comments
Scale 7 items. Butcher, Graham, Williams, about their clinician’s lack of training and
and Ben-Porath’s (1989) content scale of skills to understand them, these clients trudge
Health Concerns from the MMPI-2 (see off to seek a more favorable second, third,
Chapter 5) has 23 items of its 36 items in com¬ and fourth opinion.
mon with Scale 7. The robustness of these hypochondria¬
Thus, it seems whether an empirical, ra¬ cal features often amazes the neophyte clini¬
tional, or statistical procedure is used, a gen¬ cian, who seemingly can readily recognize the
eral dimension of poor physical health and client’s motives. Despite, or perhaps because
vague somatic complaints can be identified in of, the transparency of the motives, any form
the MMPI-2 item pool in a variety of popula¬ of psychological intervention is almost surely
tions, and Scale 7 adequately assesses this di¬ doomed to fail. Thus, Scale 7 can be under¬
mension. stood as a characterologic scale (i.e., it re¬
A person who is actually physically ill flects a long-term personality style that is sta¬
will obtain only a moderate elevation (T score ble over time and resistant to change).
of 58 to 64) on Scale 7. Such persons will en¬ High scorers (T scores of 65 or higher)
dorse their legitimate physical complaints, on Scale 7 in any population are character¬
but they will not endorse the entire gamut of ized by their abnormal concern over bodily
vague physical complaints tapped by the functions and vague hypochondriacal com¬
scale. Scale 2 (Depression) is more likely to be plaints, which attests to the construct validity
elevated by actual physical illness than Scale of the scale. In addition, high scorers are de¬
7. If a client with actual physical illness ob¬ scribed as pessimistic, sour on life, and evi¬
tains a T score of 65 or higher on Scale /, dencing long-standing personal inadequacy
there are likely to be hypochondriacal fea¬ and ineffectualness. They seem to relish exag¬
tures in addition to the physical condition, gerating the ills of the world and of their own
and the client is probably trying to manipu¬ situation. They rarely express hostility
late or control significant others in the environ¬ overtly; instead, they express their resent¬
ment with the hypochondriacal complaints. Al¬ ment covertly by using physical complaints to
though the client may vehemently argue that control and manipulate others. Finally, they
the complaints reflect legitimate physical con¬ are unlikely to be diagnosed psychotic, al-
138 Chapter 4

though an occasional psychotic individual four levels of elevation of Scale 7 is provided

with somatic delusions may simultaneously in Table 4-1.
elevate Scales 7 and 8.
Low scorers (T scores less than 45) on
SCALE 2; DEPRESSION (D)
Scale 7 are a heterogeneous group since their
common characteristic is the nonendorse¬ The 57 items (60 items on the MMPI) of Scale
ment of hypochondriacal complaints. They 2 measure symptomatic depression, which
are described as alert, spontaneous, and not is a general attitude characterized by poor
unduly concerned about the adverse reactions morale, lack of hope in the future, and
of others. Good and Brantner (1961) com¬ general dissatisfaction with one’s own sta¬
mented that low scorers may be denying hy¬ tus (Hathaway & McKinley, 1942). The
pochondriacal complaints, but no research major content areas deal with a lack of interest
has documented this hypothesis. Persons in activities expressed as general apathy, physi¬
who have worked or lived with hypochondri¬ cal symptoms including sleep disturbances and
acs also tend to earn low scores on Scale 7. gastrointestinal complaints, and excessive sen¬
Whether the difficulty in working with or re¬ sitivity and lack of sociability (Dahlstrom et
lating to hypochondriacs results in a rejection al., 1972).
of this personality style or some other factor Scale 2 was derived empirically, using an
is involved has not been investigated. Mental approach identical to that used for Scale 7.
health professionals also score in this range. The criterion group used in developing Scale
Test-retest reliability coefficients for 2 consisted of 50 patients who represented
Scale 7 are among the largest for any of the relatively uncomplicated cases of the de¬
MMPI clinical scales, with correlations rang¬ pressed phase of manic-depressive psychosis.
ing from .79 to .86 for up to a two-week inter¬ Their responses were contrasted with a nor¬
val and .38 to .65 for a one-year interval mal group to produce a preliminary depres¬
(Dahlstrom, Welsh, & Dahlstrom, 1975). sion scale.
Test-retest reliability coefficients for Scale 7 As with Scale 7, it was found that some
on the MMPI-2 for approximately a one- nondepressed clients also scored high on this
week interval are .85 for men and .85 for preliminary depression scale. Consequently,
women (Butcher, Dahlstrom, Graham, the responses of 50 such nondepressed pa¬
Tellegen, & Kaemmer, 1989). tients who scored high on this preliminary
Women tend to endorse a few more scale of depression were compared with those
Scale 7 items than men (Butcher et al., 1989; of the criterion group, resulting in the identi¬
Colligan, Osborne, Swenson, & Offord, fication of 11 correction items that distin¬
1989; Dahlstrom et al., 1972). Scores on guished these two groups of patients. These
Scale 7 also tend to increase slightly with age correction items were scored so that patients
in normal individuals; persons older than age who were actually clinically depressed
65 endorse on the average about two to three achieved higher scores on the scale. The addi¬
more items than their younger counterparts tion of the 11 correction items resulted in the
(Colligan et al., 1989; Dahlstrom et al., 60 items that currently appear on Scale 2 of
1972). However, medical patients do not en¬ the MMPI.
dorse more Scale 1 items with increasing age Each of these 60 items was required to
(Swenson, Pearson, & Osborne, 1973). Per¬ meet the following criterion: The frequency
sons from lower socioeconomic classes also of its endorsement had to increase progres¬
are more likely to endorse Scale 1 items. sively from the normal group through a nor¬
A summary of the interpretations for mal group with depression (normals who
 Clinical Scales 139

TABLE 4-1 Interpretation of Levels of Elevation on Scale 1: Hypochondriasis (Hs)

MMPI-2 MMPI
T Score T Score Interpretation

44 and 40 and 1. Low. These clients may be denying the presence of vague physical
below below complaints. Scores in this range are typical for persons in helping
professions and for children of a hypochondriacal parent.
45-57 41-59 2. Normal. These clients have a typical number of physical com¬
plaints.
58-64 60-69 3. Moderate. Scores in the lower end of this range are typical for
physically handicapped persons and persons with actual physical
illness. These clients have some concern about their bodily function¬
ing and are likely to be seen as immature, stubborn, and lacking
drive. Review of content and/or supplementary scales may facilitate
interpretation in this range.
65 and 70 and 4. Marked. These clients are excessively concerned about vague
above above physical complaints and use them to manipulate and control oth¬
ers. They are cynical, whiny, demanding of attention, and generally
negative and pessimistic. The prognosis for either psychological or
physical intervention is guarded. These clients focus on vague
bodily complaints and resist any form of resolution. Conservative
interventions reassuring these clients about their bodily complaints
are indicated. Review of the content and/or supplementary scales
may facilitate interpretation at the lower end of this range.

achieved high scores on the preliminary de¬ about themselves and the environment, with
pression scale) to the criterion group. The de¬ higher scores indicating dissatisfaction. As
pressed normal group was used to help estab¬ clients’ evaluations of themselves or of the
lish the meaning of intermediate scale values situation changes, scores on Scale 2 should
between the normal and criterion groups, change concomitantly.
which would have been impossible if only the It may seem unusual that a psychotic cri¬
two extreme groups were contrasted. terion group was used to develop a scale that
Examples of Scale 2 items with the devi¬ is part of the neurotic triad (Scales 7, 2, and
ant response indicated in parentheses are: 3) and is thought to measure reactive depres¬
sion (Carson, 1969; Dahlstrom et al., 1972).
“I am easily awakened by noise.” (true) This paradox may be partially explained by
“I usually feel that life is worthwhile.” Hathaway and McKinley’s interest in devel¬
(false) oping this scale to measure symptomatic de¬
“I believe I am no more nervous than pression, which can be a reaction to a variety
most others.” (false)2 of causes (such as economic crises, vocational
difficulties, or personal problems) and which
Scale 2 is thought to measure reactive or occurs in a multitude of psychopathologic
exogenous depression rather than “neurotic” conditions.
or endogenous depression. Accordingly, With Hathaway and McKinley’s (1942)
scores are expected to fluctuate as the client’s interest in assessing reactive depression, the
mood changes. Thus, Scale 2 is an index of specific diagnosis of the criterion group was
how comfortable and secure clients feel not important as long as depression was a
140 Chapter 4

central feature. Since reactive depression is ubiquitous nature of depression, the develop¬
not a stable trait and will vary markedly over ers of the various depression scales may have
time, using patients in the depressed phase of used divergent samples in constructing their
manic-depressive psychosis as a criterion measures. Research that investigates the rela¬
group insured that the depressive features tionships among these different depression
were pronounced and central. scales and their relationship to depressive be¬
Most of the items on Scale 2 are scored haviors is urgently needed.
on the other clinical scales as well. Only 10 Harris and Lingoes (1955) formed sub¬
items are unique to Scale 2 on the MM PI-2, scales on Scale 2 of the MMPI by subjectively
and 3 of these are correction items; 13 items grouping together the items that were either
are unique to Scale 2 on the MMPI, and 6 of similar in content or seemed to reflect a single
these are correction items. The overlapping attitude or trait. Following this procedure,
items are relatively evenly distributed among they identified five groups of items within
the other clinical scales. The deviant response Scale 2: Subjective Depression, Psychomotor
for two-thirds of the items on Scale 2 is Retardation, Physical Malfunctioning, Men¬
“false” so that a tendency to endorse the tal Dullness, and Brooding (see Table 4-2).
MMPI-2 items as “false” inflates the scores The Harris and Lingoes subscales have not
on Scale 2. (A score on TRINthat is less than been changed on the MMPI-2.
9 would be expected in such cases; see Chap¬ The purpose of the Harris and Lingoes
ter 3.) (1955) subscales is to facilitate interpretation
Wiener and Harmon (Wiener, 1948) of identical scores on Scale 2. A client can ob¬
judged two-thirds of the items to be obvious tain a specific raw score on the scale by en¬
in content. Clients who are severely depressed dorsing items from any one or various combi¬
are more likely to endorse obvious items, nations of the subscales. Knowing the
whereas mildly depressed persons tend to en¬ subscale scores should help the clinician un¬
dorse subtle items (Dahlstrom et al., 1972). derstand the exact nature of each client’s de¬
Thus, it seems that persons are unlikely to en¬ pression, especially when two or more clients
dorse items with obvious depressive content have identical raw scores.
until they are significantly depressed (Nelson, Harris and Lingoes (1955) did not re¬
1987). strict items to only one subscale; conse¬
Comrey’s (1957b) factor analysis of quently, there is extensive item overlap be¬
MMPI items on Scale 2 revealed that 28 of tween some of the subscales. For example, all
the items loaded on a factor that he labeled 10 items on the Brooding subscale also ap¬
neuroticism; 12 other items loaded on a fac¬ pear on the Subjective Depression subscale,
tor called poor physical health. Only 8 items and 12 of the 15 items on the Mental Dullness
from Scale 2 are found on Wiggins’ (1966) subscale are on the Subjective Depression
content scale of Depression, and 10 items are subscale. Other subscales, such as the Physi¬
contained within the Tryon, Stein, and Chu cal Malfunctioning subscale, have few or no
(Stein, 1968) cluster scale of Depression. items in common with the other subscales.
Butcher and associates’ (1989) content scale Miller and Streiner (1985) found that
of Depression from the MMPI-2 (see Chapter judges who were asked to reproduce the
5) has only 9 of its 33 items in common with groups of items from the Harris and Lingoes
Scale 2. subscales agreed reliably only on 9 of the 28
Apparently, all of these depression subscales (Physical Malfunctioning [Z9J; De¬
scales vary dramatically depending on how nial of Social Anxiety [Hy,]; Need for Affec¬
they were constructed. Moreover, due to the tion [Hy2]; Somatic Complaints [Hy4 ]; Family
 Clinical Scales 141

TABLE 4-2 Description of High Scorers on the Harris and Lingoes Subscales
for Scale 2 (Depression)

Subscale

Abbre¬ Number
Name viation of Items Description of High Scorers

Subjective Depression D1 32 These clients lack joy in doing things,

are pessimistic, and have poor morale
and low self-esteem; they complain
about psychological inertia and lack of
energy for coping with problems.
Psychomotor Retardation d2 14 These clients are nonparticipative in so¬
cial relations and are immobile.
Physical Malfunctioning d3 1 1 These clients complain about physical
malfunctioning and are preoccupied
with themselves.
Mental Dullness d4 1 5 These clients are unresponsive and dis¬
trustful of their own psychological
functioning.
Brooding d5 10 These clients are ruminative and irri¬
table.

Discord [Pd,\; Naivete [Pa3 ]; Lack of Ego and they suggested that the entire Scale
Mastery, Cognitive [Sc3]; Bizarre Sensory Ex¬ should be used rather than any set of sub¬
periences [Sc6]; and Amorality [Ma,]). For 10 scales.
of these subscales, the judges did not agree on Unless some form of computer scoring
a single item. It appears that this subset of 9 of the MMPI-2 is used, it is extremely time-
Harris and Lingoes subscales warrant clinical consuming to score all MMPI-2 supplemen¬
use and further research. tary scales and subscales, such as the Harris
Little research has been conducted on and Lingoes subscales. Given all the caveats
the Harris and Lingoes subscales. The re¬ noted above, the Harris and Lingoes sub¬
search that does exist consists primarily of re¬ scales may provide helpful interpretive infor¬
porting means and standard deviations for mation when Scale 2 is between a T score of
these subscales for various samples of indi¬ 60 and 80. When Scale 2 exceeds a T score of
viduals (cf. Gordon & Swart, 1973; Panton, 80, all subscales usually are elevated above a
1959b). T score of 70, and when there is a T score
Lingoes (1960) identified seven factors below 60, none of the subscales will likely be
in his factor analysis of all 28 of the Harris elevated above a T score of 70.
and Lingoes subscales, which suggests that These subscales should not be interpre¬
these subscales are capable of providing in¬ ted unless they exceed a T score of 70 because
formation beyond that contained within the of their restricted variance; on some sub¬
standard clinical scales. However, Bernstein scales, endorsing one additional or one fewer
and Garbin (1985) concluded that none of the item will change the client’s score by 5 to 10 T
subscales (Comrey or Harris and Lingoes) score points. These general guidelines for in¬
could explain the item structure of Scale 2 terpreting the Harris and Lingoes subscales
142 Chapter 4

for Scale 2 also apply to the Harris and Lin¬ MMPI-2 does not adequately assess suicide
goes subscales for Scales 3 (Hysteria), 4 (Psy¬ risk, and the clinician is well advised to use
chopathic Deviate), 6 (Paranoia), 8 (Schizo¬ more appropriate assessment techniques,
phrenia), and 9 (Hypomania). Harris and such as a directive clinical interview if suicide
Lingoes (1955) did not develop subscales for risk is an issue. Even using the MMPI-2 to
the other clinical scales. identify groups of individuals who may be
A description of high scorers on the Har¬ suicide risks is questionable because of the in¬
ris and Lingoes subscales for Scale 2 appears ordinate number of false positives and false
in Table 4-2. negatives that may be generated by any
Interpretation of Scale 2 varies markedly MMPI-2 index used to predict suicide risk.
depending on which other clinical scales are The use of the MMPI-2 in assessing suicide
elevated in conjunction with it; consequently, risk will be examined more fully in Chapter 5.
the scale is one of the most difficult clinical High scorers (T scores of 65 or higher) on
scales to interpret in isolation. An elevated Scale 2 have been described in a variety of
score on Scale 2 reveals that the client is upset ways; this variety reflects the fact that depres¬
and feeling depressed about something; the sive features are found as a concomitant to all
precise source of the distress, however, can¬ types of behavior and psychopathology. Gen¬
not be deduced from the scale alone. eral descriptions of high scorers indicate that
For example, a person in legal custody they are depressed, anxious, moody, and in¬
who is unhappy about being incarcerated and hibited. They display excessive sensitivity to
a client in psychotherapy who is concerned their own depressed level of functioning and
about self-worth may obtain similar raw usually are withdrawn and isolated. To the
scores on Scale 2; their depressions, however, extent that these negative attributes represent
clearly emanate from different sources. dissatisfaction with oneself, they will serve as
These sources will be evident from the eleva¬ an internal pressure to change and hence are a
tions on the other clinical scales: Scales 4 good prognostic sign.
(Psychopathic Deviate) and 9 (Hypomania) High scorers also frequently have so¬
are likely to be the two highest scales for the matic symptoms and complaints, sleep diffi¬
person in custody, whereas Scales 2 and 7 culties, and a loss of appetite. Moreover, they
(Psychasthenia) will probably be the highest are rather consistently described as not being
scales for the client in psychotherapy. hyperactive, excited, or belligerent toward
When Scale 2 is the only clinical scale el¬ others.
evated above a T score of 65, a careful evalu¬ Persons with elevated scores on Scale 2
ation of suicidal risk is indicated, particularly are acknowledging their personal discomfort
if there are no overt behavioral signs of de¬ and dissatisfaction with their current level of
pression (Carson, 1969; Graham, 1987). Sui¬ functioning. Their subjective distress may
cide risk in such clients is generally consid¬ represent anxiety and its concomitants, or it
ered to be greater than when depression is may represent a genuine depressive condi¬
more demonstrable clinically. The MMPI-2 tion. Thus, persons with elevations on Scale 2
contains four items (150, 506, 520, and 524) will not always be diagnosed as being de¬
that directly inquire about suicidal ideation pressed, since their diagnosis will reflect their
and attempts. These four items should be re¬ prominent symptoms and behaviors. Regard¬
viewed routinely in all clients regardless of less of their diagnosis, these clients are ac¬
the consistency or accuracy of item endorse¬ knowledging their dissatisfaction with their
ment. present circumstances.
Other than these four specific items, the Clinicians have been cautioned about
 Clinical Scales 143

the interpretation of high Scale 2 scores in the dercontrolled; this can be manifested by os¬
aged because of the increase in scores that oc¬ tentatiousness, sarcasm, or exhibitionism.
curs in normal individuals (cf. Colligan, Os¬ For some, their activity, aggressiveness, and
borne, Swenson, & Offord, 1983, 1989; tendency to show off interferes with their in¬
Dahlstrom et ah, 1975). It has been conjec¬ terpersonal relationships.
tured that the elderly may be reporting more It is unclear whether other scales in the
physical symptoms, which elevates Scale 2. profile will help to differentiate among these
However, Dye, Bohm, Anderten, and subtypes of persons with low scores. Very low
Won Cho (1983) reported that older psychiat¬ scores (T scores of approximately 35) seem to
ric patients (age 60 + ) did not express con¬ represent an inability to tolerate anxiety and
cern over declining physical health as symp¬ a tendency to act out, which could play a role
toms of their depression, although concern in some of the negative behaviors described
over physical well-being did become more im¬ above. Although this inability to tolerate
portant as reflected by the factor structure. anxiety makes sense logically, it has not been
They also noted that there were subtle quali¬ tested empirically.
tative changes in the expression of depression Reliability coefficients for Scale 2 on the
across their three age groups. The perfor¬ MMPI are not as low as might be anticipated,
mance of the aged on the MMPI-2 is a topic given that the scale is a measure of reactive
that needs additional research (see Chapter depression that should vary over time. Gener¬
8). ally, reliability coefficients for this scale are
Persons with moderate elevations (T comparable to those for the other clinical
scores from 58 to 64) on Scale 2 are generally scales. Reliability coefficients of .80 to .90
described in similar but less extreme terms. for intervals up to a month and .40 to .50 for
Such persons are seen as shy, prone to worry intervals up to a year or more are common
and depression, and dissatisfied, either with (Dahlstrom et al., 1975). Test-retest reliabil¬
themselves or their personal situation. ity coefficients for Scale 2 on the MMPI-2 for
Bieliauskas and Shekelle (1983) found that approximately a one-week interval are .75 for
normal males with moderate elevations on men and .77 for women (Butcher et ah,
Scale 2 of the MMPI frequently felt nervous 1989).
or upset and they spent a lower percentage of Women tend to endorse more items on
time in bed sleeping. They also were more Scale 2 than men, and older, normal persons
likely to be rated as appearing emotionally tend to endorse more items than younger,
tense. normal persons (Colligan et ah, 1983, 1989;
Low scorers (T scores below 45) should Dahlstrom et ah, 1972; Leon, Gillum,
be evaluated carefully since some kind of de¬ Gillum, & Gouze, 1979). However, scores on
pressive mood should be present in most, if Scale 2 do not increase with age in medical
not all, clients in psychiatric settings. Scores patients (Swenson et ah, 1973) or psychiatric
in this range suggest that the clients are not patients (Hedlund & Won Cho, 1979).
affected adversely by the behaviors that led The interpretation of Scale 2 at four lev¬
to their referral for an evaluation. The clini¬ els of elevation is summarized in Table 4-3.
cian should note well that scores in the nor¬
mal range or below are not appropriate in
SCALE 3: HYSTERIA (Hy)
psychiatric clients. Low scorers are generally
described as active, alert, socially outgoing, The 60 items of Scale 3 consist of two general
and effective in a variety of tasks. Some per¬ types: items reflecting specific somatic com¬
sons with low scores also are described as un¬ plaints and items that show that the client
144 Chapter 4

TABLE 4-3 Interpretation of Levels of Elevation on Scale 2: Depression (D)

MMPI-2 MMPI
T Score T Score Interpretation

44 and 40 and 1. Low. These clients tend to be alert, gregarious, and active. Be sure
below below that these behaviors are appropriate for the person's situation and
setting (i.e., clients should rarely be scoring in this range).
45-57 41-59 2. Normal. These clients have a typical number of attitudes and be¬
haviors that reflect symptomatic depression.
58-64 60-69 3. Moderate. These clients are dissatisfied with something or with
themselves, but they may not recognize this state as depression.
Their mild degree of dissatisfaction may appropriately represent the
situation. Or they may not really be concerned about what is hap¬
pening to them, or they may have learned to adjust to a chronic
depressed existence. Review of the content and/or supplementary
scales may facilitate interpretation in this range.
65 and 70 and 4. Marked. These clients exhibit a general sadness and depressed
above above mood either about life or themselves. The clinician can determine
the source of this depressed mood either by asking the client or by
examining the clinical scales. As the scores increase, the pessi¬
mism, depression, and hopelessness begin to pervade the client's
entire life. These clients tend to be depressed, withdrawn, guilty,
and self-deprecating. Review of the content and/or supplementary
scales may assist the clinician in interpretation at the lower end of
this range.

considers himself or herself well socialized reduced the validity of Scale 3. Therefore,
and adjusted. Although these two types of these items were returned to Scale 3 and de¬
items are either unrelated or negatively corre¬ leted from Scale 1 (see discussion of the con¬
lated in normal individuals, they are closely struction of Scale 1 in Chapter 1). The result
associated in persons whose personality re¬ was the current 60 items on Scale 3.
volves around histrionic dynamics. Such per¬ Examples of the items with the deviant
sons generally maintain a facade of superior answer indicated in parentheses are:
adjustment and only when they are under
stress does their proneness to develop conver¬ “Much of the time my head seems to
sion-type symptoms as a means of resolving hurt all over.” (true)
conflict and avoiding responsibility appear. “I often wonder what hidden reason an¬
Scale 3 was developed on an empirical other person may have for doing some¬
basis, using a criterion group composed of 50 thing nice for me.” (false)
patients with either a diagnosis of hysteria or “It is safer to trust nobody.” (false)3
identifiable histrionic personality compo¬
nents. The original Hysteria scale included Preliminary use of Scale 3 revealed that
numerous somatic complaint items that also clients tended to score similarly on Scales 1
appeared on Scale 1 (Hypochondriasis). In an and 3. Clinical experience, however, demon¬
effort to differentiate Scales 1 and 5, Mc¬ strated valid clinical differences in prognosis
Kinley and Hathaway (1944) eliminated the and treatment for clients who scored rela¬
duplicate items from Scale 3; this, however, tively higher on Scale 1 or Scale 3, so Me-
 Clinical Scales 145

Kinley and Hathaway (1944) decided to re¬ vidual items on Scale 3 have yielded generally
tain both scales. convergent results. Using factor analysis,
Clients who scored higher on Scale 1 Comrey (1957c) identified five factors: poor
than Scale 3 tended to have diffuse, vague physical health, shyness, cynicism, head¬
physical complaints, and the role of psycho¬ aches, and neuroticism. Through cluster
logical factors in their disability was readily analysis, Little and Fisher (1958) identified
apparent. Clients who scored higher on Scale two relatively independent clusters of items:
3 than Scale 1 were less obviously neurotic; in admission of physiologic symptoms and de¬
fact, they appeared normal psychologically nial.
except when under stress. Their physical These two clusters were used to develop
complaints tended to be specific and were the Admission (Ad) and Denial (Dn) scales.
likely to be psychosomatic in nature. Further The Ad scale correlates positively (.89-.90)
differences between high scorers on these two with Scale 1, which is logical since the two
scales will be detailed in Chapter 6. scales have 18 items in common. The Dn scale
Almost all items on Scale 3 are scored on correlates positively (.78-.88) with the K
other clinical scales; only 10 items are unique scale, with which it has 9 items in common.
to Scale 3. One-third of the Scale 3 items Clients who score high on the Ad scale com¬
overlap with Scale 1 (even after the duplicate plain about their somatic functioning and
somatic complaints items were deleted from have poor interpersonal relationships. High
Scale 1) and are scored in the same direction. scorers on the Dn scale are described as lack¬
Thus, it stands to reason that these two scales ing insight into their own behavior and mor¬
often are simultaneously elevated due to their ally virtuous. Little and Fisher (1958) believe
shared variance. that when both of these scales are elevated,
The other overlapping Scale 3 items are the person should have conversion reaction
distributed relatively evenly across the other dynamics.
clinical scales. There are 10 items on Scale 3 Harris and Lingoes (1955) identified five
that overlap with the K scale, and they are subscales within Scale 3: Denial of Social
scored in the same direction. Although Scale Anxiety, Need for Affection, Lassitude-Mal¬
3 is not A"-corrected, the fact that they share aise, Somatic Complaints, and Inhibition of
10 items functionally produces a result sim¬ Aggression (see Table 4-4). Two of these sub¬
ilar to a TCcorrection of .33. Consequently, scales overlap substantially with iheAd scale:
when a non-A-corrected profile is con¬ Lassitude-Malaise has 14 of its 15 items in
structed (see Chapter 2), the contribution of common with the Ad scale, and Somatic
the K scale to Scale 3 cannot be removed di¬ Complaints has 16 of its 17 items in common
rectly. with the Ad scale. The other three sub¬
As with the other scales in the neurotic scales—Denial of Social Anxiety, Need for
triad (Scales 1, 2, and 3), “false” is the devi¬ Affection, and Inhibition of Aggression—
ant response for 47 (78 percent) of the items overlap completely with the Dn scale, except
on Scale 3. The profile for an “all false” re¬ that Inhibition of Aggression has one item
sponse set in Chapter 3 illustrates the empha¬ that does not appear on Dn.
sis on “false” as the deviant response on the Because of this almost complete overlap
neurotic triad scales. Wiener and Harmon between the Ad and Dn scales and the Harris
(Wiener, 1948) judged the items on Scale 3 to and Lingoes subscales for Scale 5, it is unnec¬
be almost evenly split between obvious and essary to score both groups of scales. If the
subtle items. clinician chooses to score the Harris and Lin¬
Studies of the associations among indi¬ goes subscales, it is possible to obtain a close
146 Chapter 4

TABLE 4-4 Description of High Scorers on the Harris and Lingoes Subscales
for Scale 3 (Hysteria)

Subscale

Abbre¬ Number
Name viation of Items Description of High Scorers

Denial of Social Anxiety HVl 6 These clients are characterized by social

extroversion.
Need for Affection Hy2 12 These clients obtusely deny that they
have a critical or resentful attitude
toward others. They consider themselves
impunitive, and they overly protest their
optimism and faith in other people.
Lassitude-Malaise HY3 1 5 These clients complain about functioning
below par physically and mentally. They
effortfully keep up a good front, but they
need attention and reassurance.
Somatic Complaints HY4 17 These clients exhibit somatic complaints
of a kind that suggest repression and
conversion of affect.
Inhibition of Aggression hY5 7 These clients express concurrence with
others and disavow violence.

approximation of the Ad and Dn scale scores and their effects on behavior (cf. Weinstein,
by combining the raw scores on the appropri¬ Averill, Opton, & Lazarus, 1968). The Dn
ate subscales and converting these to T scores scale also shares 19 of its 26 items with Byrne,
using the means and standard deviations for Barry, and Nelson’s (1963) Repression-Sensi¬
Ad and Dn (see Appendix D). tization scale, which has been extensively in¬
Two of Wiggins’ (1966) content scales vestigated as a measure of personality. The
share items with the subscales for Scale 3: the clinician who is interested in pursuing this
Organic Symptoms scale has 11 items, and line of research should consult Dahlstrom
the Poor Health scale has 3 items in common and associates (1975) for a review.
with the Somatic Complaints subscale. Pro- High scorers on Scaled (T scores of 65 or
kop (1986) has shown how the Harris and higher) are described as self-centered, imma¬
Lingoes subscales for Scale 3 can be useful in ture, and infantile. They are demanding of
the treatment of low back pain patients, and attention and manipulative in interpersonal
Miller and Streiner (1985) found that the relationships. They tend to be uninhibited
items on three of these five subscales could be and outgoing in their social relationships, al¬
replicated by judges. The subscales for Scale though they relate with others on a superficial
3 seem to be assessing similar functions in a and immature level. As their T score on Scale
variety of populations, and they appear to 0 (Social Introversion) approaches 30, the su¬
warrant continued clinical use and experi¬ perficiality and lack of real intimacy in their
mental investigation. interpersonal relationships becomes even
The Dn scale (Little & Fisher, 1958) has more apparent.
been used in many studies of defensive styles Their insensitivity to others and lack of
 Clinical Scales 147

empathy reflect their egocentric involvement. from the other clinical scales since the behav¬
Their primary defenses are denial and repres¬ ioral and clinical correlates of Scaled are usu¬
sion, and they generally appear to be defen¬ ally stable regardless of scores on the other
sive and overcontrolled. They tend to be clinical scales. The clinician should be sure,
emotionally immature and labile. A pro¬ however, that the client has endorsed both
found fear of pain, both emotional and phys¬ sets of items within Scale 3—specific somatic
ical, may characterize high scorers. When complaints and denial of psychological prob¬
under stress, high scorers are likely to display lems—before interpreting it in isolation; that
specific physical complaints such as head¬ is, all subscales should be elevated above a T
aches, chest pains, or tachycardia. At these score of 70.
times they also display transient depressive However, McGrath and O’Malley (1986)
features and anxiety. High scorers on Scale 3 found that Scales K, 7, and 3 needed to be el¬
are rarely psychotic, although their symp¬ evated to insure that both denial of problems
tomatology may be quite dramatic during pe¬ and specific somatic complaints were present.
riods of stress. The statement is sometimes made (Duck¬
Because of their strong need to be liked worth & Anderson, 1986) that the client will
and their desire to make a good initial impres¬ see and acknowledge behaviors indicated by
sion on others, high scorers appear to be clinical scales whose elevations are higher
good candidates for psychological interven¬ than Scale 3 but deny and fail to see behav¬
tions since they probably will respond posi¬ iors indicated by scales whose elevations are
tively to the clinician. Their desire for atten¬ lower than Scale 3. Thus, if Scales 2 and 3 are
tion and support further suggests that they at T scores of 80 and 70, respectively, the cli¬
will enjoy interacting with the clinician. They ent is reputed to see and acknowledge the de¬
are, however, generally intolerant of analysis pressive features. If these T scores are re¬
of their personality dynamics and frequently versed, however, the client purportedly will
place inordinate demands on their clinician. deny the depressive features. Since no re¬
Their histrionic style usually is so deeply in¬ search has investigated this hypothesized re¬
grained that they are unaware of it. When the lationship, the clinician is cautioned against a
clinician points out the realities of their situa¬ noncritical use of such interpretations.
tion, they frequently cannot see their role in it Test-retest reliability coefficients for
and complain that the clinician does not un¬ Scale 3 on the MMPI range from .63 to .84
derstand them. Thus, despite the positive ini¬ for intervals up to two weeks and from .36 to
tial impression, any form of psychological in¬ .72 for intervals up to one year (Dahlstrom et
tervention will be a trying task for the al., 1975). Test-retest reliability coefficients
clinician. for Scaled on the MMPI-2 for approximately
Low scorers on Scale 3 (T scores less a one-week interval are .72 for men and .76
than 45) are described as socially isolated, for women (Butcher et al., 1989).
conforming, and relatively unadventurous. Females endorse more Scale 3 items than
They are likely to have limited interests. They males. Age appears to have little impact on
tend to feel that life is tough and are sarcastic Scaled in normal individuals (Colligan et al.,
and caustic. They are seen as having few de¬ 1983, 1989), medical patients (Swenson et al.,
fenses to protect them from the external envi¬ 1973), or psychiatric patients (Hedlund &
ronment and consequently are vulnerable to a Won Cho, 1979). Scale d is likely to be a peak
harsh and overwhelming environment. score in normal women and unlikely to be a
Elevated scores on Scale 3 can be inter¬ low point in either normal men or women
preted relatively successfully in isolation (Gulas, 1974). Scale d will reliably separate
148 Chapter 4

groups of clients with psychophysiological ing psychological problems into somatic

symptoms (Lair & Trapp, 1962) and conver¬ complaints. The overall elevation of this con¬
sion symptoms (Fricke, 1956) from other figuration reflects the amount of psychologi¬
groups, but pronounced overlap usually ex¬ cal distress that the client is experiencing.
ists among the individual scale scores in these As Scales 1 and 3 both approach a T
groups of clients. score of 90, the tenuousness of these defenses
The interpretation of Scale 5 at four lev¬ becomes readily apparent to everyone except
els of elevation is summarized in Table 4-5. the client. The relative elevation of Scale 2
compared to Scales 1 and 3 also reflects the
adequacy of the “conversion defenses.” The
Neurotic Triad Configurations
greater the relative elevation of Scales 1 and 3
Four configurations encompass the most fre¬ compared to Scale 2, the more severe, long¬
quently encountered relationships among the standing, and resistant to change are the
three scales in the neurotic triad—Scales 1 client’s defenses against facing the actual
(Hypochondriasis), 2 (Depression), and 3 source of distress in his or her life.
(Hysteria). Since these configurations will be The other important characteristic of
described in more detail in Chapter 6 under this configuration is the relative elevation of
their respective codetypes, the review here Scales 1 and 3. When Scale 3 is higher than
will be brief. Scale 7, the client tends to be optimistic about
The first configuration is a conversion physical complaints, which are specific and
“V” (see Figure 4-1). A client with this con¬ usually focused in the head and the extremi¬
figuration is converting personally distressing ties. In contrast, the client with Scale 7 higher
troubles into more rational or socially accept¬ than Scale 3 tends to be bitter and pessimistic
able problems; that is, the person is convert¬ about vague and general physical complaints.

TABLE 4-5 Interpretation of Levels of Elevation on Scale 3: Hysteria (Hy)

MMPi-2 MMPI
T Score T Score Interpretation

44 and 40 and 1. Low. These clients tend to be caustic, sarcastic, and socially iso-
below below lated. They have few defenses. They are seen as having narrow
interests and being socially conforming.
45-57 41-59 2. Normal. These clients have a typical number of attitudes and be¬
haviors that relate to hysteric dynamics.
58-64 60-69 3. Moderate. These clients are likely to be exhibitionistic, extro¬
verted, and superficial. They are naive, self-centered, and deny
any problems. They prefer to look on the optimistic side of life and
avoid unpleasant issues. Review of the content and/or supplemen¬
tary scales may facilitate interpretation in this range.
65 and 70 and 4. Marked. These clients are naive, suggestible, lack insight into their
above above own and others' behavior, and deny any psychological problems.
Under stress, specific physical complaints will appear. Despite the
initial positive impression they make on the clinician, any form of
psychological intervention will be difficult. They look for simplistic,
concrete solutions to their problems, solutions that do not require
self-examination. Review of the content and/or supplementary
scales may facilitate interpretation at the lower end of this range.
 Clinical Scales 149

FIGURE 4-1 Neurotic Triad Configuration: FIGURE 4-2 Neurotic Triad Configuration:
Conversion "V" Descending Slope

1 2 3 1 2 3
Scale Scale

The emphasis on physical complaints fested by hypersensitivity to even the most

along with the denial of any psychological minor dysfunction, and they have constant
basis for them makes all members of this physical complaints without adequate physi¬
group poor candidates for any form of psy¬ cal pathology.
chological treatment. It is common for this Somatic symptoms often include nau¬
neurotic triad configuration to be accompa¬ sea, dizziness, insomnia, and headaches.
nied by a similar validity scale configuration These clients typically have stable work re¬
(see page 120). cords and marital relationships. As would be
Another common neurotic triad config¬ expected, they see little if any correlation be¬
uration is the descending pattern (see Figure tween their physical complaints and any psy¬
4-2). The essential feature of this pattern is chological problems. Prognosis is poor for
that all three scales are elevated above a T any short-term psychological intervention.
score of 65, with Scale 1 being the highest, This configuration is frequently found in
followed by Scales 2 and 3 in descending males over the age of 35 who feel “over the
order. Clients with this configuration have a hill.”
long-standing somatic overconcern mani¬ Figure 4-3 illustrates a third common
150 Chapter 4

FIGURE 4-3 Neurotic Triad Configuration: scribed as dependent and immature. Such
Caret persons often have learned to tolerate great
unhappiness and a high level of discomfort;
consequently, they may have poor motiva¬
tion for treatment. They seem to operate at
low levels of efficiency for extended periods
of time.
The fourth common neurotic triad con¬
figuration is the ascending pattern (see Figure
4-4). In this configuration all three scales are
greater than a T score of 65 and each succeed¬
ing scale is higher than the previous one. This
pattern typically is found in females who
present a history of gynecologic complaints.
(Duckworth and Anderson [1986] call this the
“hysterectomy” profile.) The women report

FIGURE 4-4 Neurotic Triad Configuration:

Ascending Slope

1 2 3
Scale

neurotic triad configuration. Its main feature

is the elevation of Scale 2; although all three
scales are elevated, Scale 2 is higher than
Scales 1 and 3. These clients have a chronic
neurotic condition with mixed symptomatol¬
ogy. Multiple somatic complaints, depres¬
sion, and hysteroid features are typical, par¬
ticularly as this configuration increases in
overall elevation. When Scale 1 is below a T
score of 65 and Scales 2 and 3 are above a T
score of 70, the client frequently is over¬
controlled emotionally and reports feeling
“bottled up.”
These clients usually are fatigued, anx¬
ious, and filled with self-doubts, which pre¬ 1 2 3
vent them from doing anything. They are de¬ Scale
 Clinical Scales 151

many marital problems, including sexual forethought and with little effort to avoid
complaints such as frigidity and a lifelong being caught.
history of ill health. Males with this configu¬ All members of the criterion group,
ration are likely to be in chronic states of anx¬ which included more females than males,
iety and exhibit physical effects of prolonged were involved in legal proceedings, and many
tension and worrying, such as gastric distress were incarcerated. Hence, their emotional re¬
and ulcers. sponses of depression and boredom could
In both men and women this configura¬ have reflected their current circumstances
tion reflects a mixed neurotic pattern with de¬ rather than any real, inherent characteristics.
pression and somatization predominating. A The responses of this criterion group were
high level of anxiety with insomnia and an¬ contrasted with those of a sample of the mar¬
orexia usually accompanies the clinical pic¬ ried members of the original Minnesota nor¬
ture. Lack of psychological insight and resis¬ mative group and a sample of college appli¬
tance to psychological interpretation of cants. This procedure resulted in the 50 items
behavior are typical of clients with this con¬ currently on Scale 4.
figuration. Examples of this scale’s items with the
deviant response in parentheses are:

SCALE 4; PSYCHOPATHIC “In school I was sometimes sent to the

DEVIATE (Pd) principal for bad behavior.” (true)
“My way of doing things is apt to be
General social maladjustment and the ab¬
misunderstood by others.” (true)
sence of strongly pleasant experiences are as¬
sessed by the 50 items of Scale 4 (McKinley & “I have been quite independent and free
Hathaway, 1944). The major content areas of from family rule.” (false)4
the items are diverse and in some cases seem
contradictory. Items tap complaints about McKinley and Hathaway (1944) also
family and authority figures in general, self- cross-validated Scale 4 by examining two
and social alienation, and boredom. Other other groups’ total score on the scale. These
items assess the denial of social shyness and two groups were a sample of psychiatric inpa¬
the assertion of social poise and confidence. tients and a group of prison inmates, all diag¬
As with Scale 3, the simultaneous endorse¬ nosed as psychopathic personality. A T score
ment of apparently contradictory groups of of 70 or above on Scale 4 was achieved by 59
items was particularly characteristic of the percent of the prisoners and 45 percent of the
criterion group used to construct Scale 4. inpatients. McKinley and Hathaway called
Scale 4 was constructed empirically this scale psychopathic deviate to indicate
using a criterion group of young persons pri¬ that it was not expected to differentiate all
marily between the ages of 17 and 22 diag¬ cases of psychopathic personality. Rather,
nosed as psychopathic personality, asocial Scale 4 could identify about one-half or more
and amoral type, who were referred for test¬ of those clients diagnosed as psychopathic
ing by the courts because of their delinquent personality.
activities. None of the criterion cases was a Scale 4 has substantial overlap with most
major criminal type; most were characterized of the validity and clinical scales. It shares
by a long history of minor delinquency. five to ten items with all clinical scales except
When they engaged in delinquent behavior, Scales 1 (Hypochondriasis) and 5 (Masculin¬
they generally did so without planning or ity-Femininity), and it shares five and seven
152 Chapter 4

items with scales Fand K, respectively. Scale constructing these subscales, Harris and Lin¬
4 has almost an equal number of “true” and goes added to each subscale two to six items
“false” deviant responses, and it has slightly not found on Scale 4. They did not provide a
more obvious than subtle items (Wiener, rationale for adding these items, nor did they
1948). add items to the subscales for any of the other
Factor analyses of the items have yielded clinical scales.
similar results in a variety of populations. Several of Harris and Lingoes’ subscales
Generally five factors are identified: shyness, overlap substantially with the factors identi¬
hypersensitivity, delinquency, impulse con¬ fied by factor analysis that were described
trol, and neuroticism (Astin, 1959, 1961; above: Social Alienation overlaps with hyper¬
Comrey, 1958a). Comrey (1958a) also identi¬ sensitivity, Self-Alienation with neuroticism,
fied a family dissension factor in his mixed and Social Imperturbability with shyness.
sample of normals and psychiatric patients, Further, the Familial Discord subscale shares
but Astin (1959) did not in his sample of adult 8 of its 11 items with Wiggins’ (1966) Family
male narcotic addicts. Monroe, Miller, and Problems scale, and the Authority Conflict
Lyle (1964) extended Astin’s factor-analytic subscale shares 3 of its 11 items with the Wig¬
scales to assist in screening adult addict pa¬ gins Authority Conflict scale. Thus, it ap¬
tients. pears that regardless of the method used to
Harris and Lingoes (1955) identified construct subscales within Scale 4, four or
four subscales within Scale 4: Familial Dis¬ five factors are frequently identified in most
cord, Authority Conflict, Social Imperturb¬ populations.
ability, and Alienation (see Table 4-6). They High scorers on Scale 4 usually are de¬
further subdivided their Alienation scale into scribed in unfavorable terms: angry, impul¬
Social Alienation and Self-Alienation. In sive, emotionally shallow, and unpredictable.

TABLE 4-6 Description of High Scorers on the Harris and Lingoes Subscales
for Scale 4 (Psychopathic Deviate)

Subscale

Abbre¬ Number
Name viation of Items Description of High Scorers

Familial Discord Pd1 9 These clients struggle against familial

control.
Authority Conflict Pd2 8 These clients resent societal demands
and conventions and parental standards.
Social Imperturbability Pd 3 6 These clients deny social anxiety, exhibit
blandness, and deny dependency needs.
Social Alienation Pd4 13 These clients feel isolated from other
people; they lack feelings of belonging¬
ness, externalize blame for their difficul¬
ties, and lack gratification in their social
relations.
Self-Alienation Pd 5 1 2 These clients lack self-integration; they
avow guilt, exhibitionistically stated,
and are despondent.
 Clinical Scales 153

They are socially nonconforming, disregard¬ guilt about his behavior. Frequently, a high
ing social rules and conventions in general scorer will not elevate Scale 2 at all even when
and authority figures in particular. They har¬ external constraints are being placed on his or
bor a brooding resentment and hostility to¬ her behavior. In these circumstances, a suc¬
ward authority figures, which may or may cessful psychotherapeutic intervention is
not be overtly displayed. In the absence of an highly unlikely because the clients lack con¬
antisocial history, this hostility may have cern about their behavior. For most high
been directed inward toward the self. Thus, a scorers on Scale 4, personal maturation
marked elevation on Scale 4 indicates the rather than some other form of intervention
presence of antisocial behavior and attitudes, is usually most effective in changing their be¬
but it does not necessarily mean that these be¬ havior.
haviors will be expressed overtly. High scorers on Scale 4 are very likely to
Other clinical scales, especially Scale 9 be diagnosed as having some form of person¬
(Hypomania), usually are elevated if these ality disorder; they are unlikely to receive a
antisocial behaviors are overtly expressed. psychotic diagnosis. They often have a long
High scorers have a perfectionistic and nar¬ history of inadequate familial and social rela¬
cissistic conception of themselves, and they tionships, which seems to reflect a character-
use these personal standards as a rationaliza¬ ologic adjustment.
tion for ignoring social conventions. This Elevations on Scale 4 are positively cor¬
perfectionistic self-concept is illustrated by related with the frequency of delinquent and
the typical profile for a high scorer: Scale 4 is criminal behaviors and recidivism rates
the only clinical scale elevated above a T (Forgac, Cassel, & Michaels, 1984; Gearing,
score of 65 (a Spike 4 codetype). Rather than 1979; Holland & Levi, 1983). Hare’s (1985)
representing conscious avoidance of deviant finding of little agreement between clinical
responses, this codetype seems to reflect their judgments or behavioral measures and self-
perfectionistic conception of themselves. report measures for assessing psychopathy
High scorers (T scores of 65 or higher) should be noted by clinicians who are inter¬
are socially outgoing, energetic, and socially ested in researching this area and trying to in¬
facile; thus, they make a good initial impres¬ tegrate the information.
sion on others. Longer exposure to them soon Heilbrun (1979) found that an index of
reveals, however, their irresponsibility, un¬ psychopathy based in part on the raw score
reliability, moodiness, and resentment. Nov¬ on Scale 4 was directly related to the fre¬
ice clinicians frequently are awed by this ini¬ quency of violent crimes in a sample of white
tial impression and then dismayed when prisoners, but only in those prisoners with
underlying qualities begin to emerge. The cli¬ IQs less than 95. Since violent crimes tend to
nician should be aware of the favorable ini¬ be impulsive, Heilbrun speculated that the re¬
tial impression typical of high scorers and not lationship between violent crime and lower
be overly influenced by it. levels of intelligence may reflect a limit on or
Depression, when evident in a high lack of temporary cognitive restraints.
scorer, usually consists of depressive Gearing (1979) provides a comprehen¬
thoughts and feelings but not actually psy¬ sive review of the use of the MMPI in prison
chomotor retardation nor the other vegeta¬ settings. Walters (1985) reported that prison¬
tive signs of depression. The depression most ers with behavioral diagnoses of Antisocial
often represents dissatisfaction about the cur¬ Personality Disorder scored higher on Scale 4
rent limits being placed on the high scorer’s than prisoners without such behavioral diag¬
behavior rather than any actual concern or noses.
154 Chapter 4

Normal persons who achieve high scores month and coefficients of .49 to .61 for inter¬
on Scale 4 (T scores of 65 or higher) are de¬ vals up to a year (Dahlstrom et al., 1975).
scribed in fairly similar and unflattering Test-retest reliability coefficients for Scale 4
terms: rebellious, immature, exhibitionistic, on the MMPI-2 for approximately a one-
unconventional, and nonconforming. Such week interval are .81 for men and .79 for
persons may display a generalized deviancy women (Butcher et al., 1989).
from societal standards and conventions, al¬ There are few gender differences in the
though they are not displaying psychopathic distribution of raw scores on Scale 4. In fact,
behaviors per se. the T score equivalents of each raw score are
Nearly 10 percent of college students ob¬ identical on the standard MMPI profile
tain T scores greater than 70 on Scale 4 of the sheet, and they are generally within one or
MMPI, and the possibility of developing sep¬ two T points on the MMPI-2. Scores on Scale
arate norms for college populations has been 4 tend to decrease significantly (5 to 10 T
discussed (Forsyth, 1967; Goodstein, 1954; points) with age in all populations (Colligan
Murray, Munley, & Gilbart, 1965). King and et al., 1983, 1989; Hedlund & Won Cho,
Kelley (1977a), however, found that college 1979; Swenson et al., 1973); this pattern is
students requesting counseling who elevated thought to reflect the slow maturational
Scale 4 above a T score of 70 had a history of changes that occur in persons with elevated
legal, academic, and criminal difficulties. scores. Several normal groups score in the
These students also were likely to be diag¬ high-normal and low-moderate range on
nosed as having a personality disorder. Thus, Scale 4\ social activists, adolescents, and
at least in this sample, elevation of Scale 4 mental health professionals.
seems to reflect significant pathology. The interpretations of Scale 4 at four
Whether similar psychopathology accompan¬ levels of elevation are summarized in Table
ies elevations of Scale 4 among college stu¬ 4-7.
dents who are not seeking psychological ser¬
vices remains to be examined.
SCALE 5;
Butcher, Graham, and Bowman (1990)
MASCULINITY-FEMININITY (Mf)
reported that normal college students had
similar scores on Scale 4 and the other valid¬ The 56 items (60 items on the MMPI) com¬
ity and clinical scales of the MMPI-2 as the prising Scale 5 are very heterogeneous in con¬
MMPI-2 normative group. They concluded tent. Major content areas include interests in
that the MMPI-2 norms were appropriate for vocations and hobbies, aesthetic preferences,
college students so separate norms would not activity-passivity, and personal sensitivity.
be needed. Scale 5 was developed in a slightly differ¬
Low scorers on Scale 4 (T scores less ent manner than the other clinical scales.
than 45) are generally described as conven¬ Hathaway and McKinley (Hathaway, 1956)
tional, conforming, and submissive. These had intended to use a large sample of homo¬
persons are socially constricted, rigid, and sexual males and females in empirically de¬
have narrow interests. Males are frequently veloping a scale of masculinity-femininity,
described as being uninterested in sexual ac¬ but they quickly discovered that homosexual
tivity, particularly when Scale 4 is the low samples were too heterogeneous to use as a
point in the profile. single criterion group. They identified at least
Scores on Scale 4 on the MMPI tend to three subgroups of homosexuals within their
be fairly stable with test-retest reliability co¬ samples, each with an apparently different
efficients of .59 to .84 for intervals up to one source or cause of homosexuality.
 Clinical Scales 155

TABLE 4-7 Interpretation of Levels of Elevation on Scale 4: Psychopathic Deviate (Pd)

MMPI-2 MMPI
T Score T Score Interpretation

44 and 40 and 1 Low. These clients tend to be rigid and conventional. They usually
below below are able to tolerate much mediocrity and boredom. Males may lack
interest in heterosexual activity, particularly if this scale is the low
point.
45-57 45-59 2 Normal. These clients have a typical number of complaints about
authority, alienation, and boredom.
58-64 60-69 3 Moderate. These clients may be genuinely concerned about social
problems and issues; they may be responding to situational con¬
flicts, or they may have adjusted to an habitual level of interpersonal
and social conflict. If the conflict is situational, the score should re¬
turn to the normal range as the conflict is resolved. Review of the
content and/or supplementary scales may facilitate interpretation
of scores in this range.
65 and 70 and 4. Marked. These clients are fighting against something, which is
above above usually some form of conflict with authority figures. These con¬
flicts may not necessarily be acted out overtly; the rebelliousness
and hostility toward authority figures are readily apparent even in
these cases. They are likely to be unreliable, egocentric, and irre¬
sponsible. They may be unable to learn from experience or to plan
ahead. These clients have a good social facade and make a good
initial impression, but the psychopathic features will surface in
longer interactions or under stress. Psychological interventions are
less effective than maturation in achieving change. Review of the
content and/or supplementary scales may be helpful at the lower
end of this range.

Because of their difficulty in obtaining a ple, separate groups of normals had to be

large number of cases within each subgroup, gathered to contrast with the criterion group.
they decided to restrict their criterion sample These normal groups consisted of 54 male
to one subgroup—male homosexual inverts. soldiers and 67 female airline employees. The
Such persons were thought to engage in ho¬ initial item selection for Scale 5 resulted from
moerotic behavior as a part of their feminine contrasting the normal males with the crite¬
(i.e., inverted) personality characteristics; rion group (Dahlstrom et al., 1972). Items se¬
many such men, however, are too inhibited lected on this basis were then checked to in¬
or conflicted to express their homosexuality sure that they separated the “normal” males
overtly (Dahlstrom et al., 1972). The primary from the “normal” females.
criterion group then consisted of 13 homo¬ Finally, a group of feminine males was
sexual invert males who were selected for defined by the Attitude-Interest Analysis Test
their freedom from any form of psychopa¬ of Terman and Miles (1938), and the re¬
thology. sponses of these feminine males were con¬
Since most of the items used to identify trasted with those of the normal males. The
sexual inversion were added to the item pool 60 items that contrasted the groups in all
after the data had already been collected three comparisons became Scale 5 on the
from the original Minnesota normative sam¬ MMPI.
156 Chapter 4

Hathaway and McKinley (Hathaway, still refer to the clinical scales excluding
1956) were unsuccessful in their attempt to Scales 5 and 0 (Social Introversion), which re¬
develop a separate scale (Fm) to identify fe¬ flects the early tradition of MMPI usage.
male homosexual inversion. They found that Pepper and Strong (1958) rationally
their Fm scale correlated positively with Scale identified five subgroups of items within
5, so they abandoned it in favor of a single Scale 5: Personal and Emotional Sensitivity,
scale of Masculinity-Femininity. Sexual Identification, Altruism, Feminine
Examples of Scale 5 items with the devi¬ Occupational Identification, and Denial of
ant response for men in parentheses are: Masculine Occupations. Serkownek (1975)
developed six subscales within Scale 5 based
“I think I would like the work of a li¬ on the factor analysis carried out by Graham,
brarian.” (true) Schroeder, and Lilly (1971): Narcissism-
“I like collecting flowers or growing Hypersensitivity; Stereotypic Feminine Inter¬
house plants.” (true) ests; Denial of Stereotypic Masculine Inter¬
“I like mechanics magazines.” (false)5 ests; Heterosexual Discomfort-Passivity;
Introspective-Critical; and Socially Retiring
The items on Scale 5 have not been in¬ (see Table 4-8).
vestigated as extensively as the other clinical There is extensive item overlap between
scales. Part of this lack of research is because the Pepper and Strong (1958) item groupings
the scale was not routinely scored in the early and the Serkownek (1975) subscales. For ex¬
years of the MMPI. In fact, some clinicians ample, 12 of the 14 items on Serkownek’s

TABLE 4-8 Description of High Scorers on the Serkownek Subscales

for Scale 5 (Masculinity-Femininity)

Subscale

Abbre¬ Number
Name viation of Items Description of High Scorers

Narcissism- Mf! 18 These clients are sensitive to reactions

Hypersensitivity of others; they worry constantly and
are easily hurt and upset.
Stereotypic Feminine Mf2 12 These clients profess interests that are
Interests clearly feminine in character.
Denial of Stereotypic Mf3 8 These clients deny interests that are
Masculine Interests clearly masculine in character.
Heterosexual Discomfort- Mf4 3 These clients admit homosexual im¬
Passivity pulses but feel uncomfortable talking
about sexual matters.
Introspective-Critical Mf, 6 These clients neither enjoy nor feel
comfortable in loud, active social
gatherings.
Socially Retiring MU 9 These clients deny a liking for a num¬
ber of social and cultural activities in
which a dominant and conspicuous
role is possible.
 Clinical Scales 157

Stereotypic Feminine subscale overlap with lates of males and females at various eleva¬
Pepper and Strong’s Feminine Occupational tions on Scale 5; thus, separate interpreta¬
Identification. Consequently, both sets of tions for males and females must be used.
items do not need to be routinely scored. Hathaway and McKinley (Hathaway,
Serkownek’s subscales were selected for use 1956) apparently assumed that masculinity-
in this manual because of their empirical der¬ femininity was a bipolar dimension with mas¬
ivation and because of the lack of construct culinity at one end and femininity at the
validity for some of the Pepper and Strong other. Numerous investigators (cf. Aaron-
groups of items (Martin & Greene, 1979). son, 1959; Gonen & Lansky, 1968; Sines,
Harris and Lingoes (1955) did not de¬ 1977) have suggested that Scale 5 is not bipo¬
velop subscales for Scale 5, nor did Wiener lar and that it is more likely to be multidimen¬
and Harmon (Wiener, 1948) develop obvious sional. Both factor-analytic (Graham et al.,
and subtle subscales. Dahlstrom and col¬ 1971) and rational (Pepper & Strong, 1958)
leagues (1972) state that most of the items are subscales formed with Scale 5 items, which
psychologically obvious, which supports the were described above, also support its multi¬
contention that homosexuals can conceal dimensional nature.
their sexual orientation without being de¬ Constantinople (1973) provided an ex¬
tected by this scale (Bieliauskas, 1965). There cellent review questioning whether masculin¬
are approximately the same number of ity-femininity is a bipolar dimension that can
“true” and “false” deviant responses to be adequately measured by a single score;
Scale 5 items, which means that either an “all Baucom (1976) empirically demonstrated
true” or “all false” response set will not ap¬ that independent measures of masculinity
preciably affect raw scores on this scale. and femininity can be developed from the
The same 56 items are used to assess California Psychological Inventory (Gough,
masculinity-femininity for males and fe¬ 1957) ; and Peterson (1989) developed inde¬
males. Responses to these items are scored as pendent scales of masculinity and femininity
deviant when they reflect femininity in men for the MMPI-2 (see the Supplementary
and masculinity in women. Thus, high T Scales for Gender Role—Masculine and Gen¬
scores result when a client endorses the items der Role—Feminine in Chapter 5).
like a person of the opposite gender. For 4 of There also is a paucity of data on the be¬
the items (121, 166, 209, 268), which deal havioral correlates of Scale 5 in various pop¬
with the admission of sexually deviant behav¬ ulations, again reflecting the early history of
iors, the same response is considered deviant the MMPI in which this scale was not rou¬
for both males and females. For the other 52 tinely scored nor described in most research.
items, the scoring for a deviant response is re¬ What research has been done discusses males
versed for the genders. For these latter 52 and females separately because of the very
items, if “true” is the deviant response to a different correlates of specific T scores for
specific item for males reflecting femininity, the two genders.
“false” is the deviant response for females High-scoring males (T scores of 65 or
reflecting masculinity. higher) in psychiatric populations are de¬
This procedure was intended to provide scribed as passive, socially sensitive and per¬
a basis of uniform interpretations of elevated ceptive, having a wide range of aesthetic and
scores since these scores would indicate that social interests, and inner-directed. They also
the person was endorsing the items like a per¬ are seen as being dependent and insecure re¬
son of the opposite sex. Immense differences garding their masculine role; often they tend
exist, however, between the behavioral corre¬ to identify with a feminine role. Depression,
158 Chapter 4

anxiety, tension, and guilt frequently are re¬ scales, clients are likely to be seen as not hav¬
ported (Ward & Dillon, 1990). If males are ing a psychiatric disorder even in a psychiat¬
homosexual or have homosexual concerns ric setting (King & Kelley, 1977b; Rosen,
and are willing to acknowledge openly these 1974). High-scoring normal males are gener¬
behaviors or concerns, they will achieve high ally described in positive terms: curious, so¬
scores on Scale 5 (Aaronson & Grumpelt, cially perceptive, peaceable, tolerant, and
1961; Dean & Richardson, 1964; Friberg, psychologically complex. They also are de¬
1967; Manosevitz, 1971; Singer, 1970). If scribed as passive and prone to worry. They
they cannot acknowledge these behaviors, have wide philosophical and aesthetic inter¬
they will not elevate the scale at all since item ests.
content on Scale 5 is psychologically obvious Among females in psychiatric popula¬
(Wong, 1984). tions, high scorers (T scores of 65 or higher)
The fact that a variety of factors, such as are seen as being aggressive, unfriendly,
education and vocational interests, tend to be dominating, and competitive. In an inpatient
associated with elevated scores on Scale 5 of psychiatric setting high-scoring females have
the MMPI in males further reduces the use¬ features of a psychotic thought disorder
fulness of this scale in diagnosing homosexu¬ (Boerger et al., 1974). They have difficulty
ality (Burton, 1947; Friberg, 1967), since the remembering, are slow moving and sluggish,
elevation on the scale may reflect these fac¬ and report hallucinations, thinking distur¬
tors rather than homosexual behaviors or bances, and psychomotor withdrawal-retar¬
concerns. Raw scores on Scale 5 are posi¬ dation.
tively correlated with level of education, par¬ High-scoring normal females (T scores
ticularly in the T score range of 55 to 65. Ed¬ of 65 or higher) are seen as being adventur¬
ucation per se, however, is insufficient to ous. Somewhat surprisingly, no one has re¬
elevate scores much above a T score of 65 on ported that they have masculine interests or
the MMPI (Dean & Richardson, 1964; that they do not have feminine interests. The
Manosevitz, 1971). Consequently, high lack of additional correlates of high scores in
scores in males cannot simply be dismissed as normal females reflects the infrequency with
reflecting their humanistic and liberal arts- which such scores occur on the MMPI.
oriented college education. Low-scoring males (T scores less than
The interpretation of Scale 5 on the 40) are easygoing, adventurous, and coarse.
MMPI-2 may be problematic because of the They may display an almost compulsive mas¬
relatively high education level of the norma¬ culinity and will emphasize their masculine
tive sample (see Chapter 1). It is apparent interests. They tend to lack individuality and
that scores on Scale 5 tend to be about 10 T originality.
points lower in men and 2 to 3 T points higher Low-scoring females (T scores less than
in women on the MMPI-2 than on the origi¬ 40) are passive, submissive, yielding, and de¬
nal MMPI. Research is clearly needed to de¬ mure. They strongly identify with a tradi¬
termine whether the MMPI correlates of tional feminine role. Extremely low-scoring
Scale 5 that are described below can be ap¬ females (T scores below 35) are likely to be
plied directly to the MMPI-2. Clinicians constricted, self-pitying, faultfinding, and
should recall that linear T scores are used on self-deprecating. They frequently display al¬
Scales 5 and 0 of the MMPI-2 rather than most a caricature of an extreme feminine
uniform T scores. role. They appear helpless and utterly depen¬
When only Scale 5 is elevated without dent on significant others to take care of
accompanying elevations on other clinical them. This behavior often is manipulative
 Clinical Scales 159

but can occasionally represent genuine help¬ as being less formal, mild, opportunistic, and
lessness. silent than the undergraduates. It appears
Test-retest reliability coefficients for that women with low scores on Scale 5 may
Scale 5 of the MMPI range from .72 to .91 be described in a variety of different terms
for intervals up to two weeks, with very sim¬ depending upon their level of education.
ilar coefficients for males and females. Reli¬ Summary tables of the interpretations of
ability coefficients range from .34 to .63 for four levels of scores on Scale 5 are provided
intervals up to one year, again with little dif¬ for men (see Table 4-9) and for women (see
ference between males and females (Dahl- Table 4-10).
strom et al., 1975). Test-retest reliability coeffi¬
cients for Scale 5 on the MMPI-2 for
SCALE 6: PARANOIA (Pa)
approximately a one-week interval are .82 for
men and .73 for women (Butcher et al., 1989). Interpersonal sensitivity, moral self-righ¬
Education has a predominant influence teousness, and suspiciousness are revealed by
on raw scores on Scale 5 in males and some¬ the 40 items that make up Scale 6 (Paranoia).
what less influence on raw scores in females. The content of some items is clearly psy¬
T scores for males can be expected to be in the chotic, acknowledging the existence of delu¬
range of 55 to 70 as a result of a college edu¬ sions and paranoid thought processes.
cation and the vocational interests and train¬ Hathaway and McKinley (Hathaway, 1956)
ing that are part of the education process. A never described the criterion group of para¬
male who has a liberal arts education can be noid patients used to develop Scale 6 empiri¬
expected to score at the upper end of this cally. They considered Scale 6 to be a weak
range, whereas a male who has an engineer¬ preliminary scale although they were unable
ing or basic sciences oriented degree will to develop a better scale.
score at the lower end of this range. In any case, the exact number and com¬
Males with education below the college position of patients in the paranoid criterion
level also can be expected to score at the group are not known. These patients were
lower end of this range or below. Conse¬ judged to have paranoid symptoms, although
quently, either a low T score in a male liberal few of them were classified as “only” para¬
arts major or a high T score in a male high noid. Most were diagnosed as having a para¬
school graduate should be investigated by the noid state, paranoid condition, or paranoid
clinician because of the unusualness of such schizophrenia. It can be assumed that Scale 6
scores. To the extent that social class will af¬ was empirically derived by contrasting the
fect a male’s vocational interests and his con¬ item endorsements of this unspecified para¬
ceptions of what behaviors are appropriate noid criterion group with the original Minne¬
within a masculine role, social class affects sota normative group.
Scale 5 T scores, as does education. Examples of scale items with the deviant
Graham and Tisdale (1983) reported a response indicated in parentheses are:
number of differences between female gradu¬
ate and undergraduate students with low
“I have certainly had more than my
scores on Scale 5. The graduate students de¬
share of things to worry about.” (true)
scribed themselves and were described as
being more conscientious, insightful, reflec¬ “I have no enemies who really wish to
tive, unaffected, and unconventional than harm me.” (false)
the undergraduates. The graduate students “1 believe I am being plotted against.”
also described themselves and were described (true)6
160 Chapter 4

TABLE 4-9 Interpretation of Levels of Elevation for Men on Scale 5;

Masculinity-Femininity (Mf)

MMPI-2 MMPI
T Score T Score Interpretation

44 and 40 and 1. Low. These men identify very strongly with the traditional mascu-
below below line role, and they may be compulsive and inflexible about their
masculinity.
45-57 41 -59 2. Normal. These men are interested in traditional masculine interests
and activities. This also is the typical range for college-educated
males in the more masculine-oriented fields, such as engineering
and agriculture.
58-64 60-69 3. Moderate. These men tend toward aesthetic interests such as art,
music, and literature. They are rather passive and prefer to work
through problems in a covert and indirect manner. This is the typical
range for most college-educated males.
65 and 70 and 4. Marked. These men are passive, inner-directed, and have aes-
above above thetic interests and activities. They do not identify with the tradi¬
tional masculine role. Self-proclaimed homosexuals and persons
willing to admit overtly their homosexual concerns will score in
this range. Homosexual behavior and/or concerns, however, can be
easily concealed without elevating Scale 5 or the validity scales.
Be very wary of diagnosing a client as homosexual solely on the
basis of a score in this range.

There has been only limited study of the MMPI-2 content scale (Butcher et al., 1989)
individual items on Scale 6. Comrey’s of Bizarre Mentation (see Chapter 5) overlaps
(1958b) factor analysis of the items revealed with only 8 of the 40 items on Scale 6.
four factors of paranoia—actual persecu¬ Since Endicott, Jortner, and Abramoff
tion, imagined persecution, delusion, and (1969) reported only small positive correla¬
hopelessness, guilt-ridden—as well as several tions between clinical ratings of suspicious¬
other factors that are not clearly related to ness and Scale 6 scores, it may be that Scale 6
paranoia—neuroticism, cynicism, hysteria, is an inadequate measure of the construct of
and rigidity. Harris and Lingoes (1955) devel¬ suspiciousness, which is a central feature of
oped three subscales within Scale 6: Persecu¬ paranoid behavior. Nevertheless, it appears
tory Ideas, Poignancy, and Naivete (see that regardless of the method used to develop
Table 4-11). There is nearly complete overlap subscales within Scale 6, three factors are
between Comrey’s (1958b) paranoid, neurot¬ identified, and these can be adequately as¬
icism, and cynicism factors and Harris and sessed by the Harris and Lingoes’ (1955) sub¬
Lingoes’ (1955) three subscales of Persecutory scales.
Ideas, Poignancy, and Naivete, respectively. Wiener and Harmon (Wiener, 1948) felt
Wiggins’ (1966) Psychoticism scale overlaps that more Scale 6 items were obvious (23)
with 19 of the 40 items on Scale 6 and shares than subtle (17). Only 7 items are unique to
numerous items with Comrey’s four paranoid the scale; 1 to 4 items are shared with most of
factors. The Tryon, Stein, and Chu (Stein, the other clinical scales and even more sub¬
1968) subscale of Suspicion and Mistrust stantial overlap occurs with Scales F (9
shares only one item with Scale 6. The items), 4 (8 items), and <5(13 items). There are
 Clinical Scales 161

TABLE 4-10 Interpretation of Levels of Elevation for Women on Scale 5:

Masculinity-Femininity (Mf)

MMPI-2 MMPI
T Score T Score Interpretation

34 and 34 and 1. Markedly Low. These women tend to be coy, seductive, and ap¬
below below pear helpless. They overidentify with the feminine role and at times
are almost a caricature of it. This behavior may be subtly manipula¬
tive or they may conceive of themselves as actually helpless.
These clients should elevate the supplementary scale of Gender Ro¬
le-Feminine.
35-44 35-44 2 Low. These women are genuinely interested in traditional feminine
interests and activities. They may be passive in their role.
45-64 45-59 3 Normal and Moderate. These women are less traditionally oriented
toward a feminine role than women who score low, and they have
an interest in masculine activities as well.
65 and 60 and 4 Marked. It is unusual for most women to score in this range. Check
above above for scoring or profiling errors. The male template instead of the
female template may have been used in scoring; male norms instead
of female norms may have been used (i.e., the wrong table in an
appendix was used); or the woman's profile was plotted on the male
side of the profile sheet instead of the female side. These women
may or may not have actual masculine interests, but they are defi¬
nitely not interested in appearing or behaving according to a tradi¬
tional feminine role. They may become anxious if they are expected
to limit their behavior to what is prescribed by a traditional femi¬
nine role. Aggressive behavior is likely to be seen. Homosexual be¬
havior is unlikely to be seen.

slightly more “true” (24) than “false” (16) Dahlstrom and associates (1972) ob¬
deviant responses to Scale 6 items. As the “all served that Scale 6 is quite sensitive to fluctu¬
true” profile (page 125) illustrates, Scale 6 ations in the degree and intensity of delu¬
and the succeeding scales have an increas¬ sional material in psychiatric cases; the
ingly larger proportion of “true” responses research support for this statement, however,
as deviant. Thus, “all true” response sets can is limited at best. Vestre and Watson (1972)
be identified by the extreme elevations on the could not identify a linear or a curvilinear re¬
psychotic tetrad (Scales 6, 7, 8, and 9). lationship between Scale 6 scores and para¬
High scorers on Scale 6 are generally de¬ noid symptomatology, and Endicott and col¬
scribed as being suspicious, hostile, guarded, leagues (1969) actually found an inverse
overly sensitive, argumentative, and prone to relationship between Scale 6 scores and clini¬
blame others. They often express their hostil¬ cal ratings of suspiciousness.
ity overtly and rationalize it as a result of It also is commonly reported that Scale 6
what others have done to them. In addition, on the MMPI has few false positives (non¬
an egocentric self-righteousness seems to per¬ paranoid persons scoring above a T score of
meate all of their behavior. Although they 70), presumably because of the rather obvi¬
may not actually evidence a psychotic ous nature of the scale items (cf. Carson,
thought disorder, usually the paranoid char¬ 1969). Vestre and Watson (1972) found, how¬
acter structure is evident. ever, that 9 of 22 patients with T scores
162 Chapter 4

TABLE 4-11 Description of High Scorers on the Harris and Lingoes Subscales
for Scale 6 (Paranoia)

Subscale

Abbre¬ Number
Name viation of Items Description of High Scorers

Persecutory Ideas Pa1 1 7 These clients have ideas of external influence;

they externalize blame for their problems,
frustrations, and failures; in the extreme de¬
gree, they have persecutory ideas; they also
project responsibility for their negative feel¬
ings.
Poignancy Pa 2 9 These clients consider themselves something
special and different from other people; they
are high-strung and "thin-skinned"; they
cherish sensitive feelings; they are overly
subjective.
Naivete Pa 3 9 These clients affirm moral virtue; they are ex¬
cessively generous about the motives of oth¬
ers; they are righteous about ethical matters;
they display an obtuse naivete; they deny
distrust and hostility.

greater than 75 on Scale 6 had no rated the Dy scale seems warranted to determine
paranoid symptomatology. Consequently, whether it can reliably separate paranoid
high scorers are likely to show a paranoid groups from other diagnostic groups.
thought process if not a psychotic thought Normal individuals who score in the
disorder, but there are more exceptions to moderate range (T scores of 58 to 64) on
this statement than was formerly thought. Scale 6 are described very differently from
Because of the rigidity and suspiciousness high scorers in psychiatric populations; this
of high scorers, interpersonal contact is dif¬ appears to reflect the shift in behavioral cor¬
ficult. relates in moving from moderate to marked
Numerous investigators have found that scores as well as the population differences.
Scale 6 does not reliably separate groups of High scorers in psychiatric settings have been
paranoid clients from other diagnostic groups described. High-scoring normals are de¬
(Harris, Wittner, Koppell, & Hilf, 1970; scribed as being interpersonally sensitive,
Scagnelli, 1975; Tarter & Perley, 1975). Scag- emotional, rational, and clear thinking. To
nelli (1975) reported that the Dependency (Dy) the extent that sensitivity to others and empa¬
scale (see Chapter 5) reliably separated female thy reflects some of the same underlying dy¬
paranoid patients from other diagnostic namics as suspiciousness and projection, this
groups, with the paranoid patients scoring shift in correlates from moderate to marked
lower on the Dy scale. Further research on scores is logical.
 Clinical Scales 163

Clients who score in the normal range on ability coefficients for Scale 6 on the MMPI-
Scale 6 (T scores of 45 to 57) fall into two 2 for approximately a one-week interval are
major categories: clients without any para¬ .67 for men and .58 for women (Butcher et
noid symptomatology, and clients whose al., 1989).
paranoid symptomatology is well ingrained There are no reported gender differences
and who have sufficient reality testing to on Scale 6 of the MMPI since the same T
avoid endorsing the obvious items on Scale 6. scores are used for men and women; T scores
The latter group would be expected to elevate are very similar in men and women on the
the subtle subscale of Scale 6 (Wiener, 1948), MMPI-2. Scores tend to decrease slightly (2
but no research has addressed this issue. This to 4 T points) with age in all populations
group also provides ample evidence for the (Colligan et al., 1983, 1989; Hedlund & Won
statement that scores in the normal range on Cho, 1979; Swenson et al., 1973).
a scale may not reflect normality. A summary of the interpretations of
It is sometimes stated that extremely four levels of elevation on Scale 6 is presented
paranoid patients will get scores in the low in Table 4-12.
range (T scores less than 40) (Carson, 1969;
Good & Brantner, 1961), but the available re¬
Scale 4 5-6 Configurations
-

search indicates that their scores are more

likely to fall above the low range (Endicott One configuration of Scales 4, 5, and 6 is seen
et al., 1969; Vestre & Watson, 1972). frequently in women. Scales 4 and 6 are
Tow-scoring individuals (T scores less above a T score of 65, and Scale 5 is below a
than 45) in a psychiatric population are fre¬ T score of 35 (see Figure 4-5). It is not neces¬
quently described as stubborn, evasive, and sary that Scales 4 and 6 be the high points in
overly cautious. As previously discussed, the profile, but it is mandatory that the T
there seems to be little evidence that they are scores be in the indicated ranges since Scale 5
paranoid. Tow-scoring normals are de¬ will be 10-20 T points lower than Scales 4 and
scribed in generally positive terms: socially 6in most women. This configuration is some¬
competent, having narrow interests, trust¬ times called the “Scarlett O’Hara V.”
ing, balanced, and conventional. They some¬ These women are hostile and angry but
times are characterized as being overly trust¬ unable to express these feelings directly. They
ing and unaware of or insensitive to the resort to irritating other people into attacking
motives of others; they frequently appear them, and then seem to revel over how badly
gullible as a result. Anderson (1956) found they are mistreated. They are excessively de¬
that low-scoring college students generally manding, dependent, and have an almost in¬
were in academic difficulty due to under¬ ordinate need for affection. Unfortunately,
achievement and reported difficulties with their behaviors only serve to alienate signifi¬
their parents. He conjectured that repressed cant others, and this decreases the likelihood
or denied hostility may interfere with aca¬ that their needs will be met. As would be ex¬
demic success. pected, marital difficulties, familial prob¬
Scores on Scale 6 tend to be less stable lems, and sexual dysfunctions are common.
over time than scores on the other clinical These women are very adept at getting clini¬
scales. Test-retest reliability coefficients for cians to aggress against them, which makes
intervals up to two weeks range from .49 to therapeutic intervention very difficult.
.89, and for intervals of one year from .32 to When Scale 3 is elevated higher than this
.65 (Dahlstrom et al., 1975). Test-retest reli¬ configuration, women will be superficially
164 Chapter 4

TABLE 4-12 Interpretation of Levels of Elevation on Scale 6: Paranoia (Pa)

MMPI-2 MMPI
T Score T Score Interpretation

44 and 40 and 1. Low. These clients have narrow interests and tend to be insensi-
below below tive to and unaware of the motives of other people. Students are
frequently underachievers.
45-57 41-59 2. Normal. These clients may be very sensitive and suspicious, yet
able to avoid endorsing the obvious items. The clinician should
score the subtle and obvious scales. The client's suspiciousness
should be relatively apparent in an interview. Otherwise, scores in
this range are normal.
58-64 60-69 3. Moderate. These clients are interpersonally sensitive and think
clearly and rationally. Mental health workers frequently score in
this range. These clients may be overly sensitive to criticism and
personalize the action of others toward themselves. Review of the
content and/or supplementary scales may facilitate interpretation of
scores in this range.
65 and 70 and 4. Marked. These clients are likely to be suspicious, hostile, and
above above overly sensitive and usually overtly verbalize these qualities. A
thought disorder may be readily apparent. At the lower end of this
range review of the content and/or supplementary scales may aid
in interpretation.

sociable and deny the existence of any hostile In addition to obsessive-compulsive fea¬
feelings toward others. These women are par¬ tures, Scale 7 taps abnormal fears, self-criti¬
ticularly adept at enraging others without any cism, difficulties in concentration, and guilt
real understanding of their involvement in feelings. The item content does not reflect
the process. Their demanding, manipulative, specific obsessions or compulsive rituals; in¬
and hostile qualities are readily apparent to stead a characterologic basis for a wide vari¬
everyone but themselves. This pattern of be¬ ety of psychasthenic symptoms is tapped
havior represents a chronic means of manipu¬ (Dahlstrom et ah, 1972). The anxiety assessed
lating and controlling others, which is diffi¬ by this scale is of a long-term nature or trait
cult to alter through psychological inter¬ anxiety, although the scale is somewhat re¬
vention. sponsive to situational stress as well.
McKinley and Hathaway (1942) devel¬
oped Scale 7 empirically by contrasting a cri¬
SCALE 7: PSYCHASTHENIA (Pt)
terion group of 20 patients whose final diag¬
The 48 items of Scale 7 are designed to assess nosis was psychasthenia with a subgroup of
the neurotic syndrome of psychasthenia, the original Minnesota normative sample and
which is characterized by the person’s inabil¬ the college normative sample. At least one
ity to resist specific actions or thoughts re¬ and possibly two of the criterion group pa¬
gardless of their maladaptive nature. This tients were diagnosed incorrectly.
diagnostic label is no longer used, and such Since many patients with psychasthenia
persons are now diagnosed as having obses¬ are not so disabled as to require hospitaliza¬
sive-compulsive reactions. tion, McKinley and Hathaway (1942) were
 Clinical Scales 165

FIGURE 4-5. Scales 4 5-6 Configuration:

-
Examples of scale items with the deviant
"Scarlett O'Hara V" response indicated in parentheses are:

“I feel anxiety about something or

someone almost all the time.” (true)
“I have a habit of counting things that
are not important such as bulbs on elec¬
tric signs, and so forth.” (true)
“Bad words, often terrible words, come
into my mind and I cannot get rid of
them.” (true)7

There have been limited studies of the

items on Scale 7. Comrey (1958c) identified
seven principal factors: neuroticism, anxiety,
withdrawal, poor concentration, agitation,
psychotic tendencies, and poor physical
health. None of these factors could be labeled
psychasthenia, although some of the features
of this syndrome can be seen in the factor
names.
There is limited overlap between the
items on Scale 7 and Wiggins’ (1966) content
scales or the Tryon, Stein, and Chu (Stein,
1968) cluster scales. Scale 7 has 5 items in
4 5 6 common with Wiggins’ Phobia scale, 6 items
with Wiggins’ Poor Morale scale, and 9 items
Scale with Wiggins’ Depression scale. It has 11
items in common with Tryon, Stein, and
limited in the number of criterion cases avail¬ Chu’s Tension, Worry, and Fear scale, 8
able. They were reluctant to use outpatient items with their Depression and Apathy
cases because of the difficulty of studying scale, and 7 items with their Autism and Dis¬
such patients in depth to confirm their diag¬ ruptive Thought scale. Scale 7 has 9 items in
nosis. Once items were selected by contrast¬ common with the MMPI-2 content scale of
ing the criterion group with the two norma¬ Depression, 6 items with Anxiety, 5 items
tive samples, additional items were chosen with Obsessionality, and 4 items with Low
that correlated with the total score on this Self-esteem.
preliminary scale in a sample of 100 normal Harris and Lingoes (1955) found that the
persons. An analogous procedure was fol¬ Scale 7 items did not lend themselves to sub¬
lowed in a sample of 100 randomly selected classification, and hence they were unable to
psychiatric patients, and a few more items identify any subscales. Thus, it appears that
were added that correlated with the total the use of an internal consistency approach in
score on the preliminary scale. As a result of selecting items for Scale 7 resulted in the se¬
all these procedures, 48 items were selected lection of a wide variety of heterogeneous
for Scale 7. items that tap general maladjustment and
166 Chapter 4

feeling bad; consequently, the items cannot High-scoring normals (T scores of 65 or

be separated meaningfully into subscales. higher) are generally described in positive
Scale 7 has substantial item overlap with terms although some of them, particularly
the other clinical scales: only 9 items are women, display neurotic features to some de¬
unique to it. It shares 17 items with Scale 8, gree. High-scoring males are described as
13 items with Scale 2, and from 2 to 8 items being sentimental, responsible, conscien¬
with the other clinical scales. Consequently, tious, verbal, formal, unemotional, and ide¬
Scale 7 can be expected to covary directly alistic. High-scoring females, however, are
with the other clinical scales because of the described as being prone to worry, emo¬
extensive item overlap. Scale 7 shares only 1 tional, high strung, and generally dissatisfied
item with the F scale and none with the other with themselves.
validity scales. Wiener and Harmon (Wiener, Griffith, Upshaw, and Fowler (1958)
1948) were unable to develop subtle and obvi¬ found that high scorers on Scale 7 were more
ous scales for Scale 7 since almost all of the doubtful than high scorers on Scale 9 (Hypo-
items were obvious in nature. The deviant re¬ mania) in their judgments in a psychophysio-
sponse to most of the Scale 7 items is “true” logical weight discrimination task. Griffith
(39 of 48). and Fowler (1960) found that high scorers on
High scorers on Scale 7 (T scores of 65 or Scale 7 were more compliant with an
higher) are usually described as being anx¬ administrator’s request to participate in an
ious, tense, indecisive, and unable to concen¬ experiment than high scorers on Scale 4.
trate. They frequently display obsessive Low scorers on Scale 7 (T scores less
thoughts and ruminations, self-doubt, and than 45) are described in generally positive
associated depressive features. Although spe¬ terms. They are seen as being responsible,
cific phobias or compulsive behaviors may be easygoing, capable, and efficient. They also
seen in high scorers, they are not characteris¬ are seen as relaxed and not anxious or prone
tically seen. In fact, many rigidly compulsive to worry.
clients may not elevate Scale 7 at all, since Test-retest reliability coefficients for
their intellectual defenses are sufficient to Scale 7 tend to be higher for intervals up to
control their anxieties, feelings of insecurity, two weeks than for the other clinical scales,
and so on. with correlations ranging from .74 to .93. For
Basically, high scorers are clients whose intervals up to one year, test-retest correla¬
characteristic defenses of intellectualization, tions are much lower, ranging from .37 to .58
rationalization, and undoing are no longer (Dahlstrom et al., 1975). Test-retest reliability
capable of controlling their anxiety and ten¬ coefficients for Scale 7 on the MMPI-2 for ap¬
sion. High scorers display an extreme con¬ proximately a one-week interval are .89 for
cern over their physical functioning; their men and .88 for women (Butcher et al., 1989).
complaints typically center around the car¬ There is little change in scores on Scale 7
diovascular system, although complaints with increasing age in normal samples (Col-
about gastrointestinal and gastrourinary ligan et al., 1983, 1989), however, scores in
functioning are common. Their physical medical patients (Swenson et al., 1973) and psy¬
complaints generally reflect their high anxiety chiatric patients (Hedlund & Won Cho, 1979)
levels and the effects of anxiety on their phys¬ tend to decrease slightly. Females endorse two
ical functioning. Symptomatic treatment of to three more Scale 7 items than men.
their anxiety is frequently necessary before A summary of the interpretations of
initiating any other form of therapeutic inter¬ four levels of elevations on Scale 7 is pre¬
vention. sented in Table 4-13.
 Clinical Scales 167

TABLE 4-13 Interpretation of Levels of Elevation on Scale 7; Psychasthenia (Pt)

MMPI-2 MMPI
T Score T Score Interpretation

44 and 40 and 1. Low. These clients are secure and comfortable with themselves
below below and are emotionally stable. They are success oriented, persistent,
and capable. There is an absence of worries and a relaxed attitude
toward responsibilities.
45-57 41-59 2. Normal. These clients can handle work and personal responsibili¬
ties without undue worry or anxiety.
58-64 60-69 3. Moderate. These clients are generally punctual in meeting their ob¬
ligations and may worry if unable to do so. They do not see them¬
selves as anxious nor do others see them as anxious. Review of the
content and/or supplemental scales may facilitate interpretation in
this range.
65 and 70 and 4. Marked. These clients are worried, tense, and indecisive. Agitation
above above may develop and overt anxiety is usually apparent both to them¬
selves and to others. At extreme scores (T > 85), there usually are
agitated ruminations and obsessions that no longer control anx¬
iety. Disabling guilt feelings may be present. Psychopharmacologic
treatment of the anxiety may be necessary before other forms of
therapeutic interventions are instituted.

SCALE 8: SCHIZOPHRENIA (Sc) true to false positive cases identified. Each of

the preliminary scales adequately separated
In Scale 8 there are 78 items, which is 25 to
schizophrenics from normals, but they also
125 percent more items than in the other clin¬
identified a number of other diagnostic
ical scales. The items assess a wide variety of
groups as schizophrenic.
content areas, including bizarre thought pro¬
Hathaway and McKinley (Hathaway,
cesses and peculiar perceptions, social alien¬
1956) also attempted to develop scales to
ation, poor familial relationships, difficulties
identify each of the major subtypes of schizo¬
in concentration and impulse control, lack of
phrenia without success. They finally selected
deep interests, disturbing questions of self-
the fourth of the preliminary scales as being
worth and self-identity, and sexual difficul¬
the best, despite its problem of identifying
ties.
other diagnostic groups as schizophrenic.
Scale 8 was developed empirically by
Using the AT-correction procedure (see Chap¬
contrasting the item endorsements of the
ters 1 and 3) reduced the number of false pos¬
original Minnesota normative group with the
itives on this 78-item scale.
responses of two partly overlapping groups
Examples of Scale 8 items with the devi¬
of 50 patients who had been diagnosed as
ant response in parentheses are
schizophrenic (Hathaway, 1956). The crite¬
rion group included assorted subtypes of
“I dislike having people around me.”
schizophrenia and included slightly more
(true)
females (60 percent) than males (40 per¬
cent). “I often feel as if things are not real.”
Hathaway (1956) reported that four sep¬ (true)
arate preliminary schizophrenia scales were “I hear strange things when 1 am
derived in an attempt to improve the ratio of alone.” (true)8
168 Chapter 4

Investigations of the individual items on Scale 8 in the two decades since Comrey re¬
Scale 8 have not been extensive. Comrey’s ported his analysis.
(Comrey & Marggraff, 1958) factor analy¬ Harris and Lingoes (1955) identified
sis of this scale included only 58 items be¬ three subscales within the scale items and di¬
cause of limited computer capacity. (He vided these first two subscales into two and
omitted the 17 items that overlap with Scale 7 three smaller subscales, respectively (see
and 3 additional items.) Comrey identified Table 4-14). Thus, the Object Loss subscale
seven major factors: paranoia, poor concen¬ is divided into Social Alienation and Emo¬
tration, poor physical health, psychotic tend¬ tional Alienation subscales. The Lack of Ego
encies, rejection, withdrawal, and sex con¬ Mastery, Intrapsychic Autonomy subscale is
cern. It is surprising that in spite of the divided into Lack of Ego Mastery, Cognitive;
increased computer capacity now available, Lack of Ego Mastery, Conative; and Lack
no one has factor analyzed all 78 items on of Ego Mastery, Defective Inhibition sub-

TABLE 4-14 Description of High Scorers on the Harris and Lingoes Subscaies
for Scale 8 (Schizophrenia)

Subscale

Abbre¬ Number
Name viation of Items Description of High Scorers

Social Alienation Sc1 21 These clients feel a lack of rapport

with other people; they withdraw
from meaningful relationships with
others.
Emotional Alienation Sc2 1 1 These clients feel a lack of rapport
with themselves; they experience
the self as strange and alien; they
display flattened or distorted affect
and apathy.
Lack of Ego Mastery, Sc3 10 These clients admit autonomous
Cognitive thought processes; they have
strange and puzzling ideas.
Lack of Ego Mastery, Sc4 14 These clients have feelings of ''psy¬
Conative chological weakness"; they show
abulia, inertia, massive inhibition,
and regression.
Lack of Ego Mastery, Sc5 1 1 These clients have feelings of not be¬
Defective Inhibition ing in control of their impulses; they
experience their emotions as strange
and alien; they are at the mercy of
impulse and feeling and show disso¬
ciation of affect.
Bizarre Sensory Experiences Sc6 20 These clients have feelings of change
in the perception of themselves and
their body image; they experience
feelings of depersonalization and es¬
trangement.
 Clinical Scales 169

scales. Since Comrey (Comrey & Marggraff, is added to the raw score on Scale 8 to plot a
1958) omitted 20 items in his factor analysis A-corrected profile, approximately 20 Scale 8
of Scale 5, there is no meaningful way to items endorsed in the deviant direction are
compare his factors with the Harris and sufficient to produce a T score greater than
Lingoes (1955) subscales. 65 when the client has an average score on the
Scale 8 overlaps somewhat with Wiggins’ A scale. Consequently, a client can endorse
(1966) content scales of Psychoticism any combination of 20 or more of the 78
(17/48), Organic Symptoms (13/36), and De¬ items on Scale 8 to obtain a T score greater
pression (9/33). The rather substantial over¬ than 65.
lap with the Organic Symptoms scale and the It also is important to know the specific
small overlap with the Psychoticism scale il¬ combination of items from Scales A and 8
lustrate both the heterogeneity of the Scale 8 that the client is endorsing to produce a spe¬
items and the differences in the processes of cific T score. For example, men can obtain a
rational and empirical item selection. Scale 8 T score of 74 by endorsing any combination
overlaps minimally with the Tryon, Stein, of 40 items from Scales A and 8. A man who
and Chu (Stein, 1968) cluster scales, sharing endorsed 30 A scale items and 10 Scale 8
some items with the Autism and Disruptive items will be very different from a man who
Thought (8/23), Depression (8/28), and Ten¬ endorsed 5 A scale items and 35 Scale 8 items.
sion, Worry, and Fear (7/36) scales. Scale 8 The clinician is strongly encouraged to con¬
also overlaps with the MM PI-2 (Butcher et struct a non-A-corrected profile anytime the
al., 1989) content scales of Bizarre Mentation A scale exceeds a raw score of 18 (T score of
(9/23), Depression (7/33), Health Concerns 56) in order to have an appreciation of the
(6/36), and Family Problems (6/16). relative contributions of the A scale and the
Again, it is apparent that the method of item content of the A-corrected scales in the
item selection produces different items to standard A-corrected profile.
measure similar content areas. Further re¬ Several attempts have been made to de¬
search is needed to establish the empirical velop subscales using item analysis within
correlates of these various groupings of Scale 8 to distinguish between actual cases of
items. schizophrenia and other diagnostic groups
Scale 8 shares a substantial number of (Benarick, Guthrie, & Snyder, 1951; Har¬
items with the other clinical scales, particu¬ ding, Holz, & Kawakami, 1958). These stud¬
larly Scales 6 (13), 7 (17), and 9 (11), and ies, however, have not been successfully rep¬
Scale A(15). It also shares from 3 to 10 items licated (Quay & Rowell, 1955; Rubin, 1954).
with the rest of the clinical scales. Only 16 Virtually none of the items selected by
items are unique to Scale 8. Wiener and Benarick and colleagues (1951) or Harding
Harmon (Wiener, 1948) were unable to de¬ and colleagues (1958) overlap with either
velop subtle and obvious scales within the Wiggins’ (1966) Psychoticism scale or the
scale since most of the items were obvious Tryon, Stein, and Chu (Stein, 1968) Autism
in content. The deviant response to almost and Disruptive Thought scale. This suggests
three-fourths (59/78) of the items is that research is warranted with these latter
“true.” two scales on differentiating schizophrenics
Scale 8 is probably the single most diffi¬ from other diagnostic groups, although the
cult scale to interpret in isolation because of scales are more likely to be sensitive to psy¬
the variety of factors that can result in an ele¬ chotic behaviors in general than to schizo¬
vated score. Since the total number of the A phrenia per se.
scale items endorsed in the deviant direction Newmark, Gentry, Simpson, and Jones
170 Chapter 4

(1978) developed four criteria that were they are lacking something essential to be a
successful in identifying 72 percent of hos¬ real person. They tend to prefer daydreaming
pitalized patients with an admitting diagno¬ and fantasy to interpersonal relationships.
sis of schizophrenia. Their MMP1 criteria They feel isolated, inferior, and self-dissatis¬
follow: fied.
As Scale 8 approaches and exceeds a T
1. Scale 8 is in the T score of 80 to 100, in¬ score of 75, particularly when these T scores
clusive. are the result of a small F-correction, pecu¬
liarities in logic and thinking become more
.
2 The total raw score on Scale 8 with the
apparent or actual schizoid thought processes
K-correction includes 35 percent or
may even be evident. High scorers may ap¬
fewer K items.9
pear confused and disoriented and may exer¬
3. Scale Fis in the T score range of 75 to 95,
cise poor judgment. They frequently display
inclusive.
associated depressive features and psycho¬
.
4 Scale 8 is greater than or equal to Scale motor retardation. All of these behaviors
7. may be the result of a schizophrenic process,
a schizoid adjustment, or severe and pro¬
Newmark and associates found that only longed stress.
5.5 percent of patients in other diagnostic cat¬ Extremely high scorers (T scores greater
egories were labeled schizophrenic (false pos¬ than 100) usually are characterized by severe
itives) using their criteria. They also reported and prolonged stress, accompanied by an
that the Harris and Lingoes subscales for acute decompensation, if the items have been
Scale 8 did not accurately identify the schizo¬ endorsed consistently and accurately (see
phrenic patients. The fact that Newmark and Chapter 3). These persons typically are not
associates had to include both Scales F and 7 schizophrenic; they are more likely to be un¬
to identify correctly 72 percent of their sam¬ dergoing acute psychotic reactions. For ex¬
ple further demonstrates the futility of trying ample, an adolescent going through an iden¬
to diagnose schizophrenia on the basis of tity crisis will frequently score in this extreme
Scale 8 alone. range.
This latter point is supported by Normals who achieve high scores on
Walters’ (1984) finding that Scale 8 produced Scale 8 (T scores of 65 or higher) are de¬
a 61 percent classification accuracy in distin¬ scribed in a variety of terms that seem to re¬
guishing between schizophrenic and schizo¬ flect the changing correlates of the scale as el¬
phrenia-spectrum and general psychiatric pa¬ evations increase. Normals who achieve T
tients. However, Walters (1988) found that scores of 75 or higher, which are the result of
codetypes containing Scale 8 were more likely small F-corrections, are generally described
(64.4 percent) in schizophrenic patients than in a similar manner as high scorers in psychi¬
bipolar disorder, manic patients (35.5 per¬ atric populations (see above).
cent). Anderson and Kunce (1984) found that
High scorers on Scale 8 (T scores of 65 or their university counseling center clients,
higher) are described as cold, apathetic, whose highest clinical scale was Scale 8 (M =
alienated, misunderstood, and having diffi¬ 91), were more difficult than other clients and
culties in thinking and communication, shared a number of characteristics such as
which may reflect an actual psychotic feeling socially isolated, relationship difficul¬
thought disorder. These individuals feel that ties, stressful home life, and so on. However,
 Clinical Scales 171

these clients did not display the severe symp¬ Scale 6- 7-8 Configurations
toms that might be expected with such a high
T score on Scale 8. One configuration of Scales 6, 7, and 8 is seen
Normals with less extreme elevations (T frequently. It consists of Scales 6 and 8 being
scores of 57 to 64) are described as self-dissat¬ above a T score of 80, and Scale 7 being above
isfied, irritable, having wide interests, and a T score of 65 (see Figure 4-6). Scales 6 and <5
immature. They are unlikely to be perceived will be the high-point pair of this configuration.
as being deviant or withdrawn and may be This configuration is sometimes called the
seen as creative, individualistic, and imagina¬ paranoid valley or a psychotic “V.”
tive. They like theoretical and abstract philo¬ Clients exhibiting this configuration are
sophical issues. likely to be emotionally withdrawn, socially
Low scorers (T scores less than 45) are isolated, suspicious, hostile, and lacking in¬
seen as being compliant, submissive, and sight into their own behavior. They also may
overly accepting of authority. They tend to have thought disorders, delusions, and hallu¬
have very practical interests with little con¬ cinations. They usually are labeled as being
cern about theoretical or philosophical is¬ psychotic with the most frequent diagnosis
sues. They also have difficulty understanding being paranoid schizophrenia. (The clinician
persons who approach issues in a theoretical should examine the description of a 6-8/8-6
or philosophical manner. codetype in Chapter 6 for more information
Test-retest reliability coefficients for on this configuration.)
Scale 8 for intervals up to two weeks This configuration probably occurs
range from .74 to .95, and for intervals up most frequently in invalid profiles! It is char¬
to one year from .37 to .64 (Dahlstrom et acteristic of an “all true” response set (see
al., 1975). Test-retest reliability coeffi¬ page 125) and many random response sets
cients for Scale 8 on the MMPI-2 for ap¬ (see page 127) so the clinician needs to exam¬
proximately a one-week interval are .87 for ine measures of the consistency of item en¬
men and .80 for women (Butcher et al., dorsement (see Chapter 3). It also occurs fre¬
1989). quently when clients are overreporting
There are small gender differences on psychopathology; the clinician should exam¬
Scale 8, with females likely to endorse slightly ine measures of the accuracy of item endorse¬
more items than males. Scores tend to de¬ ment (see Chapter 3).
crease 5 to 10 T points with age in psychiat¬ If the clinician has determined that this
ric patients (Hedlund & Won Cho, 1979) 6-7-8 configuration is valid (i.e., measures of
and 2 to 5 T points in medical patients the consistency and accuracy of item endorse¬
(Swenson et al., 1973) and normal indi¬ ment are in the appropriate ranges), Scales 2
viduals (Colligan et al., 1983, 1989). (Depression) and 0 (Social Introversion) may
Scores also decline with chronicity so that be helpful in distinguishing between clients
older patients frequently score in the nor¬ with a thought disorder and clients with a
mal range (Davis, 1972; Wauck, 1950). mood disorder with psychotic features (Post,
Marital status has been reported to be unre¬ Clopton, Keefer, Rosenberg, Blyth, & Stein,
lated to Scale 8 scores (Lacks, Rothenberg, 1986). Walters and Greene (1988) found sim¬
& Unger, 1970). ilar mean differences on Scales 2 and 0 be¬
A summary of the interpretations of five tween groups of inpatients with thought and
levels of elevation on Scale 8 is presented in mood disorders, but they could not identify
Table 4-15. any decision rule to discriminate the individ-
172 Chapter 4

TABLE 4-15 Interpretation of Levels of Elevation on Scale 8: Schizophrenia (Sc)

MMPI-2 MMPI
T Score T Score interpretation

44 and 40 and 1. Low. These clients are conventional, realistic, and uninterested in
below below theoretical or philosophical issues. They are unimaginative and
concrete and may have difficulty with persons who perceive the
world differently.
45-57 45-59 2. Normal. Chronic schizophrenics who have adjusted to their psy¬
chotic process may score in this range. Otherwise, scores in this
range are normal.
58-64 60-69 3. Moderate. These clients think differently from others, though this
may reflect creativity, an avant-garde attitude, or actual schizoid¬
like processes. These clients tend to avoid reality through fantasy
and daydreams. Review of the content and/or supplementary
scales may help to differentiate among these alternatives; the clini¬
cian also should examine what other clinical scales are elevated.
65-90 70-99 4. Marked. These clients feel alienated and remote from their envi¬
ronment, which may reflect an actual schizophrenic process or sit¬
uational or personal distress. Review of the content and/or supple¬
mentary scales may be helpful at the lower end of this range.
Difficulties in logic and concentration and poor judgment become
apparent as scores move higher in this range. As scores approach
a T score of 80, the presence of a thought disorder is likely. Thera¬
peutic interventions should be directive and supportive and fre¬
quently require psychotropic medications.
91 and 100 5. Extreme. These clients are under acute, severe situational stress.
above and A client going through an identity crisis will score in this range.
above These clients typically are not schizophrenic.

ual patients accurately. Consequently, clini¬ mood, psychomotor excitement, and flight of
cians should consider this decision rule to be ideas are covered by the 46 items comprising
tentative. Scale 9 (McKinley & Hathaway, 1944). The
Scales 2 and 0 usually are above a T items range over a wide variety of content
score of 60 in clients with thought disorders areas including overactivity, both behavior-
and below a T score of 55 in clients with a ally and cognitively, grandiosity, egocentric-
manic mood disorder. Wiggins’ (1966) Psy- ity, and irritability.
choticism scale and the MMPI-2 content The criterion group for Scale 9 consisted
scale (Butcher et ah, 1989) of Bizarre Menta¬ of 24 manic patients of moderate or mild sever¬
tion will not distinguish between these two ity since more severe cases would not cooperate
disorders since they are general measures of with testing. The item endorsements of this cri¬
psychoticism. However, Wiggins’ Hypoma- terion group were contrasted with those of the
nia scale will be higher in clients with a manic original Minnesota normative group to develop
mood disorder. Scale 9 empirically, resulting in the 46 items
currently on the scale. The clinician should re¬
member that the label of hypomania refers to
SCALE 9: HYPOMAMIA (Ma)
elevated scores, not T scores below 50 as the
The milder degrees of manic excitement, prefix “hypo” might suggest.
characterized by an elated but unstable A normal activity level is indicated by T
 Clinical Scales 173

FIGURE 4-6 Scales 6 7-8 Configuration:

- Comrey (1958d) found that Scale 9 had
Psychotic "V" the most diversified factor content of any of
the clinical scales and that most of the factor
content was unique to this scale. He identi¬
fied 11 major sources of variance: shyness,
bitterness, acceptance of taboos, poor reality
contact, thrill seeking, social dependency,
psychopathic personality, high water con¬
sumption, hypomania, agitation, and defen¬
siveness.
Harris and Lingoes (1955) identified
four subscales within Scale 9: Amorality,
Psychomotor Acceleration, Imperturbabil¬
ity, and Ego Inflation (see Table 4-16). There
is little overlap between Comrey’s factors and
the Harris and Lingoes subscales. The Psy¬
chomotor Acceleration subscale shares some
items with thrill seeking and agitation, and
Imperturbability shares a few items with shy¬
ness.
Scale 9 also shares few items with any of
Wiggins’ (1966) content scales; even with
Wiggins’ Hypomania scale the overlap is
minimal (7/25). There is even less overlap
with any of the Tryon, Stein, and Ghu (Stein,
6 7 8 1968) cluster scales; Scale 9 has no more than
4 items in common with any of these scales.
Scale Finally, Scale 9 shares few items with the
MMPI-2 content scales (Butcher et al., 1989).
scores in the normal range (45 to 57), and As with Scale 8, it appears that different
more elevated scores indicate increasing lev¬ methods of item grouping produce quite dif¬
els of mania. Hence, elevated scores indicate ferent results, and the correlates of these
hypomania, and increasingly higher scores groups need to be empirically investigated.
reflect mania and ultimately hypermania. Scale 9 has substantial overlap with
Since manic cases are usually readily identi¬ Scale 8 (11 items), but shares only from 1 to 6
fied behaviorally, it is the more moderate items with the rest of the validity and clinical
cases that need to be identified by Scale 9. scales. There are 15 items unique to Scale 9.
Examples of scale items with the deviant Wiener and Harmon (Wiener, 1948) found
response in parentheses are: that the scale had an equal number of obvi¬
ous and subtle items. Again, the preponder¬
“When I get bored I like to stir up some
ance of deviant responses are “true” (35/46),
excitement.” (true)
as with the other scales in the psychotic tet¬
“I am an important person.” (true) rad.
“I don’t blame people for trying to grab Scale 9 also is difficult to interpret in iso¬
everything they can get in this world.” lation. It can be conceptualized as providing
(true)10 energy to activate the qualities identified by
174 Chapter 4

TABLE 4-16 Description of High Scorers on the Harris and Lingoes Subscales
for Scale 9 (Hypomania)

Subscale

Abbre¬ Number
Name viation of Items Description of High Scorers

Amorality Ma t 6 These clients have a callousness about

their own motives and ends and those
of other people; they are disarmingly
frank; they deny guilt.
Psychomotor Acceleration IVId2 1 1 These clients are hyperactive and la¬
bile; they show flight from "inner
life" and anxiety, and pressure for ac¬
tion.
Imperturbability A4 a 3 8 These clients are confident in social
situations; they deny sensitivity; they
proclaim independence from the opin¬
ions of other persons.
Ego Inflation IVId4 9 These clients feel self-important to the
point of unrealistic grandiosity.

other elevated clinical scales. Thus, a client Post and colleagues (1986) found that bi¬
who elevates Scale 4 along with Scale 9 will polar disorder, manic patients had higher
display very different behaviors than a client scores on Scale 9 than other psychotic psychi¬
who simultaneously elevates Scales 8 and 9 atric patients and general psychiatric pa¬
(see the description of these two codetypes in tients. They found that discriminant analysis,
Chapter 6). with Scales2, 9, and das predictors, correctly
Scale 9 is often described as being ele¬ classified 82.5 percent of the patients in their
vated in persons with brain damage (cf. Car- derivation sample and 74.2 percent in the
son, 1969). Although some brain-damaged cross-validation sample. However, Walters
persons may display hyperactive and impul¬ and Greene (1988) found that Scale 9 did not
sive behaviors, virtually any type of emo¬ discriminate between bipolar disorder, manic
tional response may occur in brain-damaged patients and schizophrenic patients. Similar
persons, particularly depression. The MMPI- to Post and associates (1986), they found that
2 is an excellent instrument to assess the re¬ Scale 0 did discriminate between these two
sponse of a person to brain damage, but there groups of patients with schizophrenics earn¬
is not a typical score on Scale 9 or a typical ing higher scores.
clinical profile for brain-damaged persons. High scorers on Scale 9 (T scores of 65 or
Some persons may respond to brain damage higher) are described as being impulsive,
by becoming depressed; others may become competitive, talkative, narcissistic, amoral,
apathetic, withdrawn, and display psychotic extroverted, and superficial in social relation¬
features; still others may be virtually unaf¬ ships. They typically have problems in con¬
fected. Farr and Martin (1988) have reviewed trolling their behavior and display hostile, ir¬
the performance of neuropsychological sam¬ ritable qualities. They are not described as
ples on a number of MMP1 scales and in¬ depressed. They may display actual manic
dexes. features: flight of ideas, lability of mood, de-
 Clinical Scales 175

lusions of grandeur, impulsivity, and hyper¬ tations of high and low scores on Scale 9 must
activity. take into account these normal variations as a
High-scoring normals (T scores of 65 or function of age. Clients whose scores on
higher) are described in generally positive Scale 9 are 15 or more T points higher or
terms: friendly, sociable, energetic, talkative, lower than these expected ranges should be
and enthusiastic. Basically they have a pleas¬ evaluated for the presence of a mood disor¬
ant, outgoing temperament. If scores become der.
elevated over a T score of 75, some other fea¬ A summary of the interpretations of
tures become apparent—hyperactivity, im¬ four levels of elevation of Scale 9 is presented
pulsivity, and irritability. There may even be in Table 4-17.
acting out of conflicts in such persons.
Low scorers (T scores less than 45) also
SCALE 0: SOCIAL INTROVERSION (Si)
are described in positive terms. They are con¬
sidered dependable, reliable, mature, and The 69 items (70 items on the MMPI) in Scale
conscientious. They frequently participate 0 were selected to assess the social introver¬
very little in social activities. sion-extroversion dimension with high scores
Extremely low scores (T scores below 40 reflecting social introversion. The social in¬
in most groups; see below) are described as trovert is uncomfortable in social interactions
apathetic, having little energy, and listless. and typically withdraws from such interac¬
Such persons usually are significantly de¬ tions when possible. This individual may
pressed regardless of their score on Scale 2. have limited social skills or simply prefer to
The possibility of serious depression should be alone or with a small group of friends. The
be considered when Scale 9 is in this range social extrovert is socially outgoing, gregari¬
even if the rest of the profile is within normal ous, and seeks social interactions. Item con¬
limits. In fact, low scores on Scale 9 are gen¬ tent on Scale 0 reflects personal discomfort in
erally better indicators of the presence of sig¬ social situations, isolation, general malad¬
nificant depression than high scores on Scale justment, and self-deprecation.
2. The clinician should evaluate suicide po¬ Scale 0 was not based on a psychiatric
tential in these clients, particularly if the cli¬ syndrome; rather it was developed by using a
ent seems to be getting more energized. psychological test—the Minnesota T-S-E In¬
Test-retest reliability coefficients for ventory (Evans & McConnell, 1941)—to
Scale 9 for intervals up to two weeks range form criterion groups. The Minnesota T-S-E
from .63 to .96; for intervals up to one year Inventory assesses introversion-extroversion
they range from .43 to .64 (Dahlstrom et al., in three areas: thinking (T), social (S), and
1975). Test-retest reliability coefficients for emotional (E).
Scale 9 on the MMPI-2 for approximately a Drake (1946) limited his investigation of
one-week interval are .83 for men and .68 for introversion-extroversion to the social area as
women (Butcher et ah, 1989). assessed by the Minnesota T-S-E Inventory.
There are no gender differences on Scale He selected items for Scale 0 by contrasting
9; in fact, identical T score conversions are groups of students in the guidance program
used for both males and females in the stan¬ at the University of Wisconsin. Two groups
dard MMPI profile. Scores on Scale 9 change were formed: 50 female students scoring
significantly with increasing age. Adolescents above the 65th percentile on the social intro¬
and college students usually score in the T version-extroversion subscale of the Minne¬
score range of 55 to 65, whereas aged persons sota T-S-E Inventory, and 50 female students
score in the range of 40 to 50. Thus, interpre¬ scoring below the 35th percentile. After those
176 Chapter 4

TABLE 4-17 Interpretation of Levels of Elevation on Scale 9: Hypomania (Ma)

MMPI-2 MMPI
T Score T Score Interpretation

44 and 40 and 1. Low. These clients have a low energy and activity level that may
below below reflect situational circumstances such as fatigue or actual depres¬
sion. Extremely low scores (T < 35) indicate depression irrespec¬
tive of the elevation of Scale 2. Normal, aged persons score in the
upper end of this range.
45-57 41-59 2. Normal. Normal college students and adolescents score in the up¬
per end of this range. The client has a normal activity level.
58-64 60-69 3. Moderate. These clients are active, outgoing, and energetic. Exter¬
nal restrictions on their activity level may result in agitation and
overtly expressed dissatisfaction. Review of the content and/or sup¬
plementary scales may facilitate interpretation in this range.
65 and 70 and 4. Marked. These clients are overactive, emotionally labile, and may
above above experience flight of ideas. Although the client's mood is typically
euphoric, outbursts of temper may occur. These clients are impul¬
sive and may have an inability to delay gratification. Manic fea¬
tures become increasingly pronounced with more elevated scores.
The narcissistic and grandiose features also become more appar¬
ent. Review of the content and/or supplementary scales may facil¬
itate interpretation at the lower end of this range.

items were eliminated that had a very high or (1955) did not attempt to create subscales for
very low frequency of endorsement in either this scale. Graham and associates (1971)
or both groups, 70 items were selected that identified six factors among Scale 0 items: In¬
discriminated these two groups. feriority and Discomfort, Affiliation, Social
Drake (1946) later tested male students Excitement, Sensitivity, Interpersonal Trust,
on Scale 0 and computed separate norms for and Physical-Somatic Concerns. Serkownek
males and females. The distributions of the (1975) used the results of Graham and
total raw score for these two groups were so associates’ factor analysis to create six sub¬
similar that Drake combined them into a single scales for Scale 0 (see Table 4-18). Williams
group to establish the norms on the MMPI. (1983) found that Serkownek’s subscales were
Examples of Scale 0 items with the devi¬ in the acceptable ranges of reliability in col¬
ant response in parentheses are: lege students and provided preliminary inter¬
pretive information for males and females.
“At parties I am more likely to sit by Clinicians can continue to use the Serkownek
myself or with just one other person than (1975) subscales on the MMPI-2 if desired
to join in with the crowd.” (true) since only one item was dropped from Scale
“Whenever possible I avoid being in a 0.
crowd.” (true) Ben-Porath, Hostetler, Butcher, and
“If given the chance I would make a Graham (1989) developed new content-ho¬
good leader of people.” (false)11 mogenous subscales for Scale 0 as an alterna¬
tive to the Serkownek (1975) subscales. Ben-
Comrey (1957a) did not factor analyze Porath and colleagues identified three
the items on Scale 0, and Harris and Lingoes subscales in college students: Shyness/Self-
 Clinical Scales 177

TABLE 4-18 Description of High Scorers on the Serkownek Subscales

for Scale 0 (Social Introversion)

Subscale

Abbre¬ Number
Name viation of Items Description of High Scorers

Inferiority-Personal Sii 27 These clients are unhappy and uncomfort¬

Discomfort able because of their perceived lack of in¬
terpersonal skills.
Discomfort with Others S/2 13 These clients do not enjoy being with other
people and have problems being asser¬
tive.
Staid-Personal Rigidity S/3 16 These clients participate in few active so¬
cial groups.
Hypersensitivity S/4 10 These clients are sensitive to the reactions
of others; they are shy, easily embar¬
rassed, and generally anxious; they are
likely to avoid facing up to stressful situa¬
tions.
Distrust S/5 1 2 These clients feel that others are dishon¬
est, insincere, selfish, and generally anx¬
ious.
Physical-Somatic S/6 10 These clients admit somatic ailments and
concern about their physical appearance.

Consciousness, Social Avoidance, and only 1 item with Scale 0. Wiener and Harmon
Self/Other Alienation (see Table 4-19). (Wiener, 1948) did not develop subtle and ob-
These three subscales accounted for nearly 90 vious scales for this scale. Christian and col¬
percent of the variance in Scale 0 scores. leagues (1978) found that students rated Scale
Scale 0 overlaps significantly with 0 items as being neutral, neither subtle nor
Wiggins’ (1966) Social Maladjustment scale obvious. There are almost exactly the same
(21/27), which Wiggins characterized as cor¬ number of “true” (34) as “false” (35) devi¬
responding roughly to an introversion-extro¬ ant responses to Scale 0 items.
version dimension. Scale 0 also has substan¬ High scorers on Scale 0 (T scores of 65 or
tial item overlap with the Tryon, Stein, and higher) in both psychiatric and normal popu¬
Chu (Stein, 1968) Social Introversion scale lations are described similarly. They are seen
(20/26). Regardless of the method of item as socially introverted, shy, and withdrawn.
grouping, a dimension of social introversion- More extreme groups (T scores greater than
extroversion appears to permeate the MMPI 70 in most groups; see below) are described as
items. socially inept, aloof, self-deprecating, and
Scale 0 shares fewer items proportionally anxious in their interactions with others.
with other validity and clinical scales than Scale 0 scores tend to be unrelated to psycho¬
any other scale. There are 26 items unique to pathology since elevations may reflect a
Scale 0. Scales 2 and 7 share 7 and 8 items, schizoid withdrawal from interpersonal rela¬
respectively, with Scale 0. Most scales share tionships, neurotic withdrawal, and self-dep-
178 Chapter 4

TABLE 4-19 Description of High Scorers on the Ben-Porath et at. Subscales

for Scale 0 (Social Introversion)

Subscale

Abbre¬ Number
Name viation of Items Description of High Scorers

Shyness/Self- Sh 14 These clients are shy around others and

Consciousness easily embarrassed. They are uncomfort¬
able in social and new situations, and
avoid contact with people.
Social Avoidance Si2 8 These clients dislike and avoid group
activities of all types, and act to keep
people at a distance.
Self/Other Alienation Si3 17 These clients have low self-esteem and
lack self-confidence. They question their
own judgment. They are nervous, fearful,
and indecisive. They lack interest in
things.

recation as a function of personal distress, or liability coefficients for Scale 0 on the

merely an introverted orientation. In other MMPI-2 for approximately a one-week inter¬
words, the exact interpretation of Scale 0 will val are .92 for men and .91 for women
depend on the client’s situation and which (Butcher et al., 1989).
other clinical scales are elevated. Scores on Scale 0 increase with age. Ad¬
Low scorers (T scores less than 45) are olescents and college students usually achieve
essentially described as being extroverted. T scores in the range of 40 to 50, while aged
They are sociable, outgoing, and versatile in persons score from 50 to 60. Scale 0 appears
their interactions with others. They partici¬ to have an important role in marital relation¬
pate in many social activities. They may be ships. Couples with T scores on Scale 0 that
unable to delay gratification and be un¬ differ by 20 points or more will frequently re¬
dercontrolled emotionally. Extremely low port marital conflicts over their social rela¬
scorers (T scores below 35) are described as tionships. One partner prefers to be alone or
being flighty, superficial in their relationships with small groups of friends while the other
with others, and lacking any real intimacy. prefers larger social functions. This differ¬
These characteristics are especially likely if ence in social orientation may have appealed
the client simultaneously elevates Scale 3 or to them initially, but it can become a source
4. of marital conflict.
Test-retest reliability coefficients for Elevation of Scale 0 typically suppresses
Scale 0 on the MMPI usually are higher than the acting out seen with elevations of Scales 4
for the other clinical scales. Correlations for and 9, whereas it may accentuate the rumi¬
intervals up to two weeks range from .80 to nating behaviors seen with elevations on
.96 and for intervals up to one year from .54 Scales 2 or 7 and especially Scale 8.
to .76 (Dahlstrom et al., 1975). Test-retest re¬ Linear T scores are used on Scale 0 for
 Clinical Scales 179

TABLE 4-20 Interpretation of Levels of Elevation on Scale 0: Social Introversion (Si)

MMPI-2 MMPI
T Score T Score Interpretation

44 and 40 and 1. Low. These clients are socially extroverted, gregarious, and so-
below below daily poised. A person with an extremely low score (T < 35) may
have very superficial social relationships without any real depth. Ad¬
olescents and college students normally score at the upper end of this
range. The clinician should be very cautious of labeling these clients
as being schizophrenic.
45-57 41-59 2. Normal. These clients report a balance between socially extro¬
verted and introverted attitudes and behaviors.
58-64 60-69 3. Moderate. These clients prefer to be alone or with a smali group
of friends. They have the ability to interact with others, but gener¬
ally prefer not to. Review of the content and/or supplementary
scales may facilitate interpretation in this range.
65 and 70 and 4. Marked. These clients are introverted, shy, and socially insecure.
above above In addition, they withdraw from and avoid significant others,
which serves to exacerbate their problems since others might be
able to help them. The likelihood of acting out is decreased and
ruminative behavior is increased. Intervention should specifically ad¬
dress the client's tendency to withdraw. Review of the content and/
or supplementary scales may facilitate interpretation at the Sower
end of this range.

the MM PI-2 as was the case on the MMPI. T by the University of Minnesota. Published by the
scores on Scale 0 of the MMPI seldom get University of Minnesota Press. All rights reserved.
very high or very low. It is one of the few 2. Ibid.
3. Ibid.
MMPI scales on which T scores closely repre¬
4. Ibid.
sent the frequency with which a score will be
5. Ibid.
obtained (i.e., a T score of 65 or higher will
6. Ibid.
occur two to three times in 100 cases). Conse¬ 7. Ibid.
quently, clinicians can begin to interpret 8. Ibid.
Scale 0 on the MMPI when T scores reach 65 9. This second criterion can be stated more
or higher and 35 and lower. simply as follows: The raw score on Scale 8 with¬
A summary of the interpretations of out the /©correction should be twice the raw score
four levels of elevation of Scale 0 is given in on the K scale.
Table 4-20. 10. Reproduced from the MMPI-2 by per¬
mission. Copyright © 1943, (renewed 1970), 1989
by the University of Minnesota. Published by the
ENDNOTES
University of Minnesota Press. All rights reserved.
1. Reproduced from the MMPI-2 by per¬ 11. Ibid.
mission. Copyright © 1943, (renewed 1970), 1989
'
 CHAPTER 5

Supplementary Scales,
Content Scales, Critical Items,
and Short Forms

The popularity of the MMPI quickly led to tional sources of information about the client
the proliferation of additional means of gar¬ must be sacrificed if only the first 370 items
nering information from the item pool. The are administered.
apparent ease with which investigators can Next, certain items within the MMPI
identify a criterion group and contrast their have been identified as “critical” or “stop”
responses to all 550 items on the MMPI with items because by endorsing any one of these
a normal group has resulted in more than 450 items, the client is acknowledging the exis¬
supplementary scales (Dahlstrom, Welsh, & tence of behavior or psychopathology that
Dahlstrom, 1975). may demand immediate attention.
A large number of these MMPI scales This chapter will review various lists of
are too limited or specialized for widespread critical items and the problems of identifying
use and few of them are cross-validated. Con¬ what and how many items are actually “criti¬
sequently, only some of the more widely used cal.” Finally, this chapter will review the
supplementary scales will be considered here. MMPI short forms (i.e., abbreviated versions
It is necessary to administer all 567 MMPI-2 of the MMPI that attempt to predict the
items if all the supplementary and content scores on the validity and clinical scales from
scales and critical items are to be evaluated. a limited number of items), and their use in
The client can stop after answering the first clinical practice.
370 items of the MMPI-2 (see Chapter 2) and
the clinician can still score the standard valid¬
SUPPLEMENTARY SCALES
ity and clinical scales and plot the standard
profile. Before undertaking the development of a new
The clinician must realize, however, that supplementary scale on the MMPI-2, the cli¬
most of the supplementary and content scales nician is strongly encouraged to review
and critical item lists that will be discussed in Clopton’s (1974, 1978b, 1979a, 1982; Clop-
this chapter also require scoring many of the ton & Neuringer, 1977a; Levitt, 1978) de¬
last 200 items. Consequently, these addi¬ scription of the methodology for the

181
182 Chapter 5

development of supplementary scales and and the Wiener and Harmon (Wiener, 1948)
Butcher and Tellegen’s (1978) description of Depression-Subtle subscale. Colligan, Os¬
the common methodologic problems in borne, Swenson, and Offord (1989) have pro¬
MMP1 research. vided percentile ranks from their contempo¬
The associated issue of whether the sup¬ rary normative sample by gender for many of
plementary scales generate redundant infor¬ the scales reviewed in this chapter.
mation already provided by the standard va¬ Templates for hand scoring many of the
lidity and clinical scales has not been MMPI-2 and MMPI supplementary scales
investigated widely. Clopton and Klein are available from National Computer Sys¬
(1978) found that three supplementary scales tems (P.O. Box 1416, Minneapolis MN
(Prejudice, Ego Strength, and MacAndrew 55440, 800-627-7271); Psychological Assess¬
Alcoholism) were highly related to the scores ment Resources (P.O. Box 998, Odessa LL
on the standard validity and clinical scales. 33556, 800-331-8378) provides templates
They found, however, that a client’s individ¬ only for the MMPI.
ual scores on these three scales could not be
predicted accurately. This issue of the possi¬
ble redundancy of the supplementary scales TRADITIONAL SUPPLEMENTARY
with the standard validity and clinical scales SCALES
needs further investigation.
The interest in supplementary scales of
Welsh Anxiety (A) and Repression
the MMPI has increased in the last several
(R ) Scales
years as evidenced by two recent texts (Cald¬
well, 1988; Levitt, 1989). Caldwell has pro¬ Lactor-analytic studies of the MMPI clinical
vided interpretive information on 104 differ¬ scales have consistently identified two factors
ent supplementary scales and Levitt on 68. that are variously labeled and interpreted.
Both Caldwell and Levitt reviewed the Harris Welsh (1956) constructed his Anxiety {A) and
and Lingoes (1955) subscales, Wiggins’ (1966) Repression (R) scales to measure these two
Content scales, and the Tryon, Stein, and Chu factors. The first factor has been interpreted
Cluster scales (Stein, 1968). The interpretive in¬ in two different ways, and investigators de¬
formation in both books reflects the authors’ bated for nearly ten years over which inter¬
clinical experience in using the various scales pretation was more appropriate.
with little empirical data presented. Welsh found that this factor has high
Levitt (1989) also included a list of sup¬ positive loadings on Scales 7 (Psychasthenia)
plementary scales that are not recommended and 8 (Schizophrenia) and high negative
for clinical use (pp. 113-114). The rationale loadings on the K scale. This factor has been
for including a specific scale in this list was identified as reflecting a personality factor la¬
not made explicit other than a general state¬ beled anxiety (Welsh, 1956), lack of ego resil¬
ment about lack of experimental evaluation. iency (Block, 1965), and general maladjust¬
No doubt some clinicians will be concerned ment (Tyler, 1951). This factor also has been
that their favorite scale is on Levitt’s “hit” identified as reflecting a response bias factor
list. Levitt also recommended the use of sev¬ labeled social desirability (Edwards & Diers,
eral scales in his text that would seem to re¬ 1962) or the deviation hypothesis (Berg,
flect the same lack of experimental validation 1955, 1957). In the deviation hypothesis, the
such as the Tryon, Stein, and Chu Cluster actual item content and personality charac¬
scales (Stein, 1968), Over-controlled Hostil¬ teristics are deemed to be unimportant since a
ity (Megargee, Cook, & Mendelsohn, 1967), general response tendency is the primary de-
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 183

terminant of whether or not the client en¬ trasted the responses of these extreme groups
dorses a specific item. on all 550 MMPI items. The A scale consists
After a decade of debate, Block (1965) of 39 items that showed at least a 75 percent
demonstrated that the factor structure of the separation between high and low scores in
MMPI was virtually unchanged when the po¬ two separate VA samples. The R scale con¬
tential of response bias, particularly social sists of 37 items (40 items on the MMPI) that
desirability and acquiescence, was con¬ showed at least a 60 percent separation of
trolled. Although Jackson (1967) questioned high and low scorers in the same two samples.
Block’s research on several methodologic is¬ The items on the A and R scales are provided
sues, Block’s findings convinced most inves¬ in Appendix A for the MM PI-2 and Appen¬
tigators that the first factor should be inter¬ dix D for the MMPI.
preted as a personality factor rather than a The major content areas represented in
response bias factor. More recently, Shweder the A scale are:
(1977a, 1977b) renewed the debate when he
posited that the first factor reflects a concep¬ 1. Problems in thinking and thought pro¬
tual linkage among test items that is not a cesses
personality factor. 2. Negative emotional tone and dysphoria
The reader is encouraged to examine
3. Lack of energy and pessimism
Shweder’s (1977a) proposed interpretation,
Block’s (1977) and Edwards’ (1977) response 4. Personal sensitivity
to Shweder, and Shweder’s (1977b) reply to 5. Deviant thought processes
Block and Edwards for further information
on this heated debate. Dahlstrom and col¬ In the R scale the major content areas
leagues (1975) provided an in-depth analysis are the denial of or constriction of interests
of the major sources of variance in the in:
MMPI, which is also pertinent to this debate.
Most MMPI researchers concur with Block’s 1. Health and physical symptoms
position; however, the informed reader can 2. Emotionality, violence, and activity
decide for himself or herself which interpre¬
3. Enjoyable reactions to others in a social
tation of the first factor of the MMPI seems
situation
most appropriate.
Unlike the first factor, the second factor 4. Social dominance, feelings of personal
of the MMPI has not involved controversy. adequacy, and interest in personal ap¬
Through factor analysis, Welsh (1956) found pearance
that the second factor has moderate positive 5. Interest in personal and vocational pur¬
loadings on Scales 2 (Depression), 3 (Hyste¬ suits
ria), 5 (Masculinity-Femininity), and 6 (Para¬
noia), and a moderate negative loading on High A scores (T scores of 65 or higher)
Scale 9 (Hypomania). He identified it as re¬ are related to behaviors in which anxiety is
flecting a personality factor, which he labeled prominent, whereas high R scores (T scores
repression. of 65 or higher) are characterized by repres¬
To develop the A and R scales, Welsh se¬ sion and denial. The anxiety being assessed
lected male Veterans Administration (VA) by the A scale has been found to reflect situa¬
patients who scored at or beyond the upper tional anxiety rather than long-term
and lower 10 percent of a preliminary scale characterologic anxiety, which is assessed by
designed to assess each factor; he then con¬ Scale 7 (Psychasthenia). The deviant re-
184 Chapter 5

sponse is “true” for all but one of the A scale although these behaviors are not upsetting to
items and “false” for all of the R scale items, them. They also are described as verbally flu¬
which means that either a “true” or “false” ent and competent in social situations.
response set will substantially affect scores on A summary of the interpretations of
both of these scales. TRIN (True Response four levels of elevation of the A scale is pro¬
Inconsistency scale) scores above 9 would be vided in Table 5-1.
expected with high A scores and below 9 with High scorers on the R scale (T scores of
high R scores (see Chapter 3, pages 74-76). 65 or higher) are seen as being unwilling to
High scorers on the A scale (T scores of discuss their problems, which may reflect
65 or higher) are described as anxious, lack¬ conscious suppression and constriction of in¬
ing confidence in their own abilities, inhib¬ terests in events around them or actual re¬
ited, and overcontrolled. They are character¬ pression and denial. In the latter situation
ized as reacting to situational stress or personal they also will typically elevate the AT scale and
distress with anxiety. These persons are often Scale 3 (Hysteria), which would substantiate
seen as being generally maladjusted, and the el¬ their repressive nature. These persons appear
evation of the A scale reflects their discom¬ constricted and overcontrolled and lack in¬
fort. Because of this discomfort, they are sight into their own behavior. They are un¬
usually motivated to enter into psychological willing to discuss any form of psychopathol¬
treatment. ogy even though it may be apparent to
Low scorers on the A scale (T scores less everyone but themselves.
than 45) are usually described as well ad¬ Low scorers on the R scale (T scores less
justed and not overtly anxious. They may be than 45) are described as being able to discuss
impulsive and display behavioral problems, what problems they may perceive themselves

TABLE 5-1 Interpretation of A Scale Elevations

MMPI-2 MMPI
T Score T Score Interpretation

44 and 40 and 1. Low. These clients are not overtly anxious. They are extro-
below below verted, verbally fluent, and confident in their own abilities.
They may be impulsive. Such low scores in clients in clinical
settings would not be expected. Be sure that a score in this
range is appropriate for the client.
45-57 41-59 2. Normal. These clients have a normal amount of anxiety.
58-64 60-69 3. Moderate. These clients are reporting a significant degree of
anxiety and distress that should be readily apparent. They
usually are motivated to seek treatment because of their level
of distress.
65 and 70 and 4. Marked. These clients are reporting an ever increasing level
above above of anxiety and distress that may be totally debilitating. They
frequently are seen as maladjusted and emotionally upset,
which may reflect a situational crisis or a more chronic prob¬
lem. Extreme scores (T scores of 80 and higher) frequently are
produced by the client overreporting psychopathology (see
Chapter 3).
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 185

as having. They tend to be socially extro¬ men (Swenson, Pearson, & Osborne, 1973).
verted and outgoing in their relationships Since T scores for the A and R scales are pro¬
with others. vided by gender (see Appendix B), these gen¬
A summary of the interpretations of der differences will not affect the interpreta¬
three levels of elevation of the R scale is given tion of these scales.
in Table 5-2. The A scale has substantial item overlap
Welsh (1965) provided a summary of in¬ with a number of the standard clinical scales:
terpretations of the A and R scales when both Scale 2 (Depression)—6 items; Scale 7
scales are employed conjointly. (This sum¬ (Psychasthenia)—13 items; Scale 8 (Schizo¬
mary table also appears in Dahlstrom, phrenia)—8 items; and Scale 0 (Social Intro¬
Welsh, and Dahlstrom [1972], pp. 238-239.) version)—10 items (see Table 5-3). The A
Welsh cautioned clinicians against making scale also has substantial item overlap with
cookbook interpretations of these descrip¬ several of the other supplementary scales: Mt
tions; rather he suggested that the interpreta¬ (College Maladjustment)—12 items; PK
tions be considered as tentative hypotheses (Post Traumatic Stress Disorder—Keane)—9
for further investigation. Duckworth and items; and PS (Post Traumatic Stress Disor¬
Anderson (1986) reported that Welsh’s sys¬ der—Schlenger)—14 items (see Table 5-4).
tem was not accurate for college students ex¬ All of these scales that share items with
cept for the high A and high R description. the A scale can be conceptualized as “first
They furnished a more limited summary of factor” scales (i.e., general measures of mal¬
the joint elevations of the A and R scales, adjustment and emotional distress). Since
which they found useful in college students. these scales are redundant measures of the
The clinician who is interested in using the A first factor, it is not clear what additional in¬
and R scales in profile interpretation should formation is gained by scoring all of these
consult both of these sources. supplementary scales beyond the A scale. Re¬
Age does not appear to affect perfor¬ search that demonstrates the unique variance
mance on the A and R scales; however, accounted for by each of these first factor
women tend to endorse a few more items than scales is needed if all of these scales are to be

TABLE 5-2 Interpretations of R Scale Elevations

MMPI-2 MMPI
T Score T Score Interpretation

44 and 40 and 1. Low. These clients either do not have problems or are willing
below below to discuss the problems that they perceive themselves as
having.
45-57 41-59 2. Normal. These clients show an appropriate level of willing¬
ness to discuss their behavior and problems.
58 and 60 and 3. Moderate and Marked. These clients are unwilling to discuss
above above their behavior and any problems they may have. They may
be merely suppressing this material or repressing and denying
that any problems exist. Scales K and 3 (Hysteria) are typi¬
cally elevated in this latter situation. These clients typically
lack insight into their own behavior.
186 Chapter 5

TABLE 5-3 Item Overlap between the MMPI-2 Supplementary Scales and the Standard
Validity and Clinical Scales

A R Es MAC-R FB OH Do Re Mt GM GF PK PS
S 0 S 0 ci 0 S 0 S 0 S 0 S 0 S 0 S 0 S 0 S 0 S 0 S 0

L 0 0 0 0 0 0 1 0 0 0 3 0 0 1 1 0 0 1 0 0 2 0 0 1 0 0
F 0 0 2 1 0 3 5 0 0 0 0 0 0 0 0 1 0 0 0 0 0 2 5 0 4 0
K 0 5 4 0 1 1 0 0 0 0 3 0 1 0 1 0 1 3 0 0 0 0 0 3 0 2
Hs 0 0 3 0 0 7 1 1 0 0 1 0 0 0 1 0 6 0 0 5 0 0 5 0 5 0
D 6 0 10 0 1 10 0 4 0 0 1 2 0 4 1 0 16 0 0 4 2 0 1 1 2 1 1 2
Hy 2 2 4 0 1 10 1 4 0 0 4 0 1 2 3 0 9 2 0 5 0 1 9 1 10 2
Pd 3 1 0 0 1 5 7 1 0 0 2 2 2 4 0 5 5 0 1 2 1 4 1 1 0 9 0
Mf 1 0 5 3 2 5 2 5 0 0 3 1 1 1 4 1 0 0 0 9 16 0 1 0 0 0
Pa 1 0 2 0 0 3 3 0 2 0 2 1 1 0 0 3 2 2 1 2 0 1 7 0 6 0
Pt 13 0 0 0 0 9 1 2 2 0 1 1 0 6 0 1 14 0 1 4 0 0 17 0 17 0
Sc 8 0 1 2 1 9 2 5 10 0 0 0 0 2 0 2 9 0 0 2 0 1 19 0 27 0
Ma 0 1 0 2 2 3 5 0 0 0 0 3 2 3 0 3 3 1 1 1 0 5 3 0 5 0
Si 10 1 8 0 2 5 0 5 0 0 1 2 0 5 3 1 6 0 0 8 2 1 7 2 4 2

Note: "S" indicates that the deviant response for the item is the same on both scales; "0" indicates that the
deviant response is "true" on one scale and "false" on the other scale.

scored routinely. Nine A scale items also are 6 items with the MAC-R (MacAndrew Alco¬
scored on the Es (Ego Strength) scale, but holism Scale—Revised), and all of these
with the opposite response as deviant. items are scored in the opposite direction on
The R scale shares fewer items with the two scales.
other scales than the A scale. The R scale has
10 items in common with Scale 2 (Depres¬
Ego Strength (Es) Scale
sion), 5 items with Scale 5 (Masculinity-Fem¬
ininity), and 8 items with Scale 0 (Social In¬ Barron (1953) developed the Ego Strength
troversion). The R scale has even less overlap (Es) scale by correlating the MMPI item re¬
with the other supplementary scales. It shares sponses of 33 neurotic clients at two separate

TABLE 5-4 Item Overlap among the MMPI-2 Supplementary Scales

A R Es MAC-R FB OH Do Re Mt GM GF PK
SO SO so so so so so so so so so so
R 0 0
Es 0 9 1 3
MAC-R 1 2 0 6 0 4
Fb 1 0 0 0 1 0 2 1
OH 0 2 2 0 0 3 1 2 0 0
Do 0 5 0 0 4 0 2 6 0 0 1 0
Re 0 0 4 1 2 1 0 9 2 0 2 1 5 0
Mt 12 0 1 1 0 6 1 3 0 0 0 1 0 4 0 0
GM 0 3 0 2 7 1 3 0 0 1 0 2 2 0 3 1 1 4
GF 0 0 3 4 0 3 4 5 0 0 2 1 2 1 6 1 1 0 0 0
PK 9 0 0 2 0 10 3 1 2 0 1 0 0 3 0 1 10 0 0 1 0 0
PS 14 0 1 2 0 10 1 3 4 0 2 2 1 2 0 2 13 0 0 3 0 0 26 0

Note: "S" indicates that the deviant response for the item is the same on both scales; "0" indicates that the
deviant response is "true" on one scale and "false" on the other scale.
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 187

test administrations. Each client completed pected to have relatively normal profiles be¬
all 550 items on the MMPI before psycho¬ cause of the inverse relationship between the
therapy was initiated and then again after six Es scale and the clinical scales. Their normal
months of psychotherapy. The clients were profile would suggest that either they have no
divided into two groups: 17 of them were problems or they are extremely unwilling to
judged to have clearly improved after six admit any problems; yet they are requesting
months of psychotherapy, and the other 16 psychological treatment. Psychological inter¬
were judged to be unimproved after the same vention may be unnecessary in the former in¬
interval. stance and very difficult in the latter because
Barron identified 68 items (only 52 items of their defensiveness and resistance. Clients
remain on the MMPI-2) that significantly with high Es scale scores and an elevated pro¬
correlated with rated improvement in these file may be indicating that they have the ap¬
33 clients. He concluded that these items propriate resources for dealing with their
(see Appendix A for the MMPI-2 items) mea¬ problems, which probably are of recent ori¬
sured a general factor of capacity for person¬ gin. As a consequence of these different pro¬
ality integration, or ego strength—hence files that could be obtained with high Es
the name of the scale. He believed that the scores, sampling differences could produce
Es scale assessed the latent ego strength of the variety of relationships to psychotherapy
the person and that this would be an impor¬ outcome described above.
tant determinant of response to psychother¬ The plethora of studies reporting no rela¬
apy. tionship between theEs scale and psychother¬
Although all the clients were seeking apy outcome indicates that the Es scale is of
psychotherapy for some sort of psychological little usefulness in routinely predicting the re¬
difficulty, a high Es score, which indicates sponse of a given client to psychotherapy. If
good ego strength, suggested that the person further research on the Es scale is deemed
had resources that would emerge as therapy necessary, it should try to identify those sub¬
progressed. Since Es has a moderate negative groups of clients in whom positive or negative
correlation with all validity and clinical scales psychotherapy outcome is expected.
except one (it has a moderate positive correla¬ The contradictory nature of the research
tion with the K scale), Barron hypothesized that on the Es scale makes it inappropriate to de¬
clients with high Es scores were more likely to scribe correlates of high and low scores on
be facing situational stresses while clients with this scale.
low Es scores were more likely to be experienc¬ The Es scale has substantial item overlap
ing chronic, characterologic problems. with a number of the standard clinical scales
Probably no other supplementary scale with the items being scored in the opposite di¬
has generated as much research with contra¬ rection on the Es scale: Scale 1 (Hypochon¬
dictory results as the Es scale; various studies driasis)—7 items; Scale 2 (Depression)—10
have reported positive, no, and inverse rela¬ items; Scale 3 (Hysteria)—10 items; Scale 7
tionships between the Es scale and outcome (Psychasthenia)—9 items; and Scale 8
in psychotherapy. Dahlstrom and associates (Schizophrenia)—9 items (see Table 5-3).
(1975) provided a comprehensive review of The Es scale also has substantial item
these studies. They also pointed out the para¬ overlap with several of the other supplemen¬
doxical relationship between psychotherapy tary scales (see Table 5-4), and again the
outcome and Es scores as reported by Barron items are scored in the opposite direction: A
(1953). (Anxiety)—9 items; Mt (College Maladjust¬
Clients with high Es scores would be ex¬ ment)—6 items; PK (Post Traumatic Stress
188 Chapter 5

Disorder—Keane)—10 items; and PS (Post Although MacAndrew (1965) developed

Traumatic Stress Disorder—Schlenger)—10 the MAC scale on men and intended it to be
items. used only with men, researchers quickly ex¬
Thus, the Es scale could be conceptual¬ tended its use to women (cf. Rich & Davis,
ized as a negative marker for the first factor 1969; Schwartz & Graham, 1979). In fact,
rather than as a measure of ego strength (i.e., popular usage of the MA C scale does not rec¬
high scorers on the Es scale are saying that ognize that the scale was developed originally
they are not generally maladjusted or emo¬ for men; consequently the use of the MAC
tionally distressed, which would seem to scale in men and women will be reviewed
make them poor candidates for psychother¬ here.
apy). Recently MacAndrew has developed
substance abuse scales specifically for use
with young men (Substance Abuse Proclivity
MacAndrew Alcoholism [MAC) Scale
scale [SAP]: MacAndrew, 1986) and women
Several early investigators (Hampton, 1953; (MacAndrew, 1988). Research on these two
Holmes, 1953; Hoyt & Sedlacek, 1958) at¬ scales has been limited since they were only
tempted to develop supplementary MMPI recently developed. MacAndrew (1987) re¬
scales to detect alcoholism with limited suc¬ ported that the SAP scale was negatively cor¬
cess (cf. Clopton, 1978a; MacAndrew & related with age, which suggested that it was
Geertsma, 1964). MacAndrew (1965) was “tapping something more fundamental than
more successful in developing his Alcoholism the accumulated consequences of chronic
(MAC) scale. He selected items that differen¬ substance abuse” (p. 145).
tiated alcoholic outpatients from nonalco¬ MacAndrew also found that the detec¬
holic, psychiatric outpatients. All patients tion rates with the SAP and MAC scales were
were male, and most were white. MacAndrew almost identical in these young adults. The
identified 51 MMPI items that reliably sepa¬ very similar detection rates for the two scales
rated these groups. Two of these 51 MMPI are somewhat surprising since they share only
items (215 and 460) actually refer to alcohol nine items in common. If other studies reveal
use, and most investigators eliminate these that the SAP and MAC produce similar re¬
two items, leaving a total of 49 items on the sults, it would appear unnecessary to score
MAC scale.1 both scales.
In the original sample, a raw score of 24 Probably no single MMPI scale has gen¬
items or more to identify a patient as alco¬ erated more research than the MAC scale,
holic correctly classified 81.8 percent of the and numerous reviews are available (cf.
patients. In a cross-validation sample this Apfeldorf, 1978; Gottesman & Prescott,
same cutting score correctly classified 81.5 1989; Greene & Garvin, 1988; MacAndrew,
percent of the patients, which is an unusually 1981; Megargee, 1985). In the interests of
small loss of accuracy in classification. brevity only a summary of the prolific re¬
Through a factor analysis of the MAC scale, search on the MAC scale will be provided
Schwartz and Graham (1979) found that the here. This summary will draw heavily on the
scale taps several discrete dimensions of per¬ review by Greene and Garvin (1988). The in¬
sonality and behavior, including impulsivity, terested reader should consult that review or
high energy levels, interpersonal shallowness, any of the other reviews cited above for more
and general psychological maladjustment, specific information and references on the
but not general antisociality. MAC scale. Table 5-5 provides a summary of
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 189

TABLE 5-5 Summary Table of the Performance on the MacAndrew Alcoholism Scale (MAC)
as a Function of Age, Gender, and Ethnicity

MAC
(Raw Score)
False Hit
Sample N Age M SD Positives Rate

Normal Individuals
White
Male adults 2975 34.5 23.1 4.3 26.5 76.2
Male adolescents 352 17.0 22.7 3.7 1 9.2 77.8
Female adults 1460 40.6 20.3 3.8 19.0 80.4
Female adolescents 213 15.8 20.9 4.0 7.4 86.6
Black
Male adults 19 23.3 21.3 3.3 — —

Psychiatric Patients
White
Male adults 2285 37.2 23.3 4.6 31.4 72.3
Male adolescents 749 16.7 19.8 7.8 42.3 68.0
Female adults 485 36.8 21.7 4.8 21.7 77.4
Female adolescents 444 16.4 20.4 6.9 37.0 74.0
Black
Male adults 128 23.3 26.3 4.9 59.5 60.0

Medical Patients
White
Male adults 5353 49.0 — — 37.3 —

Female adults 6737 48.8 — — 20.3 —

MAC
(Raw Score)
False Hit
Sample N Age M SD Negatives Rate

Alcoholics
White
Male adults 6512 40.8 28.4 5.5 20.4 73.1
Male adolescents 409 1 7.6 27.1 4.0 1 8.4 77.6
Female adults 1045 40.8 25.3 4.3 46.5 76.4
Female adolescents 163 1 5.8 26.2 4.6 1 1.5 87.9
Black
Male adults 297 29.3 27.4 4.7 21.6 60.5

Polydrug Patients
White
Male adults 952 26.9 27.7 4.1 37.6 —

Female adults 127 41.5 — — 72.5

Black
Male adults 607 27.9 28.0 5.8 20.8 —

Note: From "Substance abuse/dependence" by R. L. Greene & R. D. Garvin in R. L. Greene (Ed.),

The MMPI: Use with specific populations, 1988, p. 181. Philadelphia: Grune & Stratton. Re¬
printed by permission.
190 Chapter 5

the performance on the MAC scale as a func¬ with psychiatric patients rather than normal
tion of sample, age, gender, and ethnicity individuals.
(Greene & Garvin, 1988). Hit rates in these white psychiatric
The weighted (by sample size) mean samples ranged around 75 percent, with ap¬
score on the MAC scale in normal, white, proximately 35 percent false positives. (The
adult male samples is 23.1 (SD — 4.3), which reader should note that these percentages do
is only slightly below MacAndrew’s tradi¬ not add to 100 since they are weighted
tional cutting score of 24. Gender differences means.) Both the hit rate and the false posi¬
in these normal, white, adult and adolescent tive percentage decreased about 5 percentage
samples are very consistent with men scoring points in the adult samples as compared with
approximately two raw-score points higher normal samples; however, the false positive
than women (M — 20.3; SD — 3.8). Normal, percentage more than doubled in the adoles¬
white adults and adolescents seem to have cent samples.
similar means and standard deviations; this Black, male adult psychiatric patients
finding is consistent with the report by Col- have a weighted mean score of 26.3 (SD —
ligan and Offord (1987b) of no age differ¬ 4.9) on the MAC scale that is significantly
ences on the MAC scale in their contempo¬ higher than MacAndrew’s (1965) recom¬
rary normative sample. Essentially no data mended cutting score of 24. These black pa¬
exist on the MAC scale in normal samples of tients had a mean score that was five raw-
nonwhite ethnic groups. Hit rates in these score points higher than their normal
normal samples ranged around 80 percent colleagues. Both the hit rate and the false
with approximately 20 percent false positives positive percentage were nearly 60 percent in
(normal individuals classified as alcoholic). these black patients. Although the sample
Both white, adult male (M — 23.3; SD size in these black patients was significantly
= 4.6) and female (M = 21.7; SD = 4.8) smaller than any of the other samples, which
psychiatric samples score slightly higher than may limit the generalizability of the results,
their normal counterparts (male: M = 23.1; clinicians should be very cautious in using the
SD = 4.3) (female: M = 20.3; SD = 3.8). MAC scale in nonwhite ethnic groups.
Again, men score approximately two raw- This caution about the use of the MAC
score points higher than women. Both male scale with blacks has been noted by several
and female psychiatric patients are more vari¬ authors (cf. Graham & Mayo, 1985; Walters,
able in their performance on the MAC scale Greene, Jeffrey, Kruzich, & Haskin, 1983).
with slightly larger standard deviations. Again, there is almost a total lack of data on
In contrast to the adult samples, both the MAC scale in nonwhite ethnic groups of
male and female adolescent psychiatric sam¬ psychiatric samples.
ples have lower weighted mean scores on the There has been only limited data on the
MAC scale than their normal counterparts. MAC scale in medical patients, which is
The adolescent psychiatric samples also are somewhat surprising since nearly 90 percent
extremely variable in their performance with of the individuals who misuse alcohol are
standard deviations that are almost twice as seen by their physician in a year (Kamerown,
large as those found in their normal counter¬ Pincus, & Macdonald, 1986). Davis, Col-
parts. This increase in variability in both ligan, Morse, and Offord (1987) reported
adult and adolescent psychiatric samples im¬ that their white, male medical patients had an
plies that hit rates and classification accuracy average of 37.3 percent false positives, and
will decrease when alcoholics are contrasted their female patients averaged 20.3 percent
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 191

false positives. These false positive percent¬ between black alcoholic and psychiatric pa¬
ages in medical patients are comparable with tients make it nearly impossible for the M/1 C
those reported in psychiatric patients, so it scale to discriminate between these two
would be expected that hit rates also would be groups; hence the caution about using the
comparable. MAC scale with nonwhite ethnic groups that
Colligan, Davis, Morse, and Offord was noted above. As expected because of the
(1988) found that none of seven alcoholism small mean differences between groups, hit
scales, one of which was the MAC, could be rates in black, male alcoholics averaged only
recommended for use in a medical setting be¬ 60 percent. There are limited data on the use
cause of the unsatisfactory classification of the MAC scale in black, female alcoholics
rates. They did find that MMPI item #215 and essentially no data for other ethnic
(MMPI-2 item #264), “I have used alcohol groups. Again, the need for research on the
excessively,” provided better classification MAC scale in all nonwhite ethnic groups
rates for their medical patients than any of should be evident.
the alcoholism scales they reviewed. The need White, adult male polydrug abusers had
for additional research on how to detect sub¬ a weighted mean score of 27.7 (SD = 4.1);
stance abuse and dependence in medical pa¬ black, adult male polydrug abusers had a
tients, particularly in other ethnic groups, weighted mean score of 28.0 (SD = 5.8). The
should be evident. false negative percentage in white polydrug
White, adult, male alcoholics had abusers was nearly double (37.6 percent) that
weighted mean scores significantly higher (M seen in their alcoholic counterparts (20.4 per¬
= 28.4; SD = 5.5) than their female counter¬ cent), whereas it was nearly identical in black
parts (M = 25.3; SD = 4.3), and both sam¬ polydrug abusers (20.8 percent) and alcohol¬
ples were substantially above MacAndrew’s ics (21.6 percent). It is evident that the MAC
(1965) suggested cutting score of 24. The pat¬ scale is not simply an alcoholism scale; rather
tern already noted in normal individuals and it is a more general measure of substance
psychiatric patients for men to score about abuse which includes alcohol as well as other
two raw-score points higher than women also drugs. The only data reported on white, adult
was apparent, and even larger in alcoholic female polydrug abusers revealed 72.5 per¬
samples. cent false negatives.
Male and female, white adolescent alco¬ Higher scores on the MAC sometimes
holics had weighted mean scores that were are interpreted as suggesting that the client is
somewhat comparable to those for adults. more likely to be a substance abuser or a
Hit rates and false negative (alcoholics classi¬ worse substance abuser. Actually, higher
fied as being nonalcoholic) percentages were scores on the MAC simply indicate that the
very comparable for white, male adult and client is more likely to remain above whatever
adolescent alcoholics. However, the false cutting score is being used despite any psy¬
negative percentage was significantly higher chometric error that may be associated with
in adult female alcoholics and significantly the scale. The interested reader is referred to
lower in adolescent female alcoholics. Black, Wiggins (1973) for an indepth review of this
male alcoholics had a weighted mean score of issue.
27.4 (SD = 4.7) that is only one raw-score A number of general conclusions can be
point higher than their psychiatric counter¬ drawn after this rather lengthy summary on
parts. the use of the MAC scale in a variety of sam¬
These small differences in mean scores ples:
192 Chapter 5

1. Men score about two raw-score points hit rates and classification accuracies for the
higher than women across most samples, MAC scale that have been reported above
which indicates that different cutting will be of limited usefulness in most real-life
scores will be necessary by gender. settings such as personnel selection or screen¬
2. There is not a single, optimal cutting ing medical patients. Gottesman and Prescott
score with scores anywhere from 24 to 29 (1989) have presented a cogent review of this
being used in different studies. issue that should be read by all clinicians.
One group of investigators (Hoffmann,
3. Cutting scores appear to be influenced
Loper, & Kammeier, 1974; Kammeier,
by a number of factors; clinicians need
Hoffmann, & Loper, 1973; Loper, Kamme¬
to begin to determine empirically the
ier, & Hoffmann, 1973) examined the MMPI
best cutting score for their specific treat¬
scores of male college students who were later
ment facility to optimize the percentage
treated for alcoholism. Among these men, an
of patients correctly classified as alco¬
average of 13 years elapsed between college
holic and nonalcoholic.
admission and entrance into an alcoholism
4. Clinicians need to be very cautious in treatment program.
using the MAC scale in nonwhite ethnic These investigators compared the
groups. alcoholics’ MAC scale scores upon admission
5. Hit rates and classification accuracy de¬ to college and at entrance into treatment with
crease when clinicians are trying to the scores of a control group of students who
discriminate between alcoholics and non¬ were admitted to college at the same time.
alcoholic, psychiatric patients, which is a The alcoholics had higher MAC scale scores
frequent differential diagnosis. both at college admission and at entrance
6. Hit rates and classification accuracy into treatment than the control group of stu¬
may be unacceptably low in medical dents. Using a cutting score of 26, the MAC
samples. scale correctly classified 72 percent of the al¬
coholic sample both at college admission and
In addition to the general conclusions at entrance into treatment.
that were drawn about the use of the MAC The consistency of classification by the
scale above, there is one overriding issue that MAC scale across such an extensive time in¬
cannot be disregarded and that is the base terval suggests that the MAC scale is tapping
rate of alcoholism for the setting in which the a dimension of behavior that is resistant to
MAC scale is being used. For example, a hit change. This conclusion also is supported by
rate of 75 percent and a false positive percent¬ the finding that the MAC scale scores in alco¬
age of 20 percent with the MAC scale is very holics remain elevated after treatment (Huber
respectable in a setting where the base rate for & Danahy, 1975; Gallucci, Kay, & Thornby,
alcoholism is 50 percent. The base rates for 1989; Rohan, Tatro, & Rotman, 1969).
alcoholism are much lower than 50 percent, High scorers on the MAC scale are de¬
however, in most settings. scribed as being very likely to abuse alcohol
It has been estimated that the lifetime and/or other drugs. They also are usually de¬
prevalence rates for alcoholism are approxi¬ scribed as being antisocial with a tendency to
mately 8 percent and approximately 20 per¬ act out, although Schwartz and Graham
cent of medical patients have substance abuse (1979) were unable to replicate this correlate
or substance-related problems (Robins, in their sample. They have been found to be
Helzer, Weisman, Orvaschel, Gruenberg, uninhibited, sociable individuals who appear
Burke, & Regier, 1984). Consequently, the to use repression and religion in an attempt to
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 193

control their rebellious, delinquent impulses pearance of being ‘neurotics-who-also-hap-

(Finney, Smith, Skeeters, & Auvenshine, pen-to-drink-too-much’” (p. 620).
1971). They also are described as being im¬ MacAndrew suggested that high scorers
pulsive, having a high energy level, having could be labeled primary alcoholics, and low
shallow interpersonal relationships, and scorers are reactive or secondary alcoholics.
being generally psychological maladjusted In this formulation of the MAC scale, sub¬
(Schwartz & Graham, 1979). stance abuse cannot be predicted since both
Personality and behavioral correlates of high and low scorers can abuse substances.
low scorers on the MAC scale have not been Rather the MAC scale is assessing a funda¬
reported, although MacAndrew (1981) has mental dimension of personality that will af¬
suggested one possible interpretation of low fect how the client will manifest his or her
scores. MacAndrew conjectured that the substance abuse. If this formulation by Mac¬
MAC scale taps a fundamental bipolar Andrew is accurate, clinicians would need to
dimenison of personality with high scorers avoid using the MAC scale to predict whether
(raw scores of 24 or higher) being described a client will abuse substances, which has been
as “moving (with ‘boldness’) into the world, the standard use of the MAC scale since it
albeit in a sometimes rancorous and ill-con¬ was first developed.
sidered fashion, with little regard for future Greene (1990) reported the mean MAC
consequences” (p. 618), whereas low scores scores by MMPI codetype, which would ap¬
(raw scores of 23 or lower) “give every ap¬ pear to support MacAndrew’s conceptualiza-

TABLE 5-6 Mean MAC Scores as a Function of MMPI Codetype by Gender in Psychiatric
Patients (Hedlund & Won Cho, 1979)

Codetype MAC <24 N M SD

Men
All pts 1 5.9% 6593 27.2 4.8
2-8/8-2 31.5 336 24.5 4.8
2-7/7-2 28.4 310 24.7 4.4
1-3/3-1 22.0 1 77 26.1 4.6
2-4Z4-2 1 5.6 748 26.8 4.5
7-8/8-7 1 5.3 308 27.4 4.7
4-676-4 , 8.3 278 28.3 4.3
6-878-6 i 9.8 389 28.5 4.4
6-979-6 3.3 121 29.6 3.9
4-979-4 2.3 686 30.2 3.9

Women
All pts 43.7% 2726 23.2 4.6
2-777-2 80.0 95 19.7 3.8
2-878-2 65.0 1 20 20.8 4.2
7-878-7 59.0 105 22.0 4.2
1-373-1 51.3 1 13 22.3 3.7
2-474-2 48.0 179 22.5 4.3
4-676-4 33.8 195 24.2 4.0
6-878-6 27.6 225 25.1 4.5
6-979-6 1 2.8 86 26.6 4.3
4-979-4 1 2.1 214 26.8 3.9
194 Chapter 5

tion of his scale as a measure of a dimension from the MAC scale to create a 49-item, sub¬
of personality (see Table 5-6) rather than as a tle scale of substance abuse.
measure of substance abuse per se. Some Since interpretation of the MAC scale is
codetypes (6-9/9-6, 4-9/9-4) in men and based on raw scores instead of T scores, four
women rarely have MAC scores less than 24 items were added to the MAC-R scale on the
and it would seem inappropriate to assume MMPI-2 by contrasting the item responses of
that all of these psychiatric patients abused a group of male alcoholics with male psychi¬
substances, whereas other codetypes (2-7/7-2, atric patients. Clinicians should note that
2-878-2) rarely have scores above the cutting item 387 on the MMPI-2, “I can express my
score of 24, particularly in women. true feelings only when I drink,”3 is retained
The correlates of these codetypes would on the MAC-R, even though the item content
suggest that high scorers on the MAC scale directly relates to drinking.
are impulsive, risk-taking, sensation-seeking There has been only one reported study
individuals, who also may abuse substances, of the comparability of the MAC and MAC-
whereas low scorers are depressed, inhibited, R scales (Greene, Arredondo, & Davis,
overcontrolled individuals, who also may 1990), and it appears that the two scales
abuse substances, but in a different manner. produce comparable scores (see Table 5-7).
Clinicians should see the apparent danger of The only notable difference in the MAC
identifying a client as a potential substance and MAC-R scores for these groups of al¬
abuser or not based on the MAC scale with¬ coholic and psychiatric inpatients was for
out considering the codetype. For example, male alcoholics to have slightly higher
clients with 2-777-2 codetypes, who are likely scores on the MAC-R scale than on the
to be “neurotic” and risk avoiders, would be MAC scale. It appears that the MAC-R
unlikely to have elevated MA C scores regard¬ scale can be used in a similar manner to the
less of whether they abuse substances, MAC scale, although clinicians again are
whereas clients with 6-979-6 codetypes, who cautioned to note all the caveats described
are likely to act out, be impulsive, and to be above.
risk takers, would be very likely to have ele¬ The MAC-R has only limited item over¬
vated MAC scores again regardless of lap with the other clinical and supplementary
whether they abuse substances. scales (see Tables 5-3 and 5-4). The MAC-R
Four MMPI items (58, 378, 483, and shares 5 items with the F scale, 7 items with
488) on the MAC scale were deleted in devel¬ Scale 4 (Psychopathic Deviate), and 5 items
oping the MMPI-2. The rationale for drop¬ with Scale 9 (Hypomania). The MAC-R
ping these items is apparent since they have shares 6 items with Do (Dominance) and 9
either religious content or a sexist bias. Item items with Re (Social Responsibility),
460, “I have used alcohol moderately (or not with the items being scored in the oppo¬
at all.),”2 also was deleted, which along with site direction on the Do and Re scales. It
item 215, is found on the 51-item version of appears very logical that high scorers on the
the MAC scale. The deletion of item 460, MAC-R scale are not socially responsible.
since item 215 was retained is not clear be¬
cause Colligan and colleagues (1988) found
that this item, after item 215, provided better
CONTENT SCALES
classification rates than any of the alcoholism The content of individual items was basically
scales in their medical patients. The deletion ignored in the development of the MMPI
of item 460 may not be a problem, however, scales because at that time methods of item
since items 215 and 460 are deleted frequently selection that relied extensively on item con-
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 195

TABLE 5-7 Comparability between the MMPI MAC Scale and the MMPI-2 MAC-R Scale
in Alcoholics and Psychiatric Inpatients

Alcoholics

N M SD Range <24

MAC
Men 279 26.8 4.7 10-39 24.0%
Women 147 24.6 4.2 1 5-36 41.5

MAC-R
Men 66 28.1 4.3 19-38 16.8
Women 33 23.8 4.2 14-29 42.4

Psychiatric Inpatients

N M SD Range >23

MAC
Men 797 26.4 4.9 10-40 72.9%
Women 736 23.7 4.6 1 1 -36 49.6

MAC-R
Men 45 26.7 4.5 19-36 73.3
Women 54 23.0 4.9 14-34 42.6

Data are from Greene, Arredondo, and Davis (1990).

Note: Each sample consists of two separate groups of patients taking either the MAC or MAC-R
scale in the same setting, not a single group of patients taking both scales.

tent had fallen into disfavor. Since these is¬ Wiggins' MMPI Content Scales
sues were reviewed in Chapter 1, they will not
be reiterated here. Numerous investigators Wiggins (1966) began by grouping the MMPI
(cf. Jackson, 1971), however, believed that items into the 26-item content categories (see
item content should not be ignored in test Table 1-2, page 5), originally proposed by
construction or interpretation. Hathaway and McKinley (1940). He then ap¬
Wiggins’ (1966) Content scales represent plied both psychometric and intuitive proce¬
a systematic attempt to develop a means of dures to produce 13 substantive dimensions
examining the client’s responses to the con¬ of item content. He developed his Content
tent of individual MMPI items, and the scales on normal college students and vali¬
Butcher, Graham, Williams, and Ben-Porath dated them on additional normal populations
(1989) Content scales reflect a similar ap¬ and a psychiatric sample.
proach on the MMPI-2. Since Wiggins’ Con¬ Wiggins selected the items for the Con¬
tent scales were developed prior to the tent scales so that they did not share common
Butcher and colleagues content scales on the items with one another and so that scale ho¬
MMPI-2, Wiggins" scales will be reviewed mogeneity was maximized; each content scale
first. Clinicians’ interest in critical items, dis¬ was constructed to be a homogeneous mea¬
cussed in a later section, reflect a similar in¬ sure of its substantive dimension that did not
terest in item content. overlap with other scales. The name and ab-
196 Chapter 5

breviation of each of these 13 scales appear in derreporting of psychopathology is sus¬

Table 5-8. The items found on each Content pected.
scale as well as the mean and standard devia¬ Research on the Wiggins Content scales
tion on each scale for males and females in has consistently supported their validity and
the original Minnesota normative group are generalizability to new populations. Wiggins,
provided in Dahlstrom, Welsh, and Dahl- Goldberg, and Appelbaum (1971), Jarnecke
strom (1972, 1975). and Chambers (1977), and Mezzich, Damarin,
Profile sheets to convert the raw scores and Erickson (1974) found that the Content
directly to T scores are available from Na¬ scales demonstrated the expected relationships
tional Computer Systems (P.O. Box 1416, to other test measures and behaviors. For ex¬
Minneapolis, Minnesota 55440, 800-627- ample, clients who elevated the Wiggins De¬
7271) and Psychological Assessment Re¬ pression scale were more likely to exhibit de¬
sources (P.O. Box 998, Odessa, Florida pressive behaviors and to elevate other tests
33556, 800-331-8378). Fowler and Coyle that measured depression than clients who
(1969) also have provided normative data on did not elevate this scale.
college students that can be used to derive T Jarnecke and Chambers (1977), Lachar
scores for each content scale in college popu¬ and Alexander (1978), and Mezzich and asso¬
lations. ciates (1974) found that the Content scales
Wiggins (1966) construed the Content were generalizable to psychiatric inpatients
scales as reflecting the client’s admission or and Air Force personnel. Payne and Wiggins
self-report of symptomatology; the client’s (1972) reported that hospitalized patients
self-report is one way of communicating with the same profile type, as defined by the
problems to the clinician. Since the Content Gilberstadt and Duker (1965) profile classifi¬
scales are an obvious measure of symptom¬ cation system, tended to produce similar
atology, it is possible for the client to present scores on the Content scales, and these scores
an inaccurate self-appraisal. were consistent with the established corre¬
Lachar and Alexander (1978) suggested lates of the codetype. Consequently, it seems
that the Content scales are indeed susceptible that the Wiggins Content scales adequately
to a set to deny the presence of psychopathol¬ assess their respective dimensions, and they
ogy. Therefore, low scores on the Content can be generalized to new populations with
scales could represent the absence of the spe¬ little, if any, loss in predictive power.
cific descriptors that are characteristic of The Wiggins Content scales can be un¬
high scorers or signify the client’s refusal to derstood through their relationship to the
acknowledge the presence of such symptom¬ various clinical scales; these relationships are
atology. illustrated in Table 5-9. For example,
It also should be possible for clients to Wiggins’ Content scale of Poor Health
exaggerate the severity of their symptomatol¬ (HEA) relates to Scale 1 (Hypochondriasis),
ogy and elevate their scores on the Content Depression (DEP ) relates to Scale 2 (Depres¬
scales, although no research has addressed sion), and so on.
this issue. Thus, high scores on the Content The Wiggins Content scales provide an
scales may be the result of actual symptom¬ additional source of information that the cli¬
atology or the client’s overreporting of psy¬ nician can use in interpreting an MMPI pro¬
chopathology. The procedures described in file. It is possible for two clients to have sim¬
Chapter 3 for assessing the accuracy of item ilar profiles on the standard clinical scales
endorsement should be followed routinely and very different patterns of scores on the
and particularly when overreporting or un¬ Content scales. Thus, the Content scales can
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 197

TABLE 5-8 Description of Wiggins' MMPI Content Scales

Abbreviation Name Description

HEA Poor Health High HEA is concerned about his health and has admitted
to a variety of gastrointestinal ailments centering around
an upset stomach and difficulty in elimination.
DEP Depression High DEP experiences guilt, regret, worry, unhappiness,
and a feeling that life has lost its zest. He experiences
difficulty in concentrating and has little motivation to pur¬
sue things. His self-esteem is low, and he is anxious and
apprehensive about the future. He is sensitive to slight,
feels misunderstood, and is convinced that he is unwor¬
thy and deserves punishment. In short, he is classically
depressed.
ORG Organic Symptoms High ORG admits to symptoms that are often indicative
of organic involvement. These include headaches, nau¬
sea, dizziness, loss of motility and coordination, loss of
consciousness, poor concentration and memory, speak¬
ing and reading difficulty, and problems with muscular
control, skin sensations, hearing, and smell.
PAM Family Problems High FAM feels that he had an unpleasant home life char¬
acterized by a lack of love in the family and parents who
were unnecessarily critical, nervous, quarrelsome, and
quick tempered. Although some items are ambiguous,
most are phrased with reference to the parental home
rather than to the individual's current home.
AUT Authority Conflict High AUT sees life as a jungle and is convinced that oth¬
ers are unscrupulous, dishonest, hypocritical, and moti¬
vated only by personal profit. He distrusts others, has lit¬
tle respect for experts, is competitive, and believes that
everyone should get away with whatever they can.
FEM Feminine Interests High FEM admits to liking feminine games, hobbies, and
vocations. He denies liking masculine games, hobbies,
and vocations. Here there is almost complete contamina¬
tion of content and form that has been noted in other
contexts by several writers. Individuals may score high
on this scale by presenting themselves as liking many
things since this item stem is present in almost all items.
They may also score high by endorsing interests, which,
although, possibly feminine, are also socially desirable,
such as an interest in poetry, dramatics, news of the
theater, and artistic pursuits. This has been noted in the
case of Wiggins' Sd (Social Desirability) scale. Finally,
of course, individuals with a genuine preference for
activities that are conceived by our culture as "femi¬
nine" will achieve high scores on this scale.
REL Religious High scorers on this scale see themselves as religious,
Fundamentalism church-going people who accept as true a number of
fundamentalist religious convictions. They also tend to
view their faith as the true one.

continued
198 Chapter 5

TABLE 5-8 continued

Abbreviation Name Description

HOS Manifest Hostility High HOS admits to sadistic impulses and a tendency to
be cross, grouchy, competitive, argumentative, uncoop¬
erative, and retaliatory in his interpersonal relationships.
Fie is often competitive and socially aggressive.
MOR Poor Morale High MOR is lacking in self-confidence, feels that he has
failed in life, and is given to despair and a tendency to
give up hope. He is extremely sensitive to the feelings
and reactions of others and feels misunderstood by
them while at the same time being concerned about of¬
fending them. He feels useless and is socially suggest¬
ible. There is a substantive overlap here between the De¬
pression and Social Maladjustment scales and the Poor
Morale scale. The Social Maladjustment scale seems to
emphasize a lack of social ascendance and poise, the
Depression scale feelings of guilt and apprehension,
while the present scale seems to emphasize a lack of
self-confidence and hypersensitivity to the opinions of
others.
PHO Phobias High PHO has admitted to a number of fears, many of
them of the classically phobic variety such as heights,
dark, closed spaces, etc.
PSY Psychoticism High PSY admits to a number of classic psychotic symp¬
toms of a primarily paranoid nature. He admits to halluci¬
nations, strange experiences, loss of control, and classic
paranoid delusions of grandeur and persecution. He
admits to feelings of unreality, daydreaming, and a sense
that things are wrong, while feeling misunderstood by
others.
HYP Hypomania High HYP is characterized by feelings of excitement, well¬
being, restlessness, and tension. He is enthusiastic,
high strung, cheerful, full of energy, and apt to be hot¬
headed. He has broad interests, seeks change, and is
apt to take on more than he can handle.
SOC Social High SOC is socially bashful, shy, embarrassed, reticent,
Maladjustment self-conscious, and extremely reserved. Low SOC is
gregarious, confident, assertive, and relates quickly and
easily to others. He is fun loving, the life of a party, a
joiner who experiences no difficulty in speaking before a
group. This scale would correspond roughly with the
popular concept of "introversion-extroversion."

Note: From "Substantive Dimensions of Self-report in the MMPI Item Pool" by J. S. Wiggins,
1 966, Psychological Monographs, 80 (22, Whole No. 630). Copyright © 1 966 by the American
Psychological Association. Reprinted by permission.
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 199

TABLE 5-9 Relationship between the Wiggins Content Scales and the MMPI Clinical Scales

Wiggins Content Scales

Abbreviation Name Related MMPI Clinical Scales

HEA Poor Health Scale 1 (Hypochondriasis)

DEP Depression Scale 2 (Depression)
ORG Organic Symptoms Scale 3 (Hysteria)
FAM Family Problems Scale 4 (Psychopathic Deviate)
AUT Authority Conflict Scale 4 (Psychopathic Deviate)
FEM Feminine Interests Scale 5 (Masculinity-Femininity)
REL Religious Fundamentalism Scale 5 (Masculinity-Femininity)
HOS Manifest Hostility Scale 6 (Paranoia)
MOR Poor Morale Scale 7 (Psychasthenia)
PHO Phobias Scale 7 (Psychasthenia)
PSY Psychoticism Scale 8 (Schizophrenia)
HYP Hypomania Scale 9 (Hypomania)
SOC Social Maladjustment Scale 0 (Social Introversion)

be used to supplement and expand the inter¬ (see Table 5-11), so clinicians can continue to
pretation of the standard MMPI profile. Cli¬ use these scales if desired. Only the Religious
nicians are encouraged to score and interpret Fundamentalism scale has been totally
the Wiggins Content scales routinely because dropped. Most of the Wiggins Content scales
of the valuable information that can be ob¬ retain 90 percent or more of their items on the
tained with little additional investment of MMPI-2.
time or effort.
Nichols (1987) has provided an excellent
MMP1-2 Content Scales
overview of the Wiggins Content scales and
their relationship with their respective clinical Butcher and colleagues (1989) began to de¬
scales as well as illustrative cases. Nichols’ velop the MMPI-2 Content scales by sorting
monograph should be consulted by every cli¬ the 704 items of the AX booklet (the original
nician who is interested in learning more 550 items from the MMPI plus 154 additional
about these scales. new items that were under consideration for
Wiggins’ descriptors of high scorers on inclusion in the MMPI-2) into possible cate¬
the Content scales are presented in Table 5-8. gories based on their item content. These ini¬
Lachar and Alexander (1978) also presented tial groupings of items were refined statistically
preliminary interpretations of high scores on to insure their psychometric homogeneity.
each content scale that were validated and Finally, only those items whose content
replicated in an armed service setting (see was homogeneous with the rest of the items
Table 5-10). The clinician can use these inter¬ were retained. This procedure resulted in 15
pretations of high scores in conjunction with Content scales: their general clinical area,
Wiggins’ descriptors (see Table 5-8) and names, abbreviations, and description are
Nichols’ (1987) monograph as a basis for in¬ provided in Table 5-12. An easy means of un¬
terpreting these scales. derstanding each content scale is to review
Most of the items on the Wiggins Con¬ the actual items on each scale (see Appendix
tent scales have been retained on the MM PI-2 A for the item numbers or Butcher and asso-
200 Chapter 5

TABLE 5-10 Proposed Interpretation of High Wiggins Content Scales

Abbreviation Name Description

HEA Poor Health A significant number of physical complaints are reflected

(T > 69) by item endorsement centering mainly around the diges¬
tive system. Individuals who obtain high HEA elevations
are often considerably worried about their health. Car¬
diac and pulmonary complaints are also occasionally re¬
ported.
DEP Depression This individual has admitted to symptoms associated
(T > 69) with problematic depression, such as lack of interest in
the environment, pessimism, self-criticism, and brooding.
In client populations, social withdrawal, a negative self-
concept, guilt feelings, and a reduced activity level may
be suggested.
ORG Organic Symptoms This individual has admitted to a variety of sensory, mo¬
(T > 69) tor, or general somatic concerns that may be related to
psychological discomfort and general malaise as well as
to reduced effectiveness in completing daily tasks. Cli¬
ents who obtain high ORG elevations may complain of
lack of stamina and strength, may present physical
symptoms that often indicate emotional conflict, such as
problematic headache or back pain.
FAM Family Problems Inventory responses include admission of pathology in
(T > 69) and among family members. A history of poor relation¬
ships with parents is suggested, as well as the absence
of positive supports in current family interactions,
whether with parents, spouse, or extended family. Pa¬
tient male: In adult male clients admission of family pa¬
thology may reflect not only marital conflict but may also
suggest intolerant, overactive individuals and a negative
self-concept. Drug abuse and other destructive behavior
may be associated.
AUT Authority Conflict Endorsed item content reflects the belief that interper¬
(T > 59) sonal relations are often exploitive in nature. Disregard
for principles of ethical conduct and truthfulness is sug¬
gested, as well as a tendency to minimize the negative
impact of antisocial behavior. In client populations these
attitudes may be associated with problematic overas-
sertive and manipulative social relations. Conflict with
relatives may result.
FEM Feminine Interests Inventory responses suggest an interest in pursuits tradi¬
(T > 59) tionally labeled as feminine and/or dislike of activities
stereotyped as masculine. Patient male: In male clients
this interest pattern may be associated with an indeci¬
sive, passive orientation that has proven to be problem¬
atic. Conflict may lead to confusion or self-blame. Evalu¬
ation for suicide ideation or previous attempts is
suggested.
REL Religious Endorsed item content reflects strong religious beliefs
Fundamentalism and religiously motivated behavior. In client populations
(T > 59) this orientation suggests a reduced probability of sub-

continued
TABLE 5-10 continued

Abbreviation Name Description

stance abuse, impulsive behaviors, and conflict with

family members. Expression of strong religious beliefs
may, at times, reflect a delusional system and associ¬
ated thought disorder.
HOS Mainfest Hostility This individual admits to problems in adjustment related
(T > 59) to unmodulated expressions of anger, resentment of
perceived injustices, need for interpersonal dominance,
and limited self-control. In client populations the combi¬
nation of hostility, moodiness, and impulsivity may be as¬
sociated with assaultive or other antisocial or violent be¬
havior.
MOR Poor Morale Inventory responses reflect a pervasive lack of confi¬
(T > 69) dence in one's own abilities and a history of failure,
which is related to these perceived limitations. Clients
who obtain high MOR elevations may be insecure, de¬
spondent, withdrawn, intropunitive, and oversensitive,
and may become easily upset by the actions of others.
PHO Phobias This individual admits to a variety of fears and appears to
(T > 69) be significantly uncomfortable in many situations. Cli¬
ents who obtain high PHO elevations are viewed as more
anxious, tremulous, worrisome, and phobic than most
patients. Depression and social withdrawal may also be
indicated.
PSY Psychoticism Inventory responses include admission of unusual experi¬
(T > 69) ences and beliefs, many of which may include a clearly
paranoid component. In client populations this response
pattern often suggests an individual who finds compre¬
hension of human motives and behavior difficult and is
consequently suspicious of and worried about others.
Symptoms associated with a psychotic adjustment, such
as ideas of reference, hallucinations, and autistic or dis¬
organized thought, may be present.
HYP Hypomania This individual's self-description suggests a fast personal
(T > 59) tempo characterized by enthusiasm, cheerfulness, and
perhaps irritability of emotional lability. Clients who ob¬
tain high HYP elevation are often described as imma¬
ture, hyperactive, excitable, agitated, and restless. They
are unlikely to respond intropunitively to conflict and
may manipulate others to reach their goals.
SOC Social Endorsed item content reflects a lack of social skill and
Maladjustment poise, discomfort in social interaction, and resultant in¬
(T > 69) hibition and social isolation. In client populations this lack
of social supports may be associated with a negative
self-image, feelings of despair or fearfulness, thoughts of
suicide, or a defensive orientation characterized by apa¬
thy and limited activity or compulsive attention to detail.

Note: From "Veridicality of self-report: Replicated correlates of the Wiggins MMPI content scales"
by D. Lachar & R. S. Alexander, 1978, Journal of Consulting and Clinical Psychology, 46, 1 355-
1356. Copyright © 1978 by the American Psychological Association. Reprinted by permission.
Also from "WPS TEST REPORT User's Manual for the MMPI" by D. Lachar, 1 979, 9-10. Copy¬
right © 1 979 by Western Psychological Services. Reprinted by permission of the publisher, West¬
ern Psychological Services, 12031 Wilshire Blvd., Los Angeles, CA 90025.
202 Chapter 5

TABLE 5-11 Wiggins' Content Scales Items on the MMPI-2

Items Number of
Wiggins Content Scale on MMPI Scale Items Retained on MMPI-2

Poor Health 28 19 ( 67.9%)

Depression 33 33 (100.0%)
Organic Symptoms 36 32 ( 88.9%)
Family Problems 16 16 (100.0%)
Authority Conflict 20 20 (100.0%)
Feminine Interests 30 23 ( 76.7%)
Religious Fundamentalism 12 1 ( 8.3%)
Manifest Hostility 27 25 ( 92.6%)
Poor Morale 23 22 ( 95.6%)
Phobias 27 26 ( 96.3%)
Psychoticism 48 45 ( 93.8%)
Hypomania 25 23 ( 92.0%)
Social Maladjustment 27 26 ( 96.2%)

ciates [1989, pp. 192-199] for a listing of the ever, other relationships among these scales
actual items). would not be expected: Depression has only
As can be seen in Table 5-13, most of the nine items in common with Scale 2 (Depres¬
items on the MMPI-2 Content scales are sion) and shares items with most of the other
from the original MMPI item pool except for clinical scales; Antisocial Practices, Cyni¬
the Low Self-Esteem, Work Interference, cism, and Anger have few items in common
and Negative Treatment Indicators scales. with Scale 4 (Psychopathic Deviate), and
That is, the MMPI-2 Content scales are not Cynicism and Anger share few items with any
composed primarily of new items. of the standard validity and clinical scales;
There is a substantial degree of item and Anxiety shares items with 11 of the stan¬
overlap within the MMPI-2 Content scales dard validity and clinical scales.
(see Table 5-14). Only three scales (Fears, It also is informative to examine the item
Health Concerns, and Social Discomfort) overlap between the Wiggins and the MMPI-
have no items in common with the other Con¬ 2 Content scales that share the same or sim¬
tent scales. Four scales (Obsessiveness, Cyni¬ ilar names (see Table 5-16). Only two Con¬
cism, Work Interference, and Negative tent scales (Fears and Phobias; Bizarre
Treatment Indicators) have almost one-half Mentation and Psychoticism) have substan¬
of their items in common with the other Con¬ tial item overlap.
tent scales. There also is significant overlap In most instances, MMPI-2 and Wig¬
among the standard validity and clinical gins’ Content scales with the same or similar
scales and the MMPI-2 Content scales (see names have a limited number of items in
Table 5-15). common. For example, the Wiggins Depres¬
Some of the scales that share items make sion scale and the MMPI-2 Content scale of
good intuitive sense: Health Concerns with Depression have only 16 of their 33 items in
Scales 1 (Hypochondriasis) and 3 (Hysteria); common, and both of these scales have few
Bizarre Mentation with Scales 6 (Paranoia) items in common with Scale 2 (Depression).
and 8 (Schizophrenia); and Social Discom¬ Whether these three “Depression” scales are
fort with Scale 0 (Social Introversion). How¬ measuring different facets of depressive phe-
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 203

TABLE 5-12 Description of the MMPI-2 Content Scales

Clinical Area

Abbreviation Name Description

Internal Symptomatic Behaviors

ANX Anxiety These clients report general symptoms of anxiety, ner¬
vousness, worries, and sleep and concentration difficul¬
ties. They have difficulty making decisions. They fear los¬
ing their minds and are afraid that they are about to go
to pieces. They find life a strain and work under a great
deal of tension and stress.
FRS Fears These clients report a large number of specific fears: ani¬
mals (snakes, mice, spiders); events in nature (dirt,
earthquakes, fire, lightning, windstorms, water); dark,
blood, money, high places, and so on.
OBS Obsessiveness These clients have great difficulty making decisions, ru¬
minate excessively, worry excessively, and have intru¬
sive thoughts. They dislike change. They count and save
unimportant things.
DEP Depression These clients have a depressive mood and depressive
thoughts. They feel blue and unhappy, and are likely to
brood. They are uncertain about their future and find their
lives empty and meaningless. They cry easily. They are
self-critical, guilty, and lonely. They may report suicidal
ideation or attempts.
HEA Health Concerns These clients report specific physical symptoms across
several body systems as well as general physical ail¬
ments such as nausea, vomiting, and pain. They have
pains in the chest, neck, and head. Their muscles may
be paralyzed or twitch and jump. They have fainting and
dizzy spells. They worry about their health and catching
disease. They believe that they are sicker than most
people.
BIZ Bizarre Mentation These clients report strange thoughts and experiences,
paranoid ideation, and hallucinations; in short, they re¬
port psychotic thought processes.

External Aggressive Tendencies

ANG Anger These clients report being irritable, grouchy, impatient,
hotheaded, annoyed, and stubborn. In addition, they
may swear or lose control and smash objects, pick fights,
and hurt someone in a fight.
CYN Cynicism These clients expect other people to lie, cheat, and steal,
and if they do not engage in these behaviors, it is be¬
cause they fear being caught. They trust nobody. People
use each other and are friendly only for selfish reasons.
ASP Antisocial These clients report stealing things, other problem behav¬
Practices iors, and antisocial practices during their school years.
They have attitudes similar to individuals who break the

continued
204 Chapter 5

TABLE 5-12 continued

Clinical Area

Abbreviation Name Description

law, even if not actually engaging in antisocial behavior.

They expect other people to lie.
TPA Type A These clients are hard-driving, fast-moving, and work-ori¬
ented individuals, who frequently become impatient,
grouchy, irritable, and annoyed. They do not like to wait,
be interrupted, or believe that someone has gotten the
best of them.

Negative Self-Views
LSE Low Self-Esteem These clients have very low opinions of themselves, and
they are uncomfortable if people say nice things about
them. They believe that they are unattractive, awkward
and clumsy, useless, and a burden to others, who do
not like them. They lack self-confidence. They see them¬
selves as not as good or capable as others, and they see
themselves as not being able to do anything well.

General Problem Areas:

Social, Familial, Work, and Treatment
SOD Social Discomfort These clients are very uneasy around others and are hap¬
pier by themselves. They see themselves as shy. They
dislike parties and other group events, because they do
not like meeting people and find it hard to talk.
FAM Family Problems These clients report considerable familial discord. Their
families are lacking love, support, and companionship,
and these clients wanted to leave home. Family members
are described as being nervous and having quick tem¬
pers, and are to be avoided at best, or may even be hated.
WRK Work Interference These clients report that they are not as able to work as
they once were and that they work under a great deal
of tension. They are tired, lack energy, and sick of what
they have to do. They dislike making decisions and lack
self-confidence. They give up easily and shrink from fac¬
ing a crisis or problem.
TRT Negative These clients dislike going to doctors and they believe
Treatment that they should not discuss their persona! problems
Indicators with others. They prefer to take drugs or medicine, since
talking about problems does not help them. They do not
believe that anyone understands or cares about them.
They give up quickly and do not care about what is hap¬
pening to them, since nothing can be done about their
problems. They have a hard time making decisions.

Note: Adapted from: Butcher, Dahlstrom, Graham, Tellegen, and Kaemmer (1989), Manual for
administration and scoring of the MMPI-2. Minneapolis: University of Minnesota Press; and
Butcher, Graham, Williams, and Ben-Porath (1 989), Development and use of the MMPI-2 content
scales. Minneapolis: University of Minnesota Press.
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 205

TABLE 5-13 Source of Items on MMPI-2 Content Scales

Abbre- Number MMPI New

Content Scale viation of Items Items Items

Anxiety ANX 23 20 3
Fears FRS 23 22 1
Obsessiveness OBS 16 10 6
Depression DEP 33 25 8
Health Concerns HEA 36 36 0
Bizarre Mentation BIZ 23 19 4
Anger ANG 16 10 6
Cynicism CYN 23 21 2
Antisocial Practices ASP 22 21 1
Type A TPA 19 12 7
Low Self-Esteem LSE 24 13 1 1
Social Discomfort SOD 24 21 3
Family Problems PAM 25 16 9
Work Interference WRK 33 22 1 1
Negative Treatment TRT 26 8 18
Indicators

nomenology, or different types of depression ity scale and the MMPI-2 Content scale of
remains to be determined. Anger share only 8 items. Clinicians may
Similarly, the Wiggins Family Prob- find it beneficial to score both the MMPI-
lems scale and the MMPI-2 Content scale 2 and the Wiggins Content scales to deter-
of Family Problems have only 10 items in mine whether these scales with similar
common, and the Wiggins Manifest Hostil- names and limited item overlap have dif-

TABLE 5-14 Item Overlap within the MMPI-2 Content Scales

Abbre- Number Unique Overlapping Content Scales

Scale viation of Items Items and Number of Overlapping Items

Anxiety ANX 23 18 ( 78.3%) WRK-5 OBS-2

Fears FRS 23 23 (100.0%)
Obsessiveness OBS 16 9 ( 56.3%) WRK-4 TRT-3 ANX-2
Depression DEP 33 23 ( 69.7%) TRT-6 LSE-2 CYN-1 WRK-1
Health Concerns HEA 36 36 (100.0%)
Bizarre Mentation BIZ 23 22 ( 95.7%) FAM-1
Anger ANG 16 14 ( 87.5%) TPA-3 WRK-1
Cynicism CYN 23 12 ( 52.2%) ASP-7 DEP-1 TPA-1 WRK-1 TRT-1
Antisocial Practices ASP 22 16 ( 72.7%) CYN-7 TPA-1
Type A TPA 19 14 ( 73.7%) ANG-3 WRK-2 ASP-1 CYN-1
Low Self-Esteem LSE 24 20 ( 83.3%) DEP-2 TRT-2 WRK-1
Social Discomfort SOD 24 24 (100.0%)
Family Problems FAM 25 24 ( 96.0%) BIZ-1 WRK-1
Work Interference WRK 33 18 ( 54.5%) ANX-5 OBS-4 TRT-4 TPA-1 ANG-
1 CYN-1 DEP-1 LSE-1 FAM-1
Negative Treatment TRT 26 14 ( 53.8%) DEP-6 WRK-4 OBS-3 LSE-2 CYN-1
Indicators
206 Chapter 5

TABLE 5-15 Item Overlap among the MMPI-2 Standard Validity and Clinical Scales
and the MMPI-2 Content Scales

L F K Hs D Hy Pd Mf Pa Pt Sc Ma Si Total

HEA 0 3 0 23 8 16 0 1 1 3 6 0 2 63
DEP 0 3 0 1 9 4 6 0 4 9 7 0 2 45
FAM 0 2 0 0 0 1 5 2 1 0 6 4 0 21
ASP 0 3 0 0 0 0 3 0 1 0 1 4 3 1 5
CYN 0 0 0 0 0 0 1 0 0 0 0 1 4 6
ANG 0 0 0 0 0 0 0 0 0 1 0 0 1 2
ANX 0 1 0 2 5 3 1 1 1 6 4 1 1 26
FRS 0 1 0 0 0 1 0 1 1 2 2 0 0 8
OBS 0 0 0 0 0 0 0 1 0 5 0 2 1 9
BIZ 0 9 0 0 0 0 1 0 8 1 8 0 0 27
LSE 0 1 0 0 3 0 0 0 0 4 0 0 3 1 1
SOD 0 1 0 0 2 0 0 1 0 2 3 0 18 27
WRK 0 3 0 1 5 2 3 0 1 3 4 2 6 30
TRT 0 1 0 0 1 0 1 0 1 0 3 0 2 9
TP A 0 0 0 0 0 0 0 0 0 1 0 1 1 3
Totals 0 28 0 27 33 27 21 7 19 37 44 1 5 44

Note: Since an individual item may overlap with several scales, the totals represent how many
times items overlap on the various scales when stored in the same directory, not how many indi¬
vidual items on a given scale overlap with other scales.

ferent correlates. As noted earlier, most of concluding this section. First, since the
the Wiggins Content scales can still be scored MMPI-2 Content scales are obvious mea¬
on the MMPI-2. sures of symptomatology, it is possible for
Several general comments need to be the client to present an inaccurate self-ap¬
made about the MMPI-2 Content scales in praisal. The cautions noted earlier for the

TABLE 5-16 Item Overlap between the Wiggins Content Scales and Related MMPI-2
Content Scales

Items Item Items

Wiggins on Scale Overlap8 MMPI-2 on Scale

Poor Health 28 13(36.1%) Health Concerns 36

Depression 33 16(48.5%) Depression 33
Organic Symptoms 36 20(55.6%) Health Concerns 33
Family Problems 16 10(40.0%) Family Problems 25
Authority Conflict 20 12(52.2%) Cynicism 23
Authority Conflict 20 1 2(54.5%) Antisocial Practices 22
Manifest Hostility 27 8(50.0%) Anger 16
Poor Morale 23 6(25.0%) Low Self-Esteem 24
Phobias 27 19(82.6%) Fears 23
Psychoticism 48 19(82.6%) Bizarre Mentation 23
Social Maladjustment 27 16(66.7%) Social Discomfort 24

aThe percentages indicate the percent of items on the related MMPI-2 Content Scale that are also
on the Wiggins Content scale
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 207

Wiggins Content scales also need to be con¬ criminals demonstrated lower hostility scores
sidered for the MMPI-2 Content scales. and better impulse control than the other
Low scores on the MMPI-2 Content groups). In a similar study of 21 MMPI scales
scales (T scores less than 45) could represent and indexes of hostility Deiker (1974) also re¬
the absence of the specific descriptors charac¬ ported that his extremely assaultive group
teristic of high scorers or the client’s refusal scored significantly lower than other assaul¬
to acknowledge the presence of such symp¬ tive groups on 13 of 17 scales that were signif¬
tomatology. Similarly, high scores on the icantly different.
MMPI-2 Content scales may be the result of In light of these paradoxical results,
actual symptomatology or the client’s over¬ Megargee and associates (1967) developed a
reporting of psychopathology. The proce¬ new scale to assist in the identification of as¬
dures described in Chapter 3 for assessing the saultive individuals. They began by distin¬
accuracy of item endorsement should be fol¬ guishing between undercontrolled and over-
lowed routinely and particularly when over¬ controlled assaultive individuals because they
reporting or underreporting of psychopathol¬ believed that different factors led to assaul¬
ogy is suspected. tive outbursts in these two groups. They be¬
Second, the deviant response for a ma¬ lieved that undercontrolled individuals have
jority of the items on the MMPI-2 Content failed to learn to control their aggressive im¬
scales is “true,” so a TRIN (True Response pulses, and their aggressive behaviors occur
Inconsistency scale) score of 12 or higher in response to some external form of provo¬
would be expected to produce elevations on cation. Overcontrolled individuals, however,
most of these scales. Third, a number of the rigidly defend against any expression of ag¬
items on these scales are clustered in the last gressive impulses irrespective of the provoca¬
100 items so any waning of the client’s moti¬ tion until finally some provocation or other
vation toward the end of the test could ad¬ factor results in their acting out, frequently in
versely affect these scales. Finally, the an extremely destructive fashion.
MMPI-2 Content scales use uniform T Megargee and colleagues (1967) felt that
scores. the latter group was more important to iden¬
tify since their assaultive behaviors occur un¬
expectedly and frequently very violently.
ADDITIONAL MMPI-2 They contrasted the item responses of four
SUPPLEMENTARY SCALES groups of men: 14 extremely assaultive pris¬
oners, 25 moderately assaultive prisoners, 25
nonassaultive prisoners, and 46 normals. The
Overcontrolled Hostility {O H ) Scale
55 items that differentiated between the as¬
Megargee and Mendelsohn (1962) attempted saultive and nonassaultive prisoners were
to cross-validate 12 MMPI indexes of hostil¬ then cross-validated in new groups of ex¬
ity by contrasting groups of male criminals tremely assaultive, moderately assaultive,
who were classified as extremely assaultive, and nonassaultive prisoners. They eliminated
moderately assaultive, and nonassaultive. items that did not differentiate among these
None of these 12 scales correctly identified new groups of assaultive and nonassaultive
the extremely assaultive groups of criminals. prisoners, and the 31 (28 items on the MMPI-
In fact, Megargee and Mendelsohn found 2) remaining items became the Over-
that the extremely assaultive criminals were controlled-Hostility (O-H) Scale (see Appen¬
more likely to score significantly lower on dix A).
these scales (i.e., the extremely assaultive Megargee and associates (1967) did not
208 Chapter 5

suggest a specific cutting score to be used on MMPI, but it is not clear if this factor would
this scale. Instead, they recommended that be sufficient to invalidate the results.
investigators determine the most appropriate Werner, Becker, and Yesavage (1983)
cutting score in their own treatment setting found that the O-H scale was not correlated
based on which errors are more tolerable— with assaultiveness in psychotic, male psychi¬
false positives (identifying a person as over¬ atric inpatients. Most investigators, including
controlled and hostile who is not) or false Megargee and colleagues (1967) and Deiker
negatives (identifying a person as not being (1974), found only small mean differences
overcontrolled and hostile who actually is). (two items) between assaultive and non-
Prisoners whose crimes were judged to assaultive groups, so it is possible that even
reflect overcontrolled hostility scored higher minor variations in procedure may be suffi¬
on the O-H scale than prisoners whose crimes cient to obscure these differences. Investiga¬
were judged to reflect undercontrolled hostil¬ tors also have been inconsistent in whether or
ity (Megargee et al., 1967). Deiker (1974) not they distinguish between overcontrolled
found that the O-H scale was one of the few and undercontrolled hostility in their assaul¬
MMPI scales that could accurately identify tive groups; this inconsistency may further
male prisoners who were extremely assaultive. obscure the reported results.
He questioned whether a negative response bias Graham (1978) cited two unpublished
might account for the obtained results since studies that reported that blacks and females
two-thirds of the items on the O-H scale have score higher on the O-H scale than white
“false” as the deviant response. males on whom the scale was developed. Gra¬
Megargee and Cook (1975) demon¬ ham also noted an unpublished study that
strated, however, that O-H scales balanced found that the O-H scale differentiated fe¬
for “naysaying” (equivalent numbers of male assaultive prisoners from female non-
“true” and “false” deviant responses) assaultive prisoners. Both Leonard (1977)
yielded similar if not better results than the and Lester and Clopton (1979) reported that
original O-H scale, which indicated that a the O-H scale did not reliably distinguish be¬
negative response bias cannot explain the re¬ tween psychiatric patients who completed
sults obtained with the O-H scale. Lane and suicide and nonsuicidal psychiatric patients.
Kling (1979) also found that the O-H scale re¬ Additional research on the O-H scale is
liably discriminated between overcontrolled needed to validate its use in minority and fe¬
assaultive forensic psychiatric patients and male samples. The apparent finding that even
undercontrolled assaultive patients. minor procedural changes such as adminis¬
Quinsey, Maguire, and Varney (1983) tering the O-H scale in isolation may alter the
reported that murderers who scored high on effectiveness of the O-H scale suggests that it
the O-H scale were less assertive than murder¬ should be used cautiously in identifying spe¬
ers who scored low on this scale. They sug¬ cific individuals as overcontrolled and hos¬
gested that assertiveness training may be ben¬ tile. Investigators also will need to determine
eficial with persons who have high scores on the most efficient cutting score for the O-H
the O-H scale. scale in their treatment setting, as Megargee
Several studies found that the O-H scale and colleagues (1967) recommended.
does not discriminate between violent and Finally, Gearing (1979) pointed out the
nonviolent criminals (Hoppe & Singer, 1976; similarity between the behaviors characteris¬
Mallory & Walker, 1972; Truscott, 1990). tic of persons with 4-3 codetypes (Davis &
The first two of these studies administered Sines, 1971; Persons & Marks, 1971) and the
the O-H scale out of context of the entire behaviors expected of a high scorer on the
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 209

O-H scale. Research is needed to determine mal student populations, additional research
whether similar behaviors are being assessed would be necessary to validate the use of the
in these two instances. MMPI Do scale with adults and in psycho-
High scorers on the O-H scale are de¬ pathologic groups.
scribed as displaying excessive control of Research on the MMPI Do scale has
their hostile impulses and as being socially been exceedingly sparse. Olmsted and
alienated. They are reluctant to admit any Monachesi’s (1956) finding that firefighters
form of psychological symptoms, even achieved only slightly higher scores on the Do
though they are sometimes diagnosed as scale than students suggests that student
being psychotic. They are seen as being rigid norms could be generalized to adult samples.
and not displaying anxiety overtly. They may Olmsted and Monachesi also found that fire¬
candidates for assertiveness training. fighters with the rank of captain did not have
There are no reported correlates of low higher Do scores than regular firefighters;
scorers on the O-H scale. this finding led them to question the validity
The O-H scale shares few items with any of the Do scale. On the other hand, Knapp
of the standard validity and clinical scales or (1960) found that military officers achieved
the supplementary scales (see Tables 5-3 and higher Do scores than enlisted men, and
5-4). It is one of the few supplementary scales Knapp questioned whether dominance
that appears to be assessing something played the same role in firefighters as in mili¬
unique since it shares so few items with the tary personnel.
other scales. Age does not seem to affect Do scores in
adults (Swenson et al., 1973). Duckworth and
Anderson (1986) reported that college stu¬
Dominance (Do) Scale
dents have an average T score of 60 on the Do
Gough, McClosky, and Meehl (1951) devel¬ scale (i.e., current college students score
oped the Dominance {Do) scale by contrast¬ somewhat higher on Do than the students on
ing the item responses of both high school which Gough and associates [1951] devel¬
and college students who were judged by oped the scale). The lack of further research
their peers to be most and least dominant. on the Do scale makes conclusions difficult to
Their intent in the development of the Do draw. The Dominance scale on the California
scale was to identify strong, dominant, influ¬ Psychological Inventory (Gough, 1957),
ential persons who were able to take initiative which shares 28 of 60 items with the MMPI
and exercise leadership. Do scale, is one of the better validated scales
Gough and associates (1951) identified (Megargee, 1972). The MMPI Do scale also
60 items that differentiated subgroups of deserves serious attention. Research is needed
male and female students who were most and on the MMPI Do scale to determine whether
least dominant. Only 28 of these items are it is an adequate measure of interpersonal
found on the MMPI, and they comprise the dominance and to document the generaliza¬
MMPI Do scale. These 28 items (25 items re¬ tion of student norms and test correlates to
main on the MMPI-2) are keyed so that high adult samples.
scores indicate more dominant behaviors. High scorers on the Do scale (T scores of
The total 60 items identified by Gough and 65 or higher) are described as being able to
associates make up the Dominance scale on take charge of and responsibility for their
the California Psychological Inventory lives. They are poised, self-assured, and con¬
(Gough, 1957). Gough and colleagues noted fident of their own abilities. They address
that since their items were validated in nor¬ problems in a realistic, task-oriented fashion
210 Chapter 5

and feel adequate in their ability to overcome members, ignoring such considerations as
any obstacles that they may encounter. At friendliness, popularity, and so on. The re¬
higher elevations (T scores > 75) domi¬ sponsible person was defined as “one who
neering qualities may be seen, but these shows a ready willingness to accept the conse¬
qualities appear to be a function of which quences of his own behavior, dependability,
clinical scales are elevated. For example, el¬ trustworthiness, and a sense of obligation to
evations on Scales 4 (Psychopathic Deviate) the group” (Gough et al., 1952, p. 74). Such
and 9 (Hypomania) in conjunction with a a person would have a sense of commitment
high Do scale are more likely to indicate to the group and others, is dependable, and
domineering behavior (Duckworth & An¬ possesses integrity. They identified 56 items
derson, 1986). that reliably distinguished between students
Low scorers on the Do scale (T scores who were the most and least socially responsi¬
less than 45) have been less adequately inves¬ ble. These 56 items comprise the Social
tigated. These persons prefer to have others Responsiblity Scale on the California Psy¬
take responsibility for their lives. They fre¬ chological Inventory (Gough, 1957). Only 32
quently have high scores on the MMPI De¬ of these items are found on the MMPI, and
pendency (Dy) scale, which further substanti¬ these 32 items comprise the MMPI Re scale,
ates their reliance on others to meet their and 30 of these items are retained on the
needs. Since the Dependency scale is not rou¬ MMPI-2 Re scale.
tinely scored on the MMPI-2 even though 48 Gough and colleagues (1952) described
of its 57 items have been retained, research is high scorers on the Re scale as showing
needed to determine the correlates of low greater concern for social and moral issues,
scores on the Do scale. disapproving of favoritism, emphasizing car¬
The Do scale has limited item overlap rying one’s own share of duties and burdens,
with any of the standard validity and clinical having a sense of trust and confidence in the
scales. Those scales with which the Do scale world, and being poised and self-assured.
does share the most items have the items They did not describe low scorers on the Re
scored in the opposite direction on the Do scale. They noted that since their items were
scale: Scale 2 (Depression)—4 items; Scale 4 validated in normal student populations, ad¬
(Psychopathic Deviate)—4 items; Scale 7 ditional research would be necessary to vali¬
(Psychasthenia)—6 items; and Scale 0 (Social date the use of the MMPI Re scale with adults
Introversion)—5 items (see Table 5-3). The and in clinical settings.
Do scale also has limited item overlap with There has been virtually no research on
any of the other supplementary scales (see the Re scale. Knapp (1960) reported that Ma¬
Table 5-4); again, the items are scored in the rine Corps officers had higher scores than en¬
opposite direction on the Do scale: A (Anxi¬ listed men, and Olmstead and Monachesi
ety)—5 items; and MAC-R (MacAndrew Al¬ (1956) found that fire captains had higher
coholism—Revised)—6 items. scores than firemen, although these differ¬
ences were not statistically reliable.
Duckworth and Anderson (1986) suggest
Social Responsibility (Re) Scale
that high scorers on the Re scale are accepting
Gough and colleagues (1952) developed the of a previously held value system, whereas
Social Responsibility (Re) Scale by contrast¬ low scorers are changing away from such a
ing the item responses of both high-school value system. They also provide suggested in¬
and college students who were judged by terpretations of ranges of scores on the Re
their peers to be most and least responsible scale. Such interpretations should be used
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 211

cautiously until empirical research is avail¬ nician should see Wiggins (1973) for an over¬
able to validate them. view of this topic.
The Re scale shares few items with any The Mt scale has substantial item over¬
of the standard validity and clinical scales lap with a number of the standard clinical
(see Table 5-3) or the supplementary scales scales: Scale 1 (Hypochondriasis)—6 items;
(see Table 5-4). The Re scale does share 9 Scale 2 (Depression)—16 items; Scale 3 (Hys¬
items with the MAC-R (MacAndrew Alco¬ teria)—9 items; Scale 7 (Psychasthenia)—14
holism—Revised); these items are scored in items; Scale 8 (Schizophrenia)—9 items; and
the opposite direction on the two scales. The Scale 0 (Social Introversion)—6 items (see
Re scale also shares 5 items with the Do Table 5-3).
(Dominance) scale. The Mt scale also has substantial item
overlap with several of the other supplemen¬
tary scales: A (Anxiety)—12 items; PK (Post
College Maladjustment (Mt) Scale
Traumatic Stress Disorder—Keane)—10
Kleinmuntz (1960, 1961a) developed the Col¬ items; and PS (Post Traumatic Stress Disor¬
lege Maladjustment (Mt) scale by contrasting der—Schlenger)—13 items (see Table 5-4).
the item responses of 40 students who were The Mt scale is another of the numerous
referred to a university mental hygiene clinic MMPI-2 and MMPI scales that measure the
for routine mental health screening required first factor like the A scale. As noted above,
by their teacher’s college with 40 students there is little reason to score all of these first-
who were referred for treatment and who had factor scales since they tend to provide redun¬
remained in psychotherapy for at least three dant information.
sessions. He excluded Scale 5 (Masculinity-
Femininity) items since they were selected to
Gender Role Scales
differentiate between men and women. He
identified 43 items that separated these two Peterson (1989) developed separate gender
groups of students at the .01 level; 41 of these role scales for men (Gender Role-Masculine
items are retained on the MMPI-2. The mal¬ [GM\) and women (Gender Role-Feminine
adjusted student, who was defined by a score [GF]) on the MMPI-2. She included an item
of 15 or higher, was an ineffectual, pessimis¬ on one of the gender scales if it was endorsed
tic, procrastinating, anxious, and worried by a majority of one gender and by at least 10
person who tended to somatize and who percent fewer of the opposite gender. Nine
found life to be a strain much of the time. items from GM and 15 items from GF over¬
Kleinmuntz (1961b) found that his Mt lap with Scale 5 (Masculinity-Femininity) so
scale did not accurately separate college stu¬ it is apparent that the GM and GF scales are
dents with potential maladjustment from not totally redundant to Scale 5 despite the
those students who made a satisfactory ad¬ substantial item overlap.
justment during their first year of college. The availability of separate gender role
Subsequently, Kleinmuntz (1963) developed scales for men and women avoids some of the
a computerized system based in part on the pitfalls of the bipolar Scale 5 (see Chapter 4)
Mt scale to differentiate maladjusted and and allows for the determination of the
well-adjusted students. His use of the Mt client’s score separately on each scale. A
scale and the MMP1 in general quickly quick perusal of the items on GM and GF(sqq
evolved into the issue of automated versus Appendix A) will reveal that a very stereo¬
clinical judgment in the identification of mal¬ typic characterization of each gender has
adjusted college students. The interested cli- been produced by this method of selecting the
212 Chapter 5

items for each scale. The items on each scale scale by contrasting the item responses of
are so stereotypic for each gender that it does healthy Vietnam-era veterans with veterans
not seem likely that a person would score with Post Traumatic Stress Disorder who did
high on both scales and be classified as an¬ not have any other psychiatric diagnosis. Their
drogynous. scale consists of 60 items on the MMPI-2.
No research is available that reports the The PK and PS scales share 26 items
correlates of these two scales. (i.e., almost one-half of their items). The PK
The GM and GF scales have little item and PS scales have substantial item overlap
overlap with any of the standard validity and with a number of the standard validity and clin¬
clinical scales (see Table 5-3). The largest ical scales (see Table 5-3). The PK scale shares
overlap is between the GM scale and Scale 0 11 items with Scale 2 (Depression); 9 items with
(Social Introversion), sharing eight items. Scale 3 (Hysteria); 11 items with Scale 4 (Psy¬
The GM and GF scales also have little item chopathic Deviate); 17 items with Scale 7
overlap with any of the supplementary scales (Psychasthenia); and 19 items with Scale 8
(see Table 5-4), The GM scale shares 7 items (Schizophrenia).
with the Es (Ego Strength) scale, and the GF The PS scale has an almost identical pat¬
scale shares 6 items with the Re (Social Re¬ tern of item overlap. The PS scale shares 11
sponsibility) scale. items with Scale 2; 10 items with Scale 5; 9
items with Scale 4\ 17 items with Scale 7; and
27 items with Scale 8.
Post Traumatic Stress Disorder Scales
The PK and PS scales also have substan¬
Keane, Malloy, and Fairbank (1984) devel¬ tial item overlap with the other supplemen¬
oped their Post Traumatic Stress Disorder tary scales. The PK scale shares 9 items with
(PK) scale by contrasting the item responses Scale A (Anxiety) and 10 items with Scale Mt
of 100 male veterans who had Post Trau¬ (College Maladjustment), whereas the PS
matic Stress Disorder with 100 male veterans scale shares 14 items with Scale A and 13
who had psychiatric diagnoses other than items with Scale Mt. Both the PK and PS
Post Traumatic Stress Disorder. They identi¬ scales share 10 items with the Es (Ego
fied 49 items from the first 400 items of Form Strength) with the items scored in the oppo¬
R of the MMPI that differentiated the two site direction on the Es scale.
groups at the .001 level. They found that a The PK and PS scales are saturated with
cutting score of 30 was optimal for separating first-factor variance as measures of general
the two groups with a hit rate of 82 percent. maladjustment and emotional distress rather
The MMPI-2 retains 46 of these 49 items. than Post Traumatic Stress Disorder per se,
Hit rates in cross-validation studies of as can be seen clearly in the extensive item
the PK scale have ranged from 38 percent overlap described above. Clinicians should be
(Gayton, Burchstead, & Matthews, 1986) to very cautious of diagnosing any client as hav¬
80 percent (Schlenger & Kulka, 1987) and av¬ ing Post Traumatic Stress Disorder based on
eraged in the 70 to 75 percent range (Penk, their scores on these two scales because of the
Keane, Robinowitz, Fowler, Bell, & Finkel- significant amount of first-factor variance in
stein, 1988). Denny, Robinowitz, and Penk both scales. Research is needed that outlines
(1987) and Penk and colleagues (1988) have the relative advantages and disadvantages of
provided recent reviews of the research on the PK and PS scales. In the interim little is to
various PTSD scales. be gained by scoring and using both of these
Schlenger and Kulka (1987) developed scales since they are so redundant and satu¬
their Post Traumatic Stress Disorder (PS) rated with first-factor variance.
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 213

ADDITIONAL MMPI Age does not appear to affect perfor¬

SUPPLEMENTARY SCALES mance on the Dy scores (Swenson et al.,
1973), but women tend to score higher than
men (Birtchnell & Kennard, 1983; Evans,
1984a). Duckworth and Anderson (1986) re¬
Dependency (Dy) Scale
ported that college students have an average
Navran (1954) developed the Dependency T score of 44 on the Dy scale.
(Dy) scale by having 16 judges examine all At best, the present research provides
MMPI items and select those items that they modest support for the Dy scale. Since depen¬
thought would be related to dependency. dency is an important dimension of interper¬
These judges identified 157 items, which were sonal behavior that could interact with and
administered to two groups of 50 psychiatric affect the success of treatment interventions,
patients. Through internal consistency proce¬ additional research on the Dy scale is needed
dures, Navran then identified those items to document its continued use. It appears
that correlated most highly with one another particularly important to develop a depen¬
in these two samples. dency scale that measures the full range of de¬
The 57 items (48 items on the MMPI-2) pendent behaviors and that is not con¬
that resulted from this procedure comprise founded with measures of depression.
the Dy scale (See Appendix A for the MMPI- High scorers on the Dy scale are de¬
2). The items on the Dy scale are keyed so scribed as being dependent, submissive, and
that high scores indicate more dependency. passive. They also are frequently seen as
Dy has moderate positive correlations with being maladjusted. The dependency that they
Scales 2 (Depression), 6 (Paranoia), and 8 are manifesting may either be characterologic
(Schizophrenia), and moderate negative cor¬ in nature or reflect their need for help in deal¬
relations with the K scale. Navran’s psychiat¬ ing with situational crises. Ambivalence
ric patients scored higher on the Dy scale than about dependency may be present when they
the Minnesota normative group, who scored simultaneously elevate the Dominance (Do)
higher than a group of graduate students scale or when they do not display dependent
tested by Navran. Within the psychiatric behaviors on other tests or in their interac¬
sample nonparanoid schizophrenics scored tions with others. In these circumstances they
higher than paranoid schizophrenics. are likely to display passive-aggressive or pas¬
Little systematic research has been con¬ sive-dependent behaviors.
ducted on the Dy scale. High scores on the Dy Low scorers on the Dy scale are described
scale has been found to be related to peer rat¬ as being independent and self-reliant.
ings and self-ratings of dependency (Zucker-
man, Levitt, & Lubin, 1961); greater chronic-
Low-Back Pain (Lb) Scale
ity in welfare recipients (Pruitt & Van de
Castle, 1962); and being female, having neu¬ Hanvik (1949, 1951) developed the Low-
rotic symptoms, being a psychiatric patient, Back Pain (Lb) scale by contrasting the item
being depressed, and attempting suicide responses of 30 male inpatients with verifi¬
(Birtchnell & Kennard, 1983). Birtchnell and able organic causes of their low-back pain
Kennard suggest that these correlates of the and the responses of 30 male inpatients with
Dy scale may largely reflect covariance with no clear organic causes for their low-back
depression (i.e, the Dy scale is actually mea¬ pain. The two groups were matched on age,
suring some aspects of depression rather than socioeconomic status, marital status, intelli¬
dependency per se). gence, and race.
214 Chapter 5

Hanvik found 25 items that discrimi¬ In France, Pichot and colleagues (1972)
nated the two groups. He found that the opti¬ developed a 63-item scale that separated fe¬
mal cutting score for identifying functional male patients with functional low-back pain
(nonorganic) low-back pain was a raw score from nonhospitalized persons without any
of 11 or higher; this cutting score yielded an pain complaints. Several studies have found
accuracy score of 80 percent on cross-valida¬ that the Lb scale in conjunction with the Pic¬
tion. Hanvik (1951) reported that the mean hot and colleagues scale was somewhat suc¬
profile for the clients with functional low-back cessful in separating clients with functional
pain had simultaneous elevations above a T low-back pain from clients with organic
score of 70 on Scales 1 (Hypochondriasis) and causes of their low-back pain. Until the Pic¬
3 (Hysteria), whereas Scale 2 (Depression) was hot and associates scale receives further vali¬
about a T score of 60 (i.e., a conversion “V” dation in American samples, however, it
profile [see Chapter 4 for a description of this should be used cautiously. It also seems that
profile]). the Lb scale either by itself or in conjunction
Hanvik (1951) also reported that experi¬ with the Pichot and colleagues scale may
enced judges could reliably discriminate the have limited usefulness in determining
profiles between these two groups, although whether a specific client’s pain is primarily
some judges were more accurate at the task functional in nature.
than others. Finally, Hanvik suggested that Several factors may have confounded
men with functional low-back pain demon¬ the previous research on the Lb scale. First,
strated a slower return to normal functioning investigators have used different cutting
after surgery than those whose low-back pain scores on the Lb scale; some have used a raw
had organic causes. score of 11 (T score of 57) as suggested by
Dahlstrom (1954) found similar results Hanvik (1949), while others have used a T
in patients referred to neurosurgery for score of 70. Few investigators have identified
chronic low-back pain. Patients with no and cross-validated the specific cutting score
known physical cause for their complaints that classifies clients most accurately in their
scored higher on the Lb scale and recovered particular setting.
more slowly after surgery than those with a Second, investigators have paid little at¬
known physical cause. tention to the demographic factors that may
More recent research has obtained less influence how a client perceives and tolerates
consistent results. Most investigators have pain (cf. Weisenberg, 1977). The classifica¬
found small mean differences on the Lb scale tion of clients into functional or organic sub¬
between groups of patients with functional or groups on the basis of whether there are de¬
mixed functional and organic low-back pain monstrable organic reasons for their low-back
and patients with organic low-back pain, with pain appears to assume that all clients react to
the former groups scoring higher (Calsyn, and tolerate pain in the same manner. Almost
Louks, & Freeman, 1976; Freeman, Calsyn, & all investigators of clients with low-back pain
Louks, 1976; Louks, Freeman, & Calsyn, have concluded that there is not a single pro¬
1978). None of these investigators, however, file type that is characteristic of these individ¬
have found the Lb scale capable of reliably uals. Profiles with conversion “V”s are fre¬
distinguishing between individual patients quently seen in clients with functional
with functional or organic low-back pain low-back pain as well as in clients who have
(Haven & Cole, 1972; Pichot, Perse, Le- demonstrated organic causes for their low-
beaux, Dureau, Perez, & Rychewaert, 1972; back pain.
Sternbach, Wolf, Murphy, & Akeson, 1973; Finally, investigators have classified cli¬
Towne & Tsushima, 1978). ents into categories of a functional, mixed, or
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 215

organic basis for their low-back pain without of items as the complete set but which elimi¬
consideration of either the reliability of these nated item overlap” (p. 247). A comprehen¬
categories or their meaningfulness for treat¬ sive review of these Scales can be found in
ment intervention. Future research on low- Morey and Smith (1988).
back pain and other forms of chronic pain Only two validation studies of the Per¬
may be more fruitful if investigators delin¬ sonality Disorder Scales have been published
eate subgroups among pain patients on the to date. Morey, Blashfield, Webb, and Jewell
MMPI and then determine what forms of (1988) found that these Scales discriminated
treatment interventions are most appropriate effectively between patients with a specific
for each subgroup (Bradley, Prieto, Hopson, personality disorder diagnosis, normal indi¬
&Prokop, 1978; Bradley, Prokop, Margolis, viduals, and patients with other personality
& Gentry, 1978). Prokop (1988) and Snyder disorder diagnoses. Dubro, Wetzler, and
(1989) have provided a recent review of the Kahn (1988) found that the Personality Dis¬
literature on the use of the MMPI in patients order Scales were only successful in identify¬
with chronic pain that should be consulted by ing patients with avoidant personality disor¬
clinicians. der from four individual disorders (avoidant,
borderline, dependent, and histrionic).
Overall, the Personality Disorder Scales
Personality Disorder Scales
had a sensitivity of 78 percent (patients with a
Morey, Waugh, and Blashfield (1985) devel¬ personality disorder/all patients) and a speci¬
oped separate MMPI Personality Disorder ficity of 67 percent (patients without a per¬
Scales for each of the 11 DSM-I1I (American sonality disorder/all patients) for the pres¬
Psychiatric Association, 1980) personality ence of any personality disorder. It is difficult
disorders: Histrionic; Narcissistic; Border¬ to draw any firm conclusions based on the
line; Antisocial; Dependency; Compulsive; limited research on the Morey and associates
Passive-Aggressive; Paranoid; Schizotypal; (1985) Personality Disorder Scales. Clini¬
Avoidant; and Schizoid. cians can use the scores from these Scales in
Morey and associates developed their conjunction with the other data from the
Personality Disorder Scales in a similar man¬ MMPI in order to assess patients with per¬
ner as Wiggins (1966) devised his Content sonality disorders. Additional research will
scales, so their methodology will not be de¬ be necessary to determine the usefulness of
scribed here. The MMPI items found on each these Scales.
Personality Disorder Scale as well as the The Morey and colleagues (1985) Per¬
mean and standard deviation on each Scale sonality Disorder scales are virtually intact on
for males and females in the original Minnesota the MMPI-2. No scale has lost more than two
normative group are provided in Dahlstrom, items and eight of the nonoverlapping scales
Welsh, and Dahlstrom (1972, 1975). and six of the overlapping scales have lost no
Since many items were common to more items.
than one Scale, Morey and colleagues devel¬
oped both a complete and a nonoverlapping
Tryon, Stein, and Chu (TSC)
version of these scales. They created the non¬
Cluster Scales
overlapping version of each Scale by assign¬
ing overlapping items to the one scale with Cluster scales on the MMPI based on an ex¬
which these items had their highest correla¬ tensive cluster analysis of the entire MMPI
tion. “As a result, two sets of scales were de¬ item pool were developed by Tryon, Stein,
veloped: a complete set and a nonoverlapping and Chu (Stein, 1968). By analyzing re¬
set, which contained the same total number sponses to individual items rather than scale
216 Chapter 5

scores, they avoided the problem of items The items in each cluster scale are listed in
that are scored on more than one scale. For Dahlstrom, Welsh, and Dahlstrom (1972, 1975).
the 10 clinical scales the percentage of over¬ High scores on each cluster scale indicate
lapping items ranges from 17 to 83 percent. the presence of the behavior for which the
This item overlap introduces a source of in¬ scale is named, whereas low scores indicate
terrelatedness among the clinical scales, the absence of this behavior. For example, a
which is a problem in cluster analysis, factor high score on the / (Social Introversion) clus¬
analysis, or any other statistical procedure ter scale indicates that the client has endorsed
that assumes independent sources of vari¬ the items like a person who is socially intro¬
ance. verted, and a low score on this scale would in¬
For their analysis Tryon, Stein, and Chu dicate that the client has not endorsed the
used three samples totaling 310 males: 70 items like a socially introverted person.
Veterans Administration (VA) hospital out¬ Clusters /, B (Bodily Symptoms), and S
patient schizophrenics, 150 VA outpatient (Suspicion) are the pivots or most indepen¬
neurotics diagnosed as anxiety reaction, and dent clusters with modest intercorrelations
90 military officers. The officers and the VA (the highest correlation of .33 was between I
patients were matched for age and education. and B). The other four clusters are highly cor¬
Items were selected for a cluster scale on related with all the clusters. The reliabilities
both statistical and content bases. Through of these clusters range from .85 to .94 with a
cluster analysis, Tryon, Stein, and Chu calcu¬ median correlation of .92.
lated communality estimates for each item There has been virtually no reported re¬
within the three samples. They eliminated 317 search on the TSC scales. Graham (1987)
items for reasons of trivial communality and provided some tentative interpretations of
an additional 57 items for being rationally high scores on the TSC scales based on un¬
ambiguous in relation to the general content published research by Boerger (1975). Boer-
meaning of items within a cluster scale. Seven ger was unsuccessful in identifying correlates
cluster scales containing 192 items were derived for low scores on these scales. Since low
following these procedures (see Table 5-17). scores indicate the absence of the qualities

TABLE 5-17 Description of TSC Cluster Scales

Abbreviation Number of Items Description

/ 26 Social Introversion versus Interpersonal Poise

and Outgoingness
B 33 Bodily Symptoms versus Lack of Physical Complaints
S 25 Suspicion and Mistrust versus Absence of Suspicion
D 28 Depression and Apathy versus Positive and Optimis¬
tic Outlook
R 21 Resentment and Aggression versus Lack of Resent¬
ment and Aggression
A 23 Autism and Disruptive Thought versus Absence
of Such Disturbance
T 36 Tension, Worry, Fear versus Absence of Such Com¬
plaints

Note: The descriptions are from Stein (1 968).

 Supplementary Scales, Content Scales, Critical Items, and Short Forms 217

manifested by high scores, it is not surprising The clinical literature on the MMPI is re¬
that there are not consistent correlates of low plete with references to specific scales or
scores. This situation may be similar to that codetypes that are frequently associated with
on Scale 1 (Hypochondriasis), where high suicide. For example, significant elevations
scorers are fairly homogeneous but low scor¬ (T scores > 70) on Scales 2 (Depression)
ers are heterogeneous with the absence of so¬ and/or 7 (Psychasthenia) are described as in¬
matic complaints as their only common char¬ creasing the likelihood of suicide attempts
acteristic. (cf. Carson, 1969; Dahlstrom et al., 1972;
The TSC scales need more empirical in¬ Graham, 1987).
vestigation in order to determine their func¬ Dahlstrom and colleagues (1972) also
tion with various psychopathologic groups in noted that when the client has a Spike 2 pro¬
a variety of settings. file (Scale 2 [Depression] is the only clinical
scale elevated above a T score of 70) but de¬
nies depressive thoughts and feelings, the risk
Suicide Scales
of suicide is increased. On the other hand,
Numerous attempts have been made to use numerous studies have found no difference in
the MMPI to predict the occurrence of sui¬ Scale 2 scores of suicidal and nonsuicidal in¬
cide and/or suicide threats through supple¬ dividuals (Clopton & Jones, 1975; Farberow,
mentary scales such as the Suicide Threat 1956; Simon & Gilberstadt, 1958). Suicidal
scale (Farberow & Devries, 1967); profile and nonsuicidal persons also do not differ
analysis (cf. Clopton, Pallis, & Birtchnell, consistently on any of the other standard
1979; Leonard, 1977); and clinical judgment MMPI scales (Clopton, 1979b; Clopton,
(Clopton & Baucom, 1979). Clopton (1979b) Post, & Larde, 1983; Spirito, Faust, Myers, &
provided a comprehensive review of the use Bechtel, 1988; Watson, Klett, Walters, &
of the MMPI in predicting suicide, a review Vassar, 1984).
that the interested reader should consult. Both Leonard (1977) and Clopton and
The initial hurdle faced in predicting sui¬ associates (1979) found that multivariate sta¬
cide with the MMPI or any other assessment tistical procedures, which simultaneously
device is the extremely low frequency with consider scores from a number of the clinical
which suicide occurs in most populations. scales, could reliably distinguish female suici¬
Even if a test were 75 percent accurate in pre¬ dal groups from control groups but not male
dicting suicide, which would be unusually suicidal groups from control groups. Leon¬
high for most tests, a more accurate predic¬ ard (1977) did not cross-validate her results;
tion can be made by simply stating that all pa¬ Clopton and associates (1979) found upon
tients will be nonsuicidal since the frequency cross-validation that the percentage of female
of suicide is less than 25 percent in any group. patients correctly classified decreased from
Consequently, any index of suicide will yield 36 to 28 percent.
a large number of false positives (clients iden¬ Clopton and colleagues (1979) reported
tified as suicidal who are nonsuicidal) be¬ the following relationship between Scales 1
cause of this low frequency of occurrence. (Hypochondriasis) and 2 (Depression) in
Although it would seem that false posi¬ their study, which deserves further investi¬
tives are of less concern than false negatives gation. Among clients with 7-8/8-7 code¬
(clients who are identified as nonsuicidal who types, the relative elevation of Scales 1 and
commit suicide), the ethical and practical im¬ 2 was significantly associated with whether
plications of falsely identifying a client as sui¬ the client had attempted suicide. Scale 1
cidal also must be considered (Rosen, 1954). was greater than Scale 2 in 60 percent of the
218 Chapter 5

nonsuicidal clients, whereas Scale 2 was identifying some subgroups of suicidal cli¬
greater than Scale 1 in 64 percent of the suici¬ ents.
dal clients.
Clopton and associates (1983) did find
Psychotic-Neurotic Indexes
that discriminant analysis could reliably dis¬
tinguish between patients who recently at¬ Numerous MMPI indexes, those that in¬
tempted suicide and nonsuicidal patients that volve combining various validity and/or
held up on cross-validation. However, nei¬ clinical scales in a linear or a configural pat¬
ther the original results (58.8 percent for fe¬ tern, have been proposed as an additional
males; 63.3 percent for males) nor the cross- means of determining how a client should be
validation results (53.9 percent for females; diagnosed. Peterson (1954) developed six di¬
58.6 percent for males) were particularly im¬ agnostic signs that he found were character¬
pressive. istic of psychotic (schizophrenic) MMPI
Clopton and Baucom (1979) presented patterns:
six psychologists who had extensive experi¬
ence in MMPI interpretation with the pro¬ 1. Four or more clinical scales are greater
files of male suicidal and nonsuicidal clients. than a T score of 70.
None of the psychologists could reliably 2. The F scale is greater than a T score of
identify the clients in each group. The psy¬ 64.
chologists’ ratings of eight variables thought
3. Scales 6 (Paranoia), 8 (Schizophrenia),
to be related to suicide risk also did not differ and 9 (Hypomania) are greater than
for the suicidal and nonsuicidal clients.
Scales 1 (Hypochondriasis), 2 (Depres¬
Thus, any method using the MMPI or
sion), and 3 (Hysteria).
the MMPI-2—whether it involves single
4. Scale 2 is greater than Scales 1 and 3.
scales, profile analysis, supplementary scales,
or item analysis—appears disappointing in 5. Scale 8 is greater than Scale 7 (Psychas-
the prediction of suicide. The few studies that thenia).
reported statistical significance in identifying 6. Scales 6 or 9 is greater than a T score of
female suicidal groups appears to have lim¬ 70.
ited clinical utility because of the small mean
differences between groups on the individual The presence of three or more of these
scales. signs was characteristic of a psychotic profile
As Clopton (1979b) pointed out, the re¬ pattern.
search question of most interest is whether Taulbee and Sisson (1957) developed 16
the MMPI can increase the accuracy of iden¬ signs, which involve comparison of one clini¬
tifying suicidal clients, not whether the cal scale with another clinical scale, as an
MMPI by itself is sufficient to predict sui¬ index of whether the profile suggests a neu¬
cide. Some investigators appear to assume rotic or a psychotic disorder (see Table 5-18).
that both suicide gestures or attempts and ac¬ Each of these signs is scored as being present
tual suicides result from a single cause with¬ or absent. The presence of 13 or more of
out fully appreciating the multitude of fac¬ these signs suggests a neurotic pattern,
tors that lead the client to attempt or whereas the presence of 6 or fewer signs sug¬
commit suicide. Future research should dis¬ gests a schizophrenic pattern.
criminate among the various causes and Meehl and Dahlstrom (1960) developed
types of suicide to determine whether spe¬ a set of complex configural rules (Meehl-
cific scale patterns can assist in successfully Dahlstrom rules) to classify a MMPI profile
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 219

TABLE 5-18 Taulbee-Sisson Signs for of the most accurate in making this distinction
Neurotic Patterns3 between neurotic and psychotic profiles.4
Of course, these findings that a linear
Sign Sign
model can accurately estimate a clinician’s
Scale 1 > Scale 3 Scale 2 > Scale 6 judgments of whether an MMPI profile is
Scale 7 > Scale 4 Scale 3 > Scale 4 neurotic or psychotic does not necessarily
Scale 1 > Scale 5 Scale 3 > Scale 5 imply that the clinician makes decisions in
Scale 7 > Scale 6 Scale 3 > Scale 6 this manner; they do suggest that the clinician
Scale 1 > Scale 7 Scale 3 > Scale 9
Scale 7 > Scale 8 Scale 7 > Scale 5 should give greater considerations to such
Scale 7 > Scale 9 Scale 7 > Scale 6 models in making judgments. Wiggins (1973)
Scale 2 > Scale 4 Scale 7 > Scale 8 provided an extensive examination of the is¬
sues involved in clinical prediction, which the
Note: The signs are from faulbee and Sisson
interested clinician is urged to read.
(1957).
aEach sign is scored as present or absent. Little research has been conducted to
Scores from 1 3 to 16 are indicative of a neu¬ validate these indexes to discriminate be¬
rotic pattern; scores from 0 to 6 are indicative tween neurotic and psychotic MMPI profiles.
of a psychotic pattern. Meehl (1959) found that both the Meehl-
Dahlstrom rules and the Taulbee-Sisson signs
were better than individual clinicians in deter¬
as neurotic, psychotic, or indeterminate. mining whether an MMPI profile should be
(These rules are found in Dahlstrom et al. classified as neurotic or psychotic. Winter
[1972].) Henrichs (1964, 1966) expanded the and Stortroen (1963) reported that the Peter¬
Meehl-Dahlstrom rules to include another son signs were more accurate than the Meehl-
category, character or behavior disorders. Dahlstrom rules or the Taulbee-Sisson signs
Three studies are particularly relevant to in discriminating among MMPI profiles from
the use of linear or configural indexes to normals, patients with physical illness, and
make decisions about MMPI profile pat¬ hospitalized schizophrenics.
terns. In a task examining linear and con¬ Since neither the Meehl-Dahlstrom rules
figural models of clinical judgment, Wig¬ nor the Taulbee-Sisson signs were designed to
gins and Hoffman (1968) asked experienced identify normal profiles or profiles of pa¬
clinicians to sort MMPI profiles on a distri¬ tients with physical illness, it is unclear what
bution from neurotic through normal to meaning to assign to the superiority of the
psychotic. Slightly more than half of their Peterson signs in their study.
clinicians (16/29) appeared to use configu¬ Giannetti, Johnson, Klingler, and Wil¬
ral cues in making their judgments, but a liams (1978) found that the Goldberg index
linear model could accurately estimate their was superior to the Meehl-Dahlstrom rules,
judgments. the Taulbee-Sisson signs, and the Peterson
In two similar studies Goldberg (1965, signs in discriminating neurotic from psy¬
1969) found that a linear model accounted chotic MMPI profiles. They also found that
for most of the variance in clinicians’ judg¬ the Meehl-Dahlstrom rules and the Taulbee-
ments of whether an MMPI profile should be Sisson signs achieved less than chance accu¬
classified as neurotic or psychotic, and a lin¬ racy in making this discrimination.
ear model was superior to all other models in The paucity of research in this area
estimating these judgments. Goldberg sug¬ makes it difficult to draw definitive conclu¬
gested a linear index [(Scale L + Scale 6 + sions. It does appear that the Goldberg index
Scale 8) — (Scale 3 + Scale 7 )] as being one may be superior to the other indexes in dis-
220 Chapter 5

criminating neurotic from psychotic MMPI TABLE 5-19 Content Areas for Caldwell
profiles, although additional research is MMPI-2 Critical Items
needed to investigate their utility thoroughly.
Number of
Items Content Areas

1 1 Distress and Depression

CRITICAL ITEMS
5 Suicidal Thoughts
Despite the inherent difficulties in under¬ 10 Ideas of Reference, Persecution,
standing responses to individual MMPI items and Delusions
(difficulties that provided the original impe¬ 9 Peculiar Experiences and Halluci¬
tus for the empirical selection of items on the nations
MMPI), clinicians have been unwilling to ig¬ 6 Sexual Difficulties
nore the information that might be contained 5 Authority Problems
in those responses. The original set of indi¬ 3 Alcohol and Drugs
vidual items, which were thought to require 7 Family Discord
careful scrutiny if answered in the deviant 10 Somatic Concerns
direction, was rationally or intuitively se¬
lected by Grayson (1951). These 38 MMPI Note: The content area names are from Cald¬
items were selected as being highly indicative well (1969).
of severe psychopathology and have accord¬
ingly been considered “stop” or “critical”
items. sis group are used, there are 67 items that are
Caldwell (1969) developed on a rational “critical” for these six crisis situations.
basis a more comprehensive set of 68 Examination of these 67 items reveals
MMPI items to identify severe, generally that the item content generally relates directly
psychotic symptomatology. The content to the crisis situations. Thus, it appears that
areas of his 66 MMPI-2 items are given in these patients were both willing and able to
Table 5-19. reveal accurate information about them¬
Koss and Butcher (1973) and Koss, selves.
Butcher, and Hoffmann (1976) examined the All but three of the original Koss and
MMPI items endorsed by patients in crisis sit¬ Butcher (1973) critical items have been re¬
uations. They obtained MMPI responses tained on the MMPI-2, although numerous
from 723 male Veterans Administration hos¬ items were added to the various item groups
pital patients in six separate crisis situations: (see Appendix C). These item changes can be
acute anxiety, depressed-suicidal ideation, seen in Table 5-20.
threatened assault, situational stress due to The rationale for these changes has not
alcoholism, mental confusion, and persecu¬ been provided other than to state that empir¬
tory ideas. ical criteria were used to add items to the de¬
The number of items that significantly pressed suicidal ideation and situational
discriminated each crisis group from a non¬ stress due to alcoholism item groups (Butcher
crisis control group ranged from 10 items in et al., 1989, p. 44). The latter point is particu¬
the threatened assault group to 89 items in the larly unclear since the original MMPI situa¬
depressed-suicidal ideation group. If all of tional stress due to alcoholism item group
the items that discriminated a crisis group had 15 items and this group has only 7 items
from the control group at a probability less on the MMPI-2.
than .001 and that are unique to only one cri- Koss and Butcher (1973) also asked eight
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 221

TABLE 5-20 Changes in Items on the Koss and Butcher Critical Items

Number of Stems Items Dropped Number of Items

Crisis Area on the MSVSPS on the MMPI-2 on the IVSSVSPi-2

Acute Anxiety 9 2 1 7
Depressed-Suicidal Ideas 25 0 22
Threatened Assault 3 0 5
Situational Stress
Due to Alcoholism 1 5 1 7
Mental Confusion 3 0 1 1
Persecutory Ideas 12 0 16
TOTAL 67 3 78

Note: Names and number of items are from Koss and Butcher (1973) and Butcher, Dahlstrom,
Graham, Tellegen, & Kaemmer (1989).

clinical judges to select those MMP1 items though a client in a crisis may endorse an
that would be relevant (face valid) to the six item, it may still be true that normal clients
crisis groups. This procedure resulted in 96 endorse the items more often than not. For
items that four of the eight judges agreed example, perhaps 75 percent of a crisis group
would be relevant. When Koss and Butcher endorsed a specific item, whereas 60 percent of
checked the responses of the patients to these a normal group endorsed the same item. This
items, they found that 67 items actually dis¬ difference could be statistically significant; if
criminated a crisis group from the control an individual endorsed the item, however,
group. The fact that 24 percent (23/96) of that would not necessarily indicate that the
these items did not significantly discriminate individual is in a crisis since 60 percent of a
a crisis group from the control group illus¬ normal sample also endorse that item. This ques¬
trates the problem of generating critical item tion needs to be explored on the items identified
lists on a rational basis at least for these six by Koss and Butcher as well as sets of critical
crisis situations. items identified by other investigators.
More importantly, Koss and Butcher Lachar and Wrobel (1979) developed a
could not find any apparent differences be¬ set of critical items designed to be face-valid
tween face-valid items that were empirically descriptors of psychological concerns. They
related to a crisis situation and those face first identified 14 categories of symptoms
valid items that were not. Koss and col¬ that summarized problems that motivate peo¬
leagues (1976) indicated that the Grayson ple to seek psychological treatment and that
(1951) and Caldwell (1969) critical items are help the clinician make diagnostic decisions.
inadequate samples of behavior of potential Then 14 clinical psychologists read each
interest to the clinician and that better critical MMPI item and nominated items that would
items could be identified. Since most of these be face-valid indicators of psychopathology
items are face valid for a specific crisis situa¬ in one of these 14 categories. Items nomi¬
tion, the clinician could use these 67 items to nated by at least 6 of the 14 clinicians, to¬
identify significant problem areas that war¬ gether with the Grayson (1951) and Caldwell
rant further exploration. (1969) critical items, were empirically vali¬
A question that needs to be addressed, dated by contrasting item response frequen¬
however, is the frequency with which these cies for normals and psychiatric samples
items are endorsed by normal clients. Even matched for sex and race.
222 Chapter 5

Lachar and Wrobel were able to validate clinical interpretations of the MMPI and
130 of the 177 items nominated. After elimi¬ MMPI-2, there is little information on what
nating 19 items that were highly duplicative meaning or clinical importance to assign to a
of item content in other items on the list, they deviant response to a specific critical item.
arrived at a final list of 111 critical items. The Most clinicians seem to assume that any devi¬
content areas for these items appear in Table ant response is worthy of further investiga¬
5-21. Lachar and Wrobel reported that 80 tion, even without any information on the
percent of the 111 items reliably differenti¬ base rate (frequency) with which a given crit¬
ated normal from psychiatric samples for ical item is endorsed by normal or pathologic
adult males, females, blacks, and whites. samples. In addition, until recently, the indi¬
They concluded that responses to these criti¬ vidual critical items had not been validated to
cal items could serve as accurate representa¬ determine whether deviant responses were
tions of the client’s psychological concerns. empirically related to the actual behavior of
All but four of the Lachar and Wrobel the individual.
(1979) critical items have been retained on the Regarding the Grayson (1951) critical
MMPI-2 (see Appendix C). Two items were items, Saunders and Gravitz (1974) found
dropped from the Sexual Concern and Devia¬ that normal females were more likely to en¬
tion group, and one item from Deviant dorse items reflecting internal conflict or
Thinking and Experience and Substance stress, whereas normal males were more
Abuse (see Table 5-21). likely to endorse items reflecting acting-out
Although critical item lists are widely behaviors. Newton (1968) reported that psy¬
employed in both automated and individual chiatric samples endorsed on the average
about 9 of the 38 Grayson critical items.
Gravitz (1968) reported that normal adults
infrequently endorse any of the items, al¬
TABLE 5-21 Content Areas for the Lachar
though 5 of the 38 items were endorsed by
and Wrobel MMPI-2 Critical Items
more than 10 percent of his sample. The fre¬
Number of quency of endorsement of these items ranged
Items Content Areas from .5 percent to 12.5 percent in males and
from .9 percent to 29.3 percent in females.
Psychological Discomfort
These results certainly question the appropri¬
1 1 Anxiety and Tension
16 Depression and Worry ateness of considering these items critical.
6 Sleep Disturbance Similarly, the university student sample
Bodily Distortions described in Greene (1980), which can be as¬
1 5 Deviant Beliefs sumed to be relatively normal, endorsed an
10 Deviant Thinking and Expe- average of six of the Grayson critical items.
rience
Only 36.4 percent of this student sample en¬
Characterologic Adjustment dorsed three or fewer items. Again, it seems
3 Substance Abuse
that the high frequency of endorsement of the
9 Antisocial Attitude
4 Family Conflict Grayson critical items by normal samples se¬
4 Problematic Anger riously questions how “critical” these items
6 Sexual Concern and Devia- actually are. In addition, none of the above
tion studies provides any empirical validation of
23 Somatic Symptoms
the Grayson critical items.
Note: The content area names are from Lachar In comparing the responses of their six
and Wrobel (1979). crisis groups with the normal group, Koss
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 223

and associates (1976) found substantial over¬ the Lachar and Wrobel (1979) critical items.
lap between the distributions of the total Finally, his normal adults endorsed an aver¬
number of their deviant responses to the age of 16 of the Koss and Butcher critical
Grayson critical items. Consequently, they items and 12 of the Lachar and Wrobel criti¬
could not identify any cutting score that ac¬ cal items.
curately classified normal and crisis samples. Holmes, Sabalis, Chestnut, and Khoury
Thus, there is little evidence that the Grayson (1984) reported that parents of children re¬
critical items are useful either in identifying ferred for outpatient psychiatric services sig¬
behaviors that need attention or in classifying nificantly increased the number of critical
clients as normal or pathologic. items that they endorsed from the period of
No research has been published on the 1970-1974 to 1975-1979.
Caldwell (1969) critical items. Caldwell used The frequency with which the Koss and
procedures similar to Grayson’s in construct¬ Butcher (1973) MMPI critical items were en¬
ing his critical items; therefore, it seems likely dorsed by men and women for four samples
that the above reservations about the Gray¬ are presented in Tables 5-22 and 5-23, re¬
son critical items also would apply to the spectively.
Caldwell critical items. Several general comments can be made
Since critical items are face valid, clients about the mean number of critical items en¬
can overreport or underreport the item con¬ dorsed by each sample. First, the university
tent if they so desire (see Chapter 3). Hence, student sample endorsed more items in the
endorsement of the critical items will indicate Situational Stress Due to Alcoholism and
the areas of psychological concern only if the Mental Confusion categories than any of the
items have been endorsed accurately. other three samples; thus, the items in these
There has been very little research with two categories may be influenced by some ex¬
either the Koss and Butcher (1973) and traneous variables, such as the willingness to
Lachar and Wrobel (1979) critical items. report problem behaviors. Second, females
Comparing items within similar content areas are more likely to endorse items within the
on these two lists reveals little item overlap, Acute Anxiety and Depressed-Suicidal Ideas
which suggests that a different meaning of categories than males. Third, the Threatened
the word critical is being used by these inves¬ Assault category does not distinguish any of
tigators. Koss and Butcher identified items the samples, probably because of the few
that are critical for specific crisis groups; items in it. Finally, even normal samples en¬
Lachar and Wrobel identified items that are dorse a moderate percentage of critical items
face-valid descriptors of psychological con¬ within any category, which suggests that ex¬
cerns. Researchers should understand these ploring every critical item endorsed by a client
differences in item selection and determine may likely be a very time-consuming process.
which critical item list is appropriate for their The frequency of endorsement of the
specific use. Lachar and Wrobel (1979) MMPI critical
Evans (1984b) found that his sample of items in the same four samples for men and
normal adults endorsed more critical items women is presented in Tables 5-24 and 5-25
than groups of psychiatric patients and alco¬ respectively. Again, some general conclu¬
holics in the areas of Acute Anxiety and sions can be drawn.
Situtational Stress Due to Alcoholism of the First, females irrespective of the sample
Koss and Butcher (1973) critical items. His were more likely than males to endorse items
normal adults also endorsed more items than in the Anxiety and Tension, Depression and
patients in the area of Problematic Anger of Worry, and Sleep Disturbance areas. Second,
224 Chapter 5

TABLE 5-22 Frequency of Endorsement of Koss and Butcher Critical Items

by Sample for Men

Sample

Clinic Medical Prison University

Clients Patients Inmates Students
Number (N = 140) (N = 86) (N = 200) (N = 96)
Crisis of
Situation Items M SD M SD M SD M SD

Acute Anxiety 9 3.58 1.49 4.86 1.53 2.78 1.41 3.10 1.28
Depressed-Suicidal
Ideas 25 9.83 5.65 5.35 4.85 5.30 4.69 6.76 5.22
Threatened Assault 3 .91 .72 .79 .81 1.38 .65 .87 .76
Situational Stress
Due to Alcoholism 1 5 7.84 2.14 8.85 2.16 9.65 1.82 8.95 1.95
Mental Confusion 3 1.08 .83 .95 .83 .82 .84 1.25 .97
Persecutory Ideas 12 2.41 1.84 1.13 1.14 2.39 1.99 1.86 1.34

males irrespective of the sample were more to question how critical the items are within
likely than females to endorse items in the these three areas.
Deviant Beliefs, Deviant Thinking, Antiso¬ Finally, even the university students,
cial Attitude, and Family Conflict areas. who are supposedly normal, endorsed about
Third, the university students endorsed more one-quarter of these critical items, and only
items in the Deviant Thinking, Substance 11 students (5.3 percent) endorsed 10 or fewer
Abuse, and Problematic Anger areas than the items. Consequently, if the clinician intends
other three samples, which would cause one to pursue every critical item endorsed by the

TABLE 5-23 Frequency of Endorsement of Koss and Butcher Critical Items by Sample
for Women

Sample

Clinic Medical University

Clients Patients Students
Number (N = 275) (N = 1 55) (N = 113)
Crisis of
Situation Items M SD M SD M SD

Acute Anxiety 9 4.03 1.49 5.44 1.40 3.54 1.31

Depressed-Suicidal
Ideas 25 1 1.80 6.01 6.82 4.75 5.30 3.38
Threatened Assault 3 .70 .74 .60 .70 .50 .71
Situational Stress
Due to Alcoholism 1 5 6.52 2.1 5 7.24 2.03 8.58 1.94
Mental Confusion 3 1.63 .76 1.22 .54 1.54 .84
Persecutory Ideas 12 2.1 1 1.50 1.55 2.22 1.73 .15
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 225

TABLE 5-24 Frequency of Endorsement of Lachar and Wrobel Critical Items

by Sample for Men

Sample

Clinic Medical Prison University

Clients Patients Inmates Students
Number (N = 140) (N = 86) (N = 200) (N = 96)
Content of
Area Items M SD M SD M SD M SD

Anxiety and
Tension 1 1 4.43 2.41 3.17 2.29 2.14 1.94 3.21 2.32
Depression and
Worry 16 5.78 3.49 3.38 2.99 2.85 1.31 4.1 3 3.19
Sleep Disturbance 6 2.05 1.53 1.47 1.43 1.59 1.42 1.75 1.56
Deviant Beliefs 1 5 2.83 2.76 1.08 1.74 2.07 2.83 2.12 1.95
Deviant Thinking
and Experience 1 1 2.86 2.29 1.41 1.58 2.38 2.20 3.1 1 2.1 1
Substance Abuse 4 1.29 1.10 .71 .92 1.35 1.15 1.54 1.30
Antisocial Attitude 9 3.41 1.98 2.21 1.87 4.41 1.90 3.62 1.97
Family Conflict 4 1.98 1.19 .74 1.03 1.14 1.00 1.31 1.19
Problematic Anger 4 1.51 1.17 1.13 1.14 .97 1.16 1.74 1.27
Sexual Concern
and Deviation 8 2.36 1.79 1.09 1.25 1.03 1.21 1.97 1.73
Somatic Symptoms 23 5.84 4.50 6.13 4.29 3.56 3.45 4.46 3.89

client, there appears to be a potential prob¬ concern in a variety of settings and popula¬
lem of expending a large amount of time be¬ tions. No criteria have been established to
cause of the frequency of endorsement of determine how many items within a given
these items even in a normal sample. area can be answered before the clinician
The sheer number of items (107 items on should investigate further; research is needed
the MMPI-2; 111 items on the MMPI) in the to establish such criteria although this vio¬
Lachar and Wrobel (1979) critical items list, lates the initial assumption that any item en¬
which includes 97 items scored on the stan¬ dorsed was deemed critical.
dard validity and clinical scales, would make Additionally, items need to be identified
routine inspection of all the items a laborious that do not effectively discriminate among
process in a clinical sample. The Lachar and groups so that the number of items can be re¬
Wrobel critical items contain 39 percent of duced to a manageable size. Some items
the 69 items that Clavelle and Butcher (1977) within the Deviant Thinking, Substance
found most strongly to discriminate MMPI Abuse, and Problematic Anger content areas
codetypes from one another. This suggests seem to be likely candidates for deletion since
that psychopathologic samples will likely en¬ the normal university students were more
dorse a large number of these items. likely to endorse items within these areas than
The Koss and Butcher and Lachar and any of the other three samples studied.
Wrobel critical items lists appear to warrant Finally, the influence of a client’s ten¬
further research to determine how well these dency to overreport or underreport psycho¬
items identify critical areas of psychological pathology on the endorsement of these criti-
226 Chapter 5

TABLE 5-25 Frequency of Endorsement of Lachar and Wrobel Critical Items by Sample
for Women

Sample

Clinic Medical University

Clients Patients Students
Number (N = 275) (N = 1 55) (N = 113)
Content of
Area Items M SD M SD M SD

Anxiety and Tension 1 1 5.10 2.65 3.46 2.26 2.95 2.00

Depression and Worry 16 6.85 3.89 4.28 3.08 3.1 7 2.18
Sleep Disturbance 6 2.80 1.76 2.05 1.51 1.96 1.38
Deviant Beliefs 1 5 2.1 5 2.07 1.05 1.43 1.52 1.25
Deviant Thinking
and Experience 1 1 2.67 1.96 1.64 1.42 2.68 1.68
Substance Abuse 4 1.02 1.16 .38 .66 .90 1.07
Antisocial Attitude 9 2.23 1.84 1.10 1.22 1.80 1.56
Family Conflict 4 2.10 1.25 .86 1.03 1.18 1.02
Problematic Anger 4 1.46 1.14 .86 .94 1.09 1.10
Sexual Concern
and Deviation 8 2.65 1.60 1.55 1.22 1.82 1.15
Somatic Symptoms 23 7.55 5.06 7.64 4.36 3.98 2.77

cal items needs to be investigated, since the other review of the literature on short forms.
fact that a client endorses a specific critical The interested reader also should consult
item does not mean that he or she is providing Greene’s (1982) response to the review by Ste¬
an accurate self-report. vens and Reilley. Butcher and Hostetler
(1990) have reviewed the research on the use
of short forms on the MMPI with suggestions
SHORT FORMS
for how these issues might be addressed on
The extensive number of items in the MMPI- the MMPI-2. Any clinician who is contem¬
2 (567) and MMPI (566) and the length of plating research on short forms on the
time required to complete the test (an hour MMPI-2 should consult this article before
for most clients and ranging upward to sev¬ starting.
eral hours for a few individuals) has led to nu¬ Kincannon’s (1968) short form of the
merous proposals to shorten or reduce the MMPI was developed based on Comrey’s
number of items on the test. Three of the factor analyses of the validity and clinical
more commonly used short forms of the scales (cf. Comrey, 1957a). Kincannon se¬
MMPI will be reviewed here—Kincannon’s lected items to represent each cluster within
(1968) Mini-Mult, Faschingbauer’s Abbrevi¬ each scale. He chose items that were scored
ated MMPI (FAM: 1974), and Overall and on the greatest number of scales (i.e., items
Gomez-Mont’s (1974) MMPI-168. with the most overlap across scales), and
The clinician who desires more in-depth most of his items are scored on three to five
analysis of these short forms of the MMPI or different scales.
other less frequently used short forms should Following this procedure, he identified
consult Faschingbauer and Newmark (1978). 71 items, which he called the Mini-Mult.
Stevens and Reilley (1980) have provided an¬ Kincannon also reworded these 71 items be-
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 227

cause he intended that they be used in an in¬ scales in the standard profile and detecting in¬
terrogative fashion in an interview format. valid profiles than other short forms.
The Mini-Mult yields an estimate of all the The MMPI-168 was developed by Over¬
validity and clinical scales except Scales 5 all and Gomez-Mont (1974) because of their
(Masculinity-Femininity) and 0 (Social Intro¬ need for a brief screening test in view of the
version). Graham and Schroeder (1972) pro¬ disappointing results that they obtained from
vided items that can be added to the Mini- evaluation of the validity of the Mini-Mult in
Mult so that Scales 5 and 0 can be scored. The their psychiatric setting. They selected the
actual items on the Mini-Mult, the proce¬ first 168 items of the MMPI as a screening
dures for transforming raw scores on the test largely because item 168 appears as the
shortened scales into estimates of the raw last item at the bottom of page 7 of the Form
scores on the original scales, and the instruc¬ R test booklet, providing a convenient stop¬
tions for administration can be found in ping point for the client. Of course, if the
Kincannon’s (1968) original article. group booklet form is used, item 168 will
Faschingbauer’s (1974) short form of the have to be marked as the last item to be an¬
MMPI, the FAM, also was developed based swered since it is in the middle of the fourth
on Comrey’s factor analyses of the standard page.
validity and clinical scales (cf. Comrey, One advantage of the MMPI-168 is that
1957). In addition, Faschingbauer used Gra¬ the regular scoring templates can be used,
ham, Schroeder, and Lilly’s (1971) factor and Overall and Gomez-Mont (1974) provide
analyses of Scales 5 and 0, which were not regression equations for estimating the scores
factored by Comrey, and the results of re¬ on all the validity and clinical scales from the
search on deficiencies in Kincannon’s (1968) obtained raw scores. They believe that most
Mini-Mult. of the information in the standard profile is
One-third of the items were selected on well represented in the first 168 items of the
the basis of the greatest amount of overlap MMPI.
with other scales, one-third for the least Research on the frequency with which
amount of overlap, and one-third for the short forms can predict the standard MMPI
greatest number of intercorrelations > 0.29 codetype has yielded mixed results. Hoff¬
with the other items. This procedure yielded mann and Butcher (1975) found that the
preliminary short-form scales, which then Mini-Mult, FAM, and MMPI-168 were com¬
were correlated with their corresponding parable in their ability to predict specific
original scales in a sample of 100 college codetypes, with hit rates ranging from .0 per¬
males. For the FAM scales that did not corre¬ cent to 65 percent, 13.0 percent to 74.0 per¬
late > 0.84 with their corresponding scale, cent, and 6.8 percent to 74.0 percent, respec¬
items were added and deleted until this crite¬ tively. Hedlund, Won Cho, and Powell
rion was met. This final step yielded the 166 (1975) found that the Mini-Mult and MMPI-
items in the FAM, which estimates all the 168 concurred with the standard MMPI high-
standard validity and clinical scales. point pair in 33 percent and 45 percent of
Faschingbauer (1974) reported that their clients, respectively.
when the FAM and MMPI were administered Evans (1984c) reported an average con¬
as two separate tests in a psychiatric sample, cordance rate of 35 percent for codetypes be¬
60 percent of the profiles had an identical tween the MMPI-168 and the standard
high-point scale and 28 percent had the same MMPI in a sample of alcoholic patients.
two high-point scales in any order. The FAM Concordance rates for specific codetypes
was more accurate in predicting high-point ranged from 78 percent for a 2-4/4-2 code-
228 Chapter 5

type to 15 percent for a 7-8/8-7 codetype. In comings of the MMPI item pool and scales
all of these studies the short forms were not (Streiner & Miller, 1986).
independently administered; rather each Only a few studies have directly com¬
short form was extracted from the standard pared the utility of short forms and the stan¬
MMPI, which would inflate the relationship dard MMPI in predicting an external crite¬
between the short form and standard MMPI. rion. Poythress and Blaney (1978) compared
Hoffmann and Butcher (1975) con¬ psychologists’ Q-sort ratings of the FAM,
cluded that there was insufficient evidence to Mini-Mult, and the standard MMPI in 36 pa¬
advocate the clinical use of any of the short tients with a wide variety of psychopathol¬
forms. They particularly cautioned against try¬ ogy. The standard MMPI yielded moderately
ing to use a short form with existing interpretive higher but not statistically significant Q-sort
systems based on the standard MMPI because ratings than the FAM, and the standard
of the low frequency of concordance between MMPI was significantly better than the Mini-
the two tests in terms of high-point pairs. Gra¬ Mult.
ham (1987) voiced the same caution. Using a similar procedure, Rand (1979)
A virtual flood of studies have reported also found that Q-sorts produced by psychol¬
comparisons between a specific short form ogists from the standard MMPI were signifi¬
and the standard MMPI to document subject cantly different from the Mini-Mult for 10
or setting characteristics. Almost all these college students. Newmark, Ziff, Finch, and
studies have focused exclusively on how well Kendall (1978) reported that the correlations
the short form can predict the standard of the FAM, MMPI-168, and standard
MMPI without considering the direct validity MMPI with direct measures of psychopathol¬
of the short form. Since the MMPI is an im¬ ogy seemed comparable. Butcher, Kendall,
perfect predictor of an external criterion, and Hoffman (1980) pointed out that New¬
using a short form with questionable validity mark and associates’ (1978) results appear to
to predict the standard MMPI only seems to represent an atypical sample and they ques¬
compound the potential for error. As tioned whether these results can be general¬
Hoffmann and Butcher (1975) suggested, it ized to other settings. Moreland (1984) found
would make more sense to use direct predic¬ that neither the FAM nor the MMPI-168
tive approaches with short-form tests could be substituted for the standard MMPI
whereby a specific criterion is predicted. in predicting ratings of psychiatric patients.
Vincent (1984) has devised an actuarial Until further research has been con¬
system for use with the MMPI-168 (Overall & ducted on the FAM, Mini-Mult, and MMPI-
Gomez-Mont, 1974) that can be seen as one 168, clinicians should be extremely cautious
attempt to determine the specific correlates in using any short form routinely. Clinicians
of a short form test. His actuarial system is should be particularly wary of trying to use a
very preliminary since it is based on a sample short form to predict the standard MMPI
of 400 patients referred to a private psychiat¬ profile and then follow existing interpretive
ric clinic. This approach is one that should be systems based on the standard MMPI since
followed in using a short form test (i.e., it concordance between codetypes is generally
needs to be construed as a new test that must limited.
be validated directly). However, if the clini¬ The research on the Mini-Mult (Kin-
cian is going to devote the time and effort to cannon, 1968) has yielded consistently nega¬
validate a short form as a new test, it would tive results, which should cause the clinician
make more sense to begin with a new item to question its appropriateness in most situa¬
pool and not be limited by any inherent short¬ tions. The fact that the MMPI-168 (Overall &
 Supplementary Scales, Content Scales, Critical Items, and Short Forms 229

Gomez-Mont, 1974) seems to yield compara¬ tinue to use them will have to demonstrate
ble results to the FAM is interesting because their usefulness empirically for whatever pur¬
of the more elaborate statistical procedures pose they have in mind. It also would seem
that Faschingbauer (1974) used in developing that these same caveats should hold for the
the FAM. If future comparisons of the FAM MMPI-2.
and MMPI-168 with external validity criteria
continue to produce similar results, the ENDNOTES
MMPI-168 would have some inherent advan¬ 1. W. G. Dahlstrom (personal communica¬
tages because the standard booklets and scor¬ tion, November 13, 1979) noted that item 356
ing templates can be retained. (“false”) on the MAC was listed incorrectly as
There seems to be little justification for item 357 (“false”) in Dahlstrom and associates
the use of short forms of the MMPI on a psy¬ (1975).
chometric basis (McLaughlin, Helmes, & 2. Reproduced from the MMPI by permis¬
Howe, 1983; Streiner & Miller, 1986), from a sion. Copyright © 1943, (renewed 1970), by the
clinical perspective (Edinger, 1981), based on University of Minnesota. Published by the Univer¬
sity of Minnesota Press. All rights reserved.
their clinical utility (Helmes & McLaughlin,
3. Ibid.
1983), or based on their concordance with the
4. These scales are all /^-corrected T scores.
standard MMPI (Evans, 1984c; Hedlund et A score greater than 45 on the Goldberg index in¬
al., 1975; Hoffmann & Butcher, 1975). Con¬ dicates a psychotic profile pattern, and a score of
sequently, short forms of the MMPI should 44 or below indicates a neurotic profile pattern.
not be used as a predictor of or substitute for
the standard MMPI, and clinicians who con¬
«
 CHAPTER 6

Co de types

The correlates of MMPI-2 and MMPI code¬ other, whereas 1-3 and 3-1 codetypes will be.
types (specific combinations of the 10 clinical When scale order within a codetype does pro¬
scales) will be considered in this chapter. duce different correlates, these will be noted
These codetypes typically have been studied explicitly.
according to high-point pairs, that is, the two The amount of material presented on a
scales with the highest elevation above a T codetype is a rough index of the frequency
score of 65 (MMPI-2) or 70 (MMPI). with which the codetype is encountered.
A codetype is referred to by writing the Some codetypes occur frequently, such as
numbers of the two scales involved with the 2-4/4-2, 4-9/9-4, and 6-8Z8-6; other code¬
most elevated one first. For example, if a types are rarely encountered in any setting,
client’s two highest scores on the MMPI-2 are such as 3-0/0-3,1 -6/6-1, and 1 -5/5-1. The ac¬
on Scales 2 and 7, and both are above a T tual frequency with which codetypes are en¬
score of 65 but Scale 7 is higher than Scale 2, countered in psychiatric and medical settings
then the client’s codetype would be 7-2. If the will be provided below.
two highest clinical scales have identical T Generally, the relationships among any
scores, they are listed in numerical order. In of the validity scales are not discussed be¬
this example, if both Scales 2 and 7 had iden¬ cause the validity scales serve primarily to es¬
tical T scores of 75, the client’s codetype tablish whether a specific clinical scale profile
would be 2-7. There are 90 possible codetypes can be safely interpreted. When an important
on the MMPI-2 and MMPI following this relationship does exist between a validity
procedure. scale and a codetype, this relationship will be
The order of the scales within the code¬ mentioned.
type will not be differentiated unless empiri¬ The correlates of profiles in which only
cal data indicate that the correlates of the one clinical scale is elevated above a T score
codetype do change depending on which scale of 65 on the MMPI-2 or a T score of 70 on the
is elevated higher. For example, 1-2/2-1 MMPI (spike profiles) also will be discussed.
codetypes will not be distinguished from each Finally, the correlates of high-point triads

231
232 Chapter 6

(three highest elevated clinical scales) will be Gilberstadt also provided actuarial rules
examined when the addition of a third scale for identifying each codetype. For example,
significantly modifies the interpretation of the rules for specifying a 1-2-3 codetype
the codetype. are:
This chapter will provide only the gen¬
eral correlates of each codetype. The clinician Scales 7, 2, and 3 > 70
is strongly encouraged to become familiar Scale 1 > Scale 2 > Scale 3
with the available references providing more
No other clinical scale greater than 70
detailed information on profile interpreta¬
tions that have been developed within a spe¬ Scales L < 65, F < 85, and K < 70
cific population, and to know under what cir¬
cumstances each source might be most Clopton (1975) has developed a computer¬
useful. ized version of the Gilberstadt and Duker
(1965) system.
Marks and Seeman (1963; Marks et al.,
MMPI COOKBOOK 1974) developed their MMPI interpretive sys¬
INTERPRETIVE SYSTEMS tem on hospitalized psychiatric clients seen in
Gilberstadt (Gilberstadt, 1970; Gilberstadt & a university medical center, two-thirds of
Duker, 1965) and Marks and Seeman (Marks whom were women. These clients were liter¬
& Seeman, 1963; Marks, Seeman, & Haller, ate, over 18 years of age, and voluntarily
1974) have developed the most widely known seeking treatment for problems of personal
actuarial cookbooks for the MMPI. Gilber¬ adjustment.
stadt developed his interpretive system on Marks and Seeman (1963) identified 9
male inpatients at a Veterans Administration preliminary codetypes in an original sam¬
hospital. He used five criteria for including a ple; in new samples they revised and refined
client in his preliminary analysis: (1) MMPI these 9 codetypes and identified 11 addi¬
administered within 21 days before or after tional codetypes. Before including a code¬
admission; (2) age range from 20 to 60; (3) type within their system, Marks and associ¬
primary diagnosis not brain damage; (4) L < ates insisted on studying at least 20 clients
60, F < 85, and K < 70; and (5) Shipley In¬ with that codetype; Gilberstadt and Duker
stitute of Living Scale IQ estimate > 105. (1965), on the other hand, used as few as 6
Gilberstadt cautioned the clinician about ap¬ clients in some of their codetypes. In their
plying his cookbook when any of these cri¬ system Marks and Seeman determined the ac¬
teria are not met. He identified 19 codetypes tual correlates of each codetype for women
among these clients, for which he provided only. When they were able to examine dif¬
the following data: ferences between males and females within
a codetype, they found no significant dif¬
1. The most probable diagnosis ferences.
Marks and Seeman (1963) originally de¬
2. The list of complaints, traits, and symp¬
fined their codetypes by complex configural
toms associated with the specific code¬
rules; later they modified their classification
type
procedure (Marks et al., 1974) in view of
3. The cardinal features of the client as a Gynther, Altman, and Sletten’s (1973) dem¬
summary description onstration that codetypes were more useful
4. Descriptive clinical information about than their original configural rules. For ex¬
the client ample, Marks and Seeman (1963) originally
 Codetypes 233

defined a 2-7 codetype by the following cri¬ codetypes (Fowler & Coyle, 1968a; Meikle &
teria: Gerritse, 1970; Shultz, Gibeau, & Barry,
1968). Even when some of the configural
Scales 2 and 7 > 70 rules for codetypes are relaxed, the number
of profiles that can be interpreted with either
Scale 2 minus Scale 8 > 15 points
system does not increase appreciably (Pau-
Scale 7 > Scales 1 and 3
ker, 1966).
Scale 7 minus Scale 4 > 10 points Two additional MMPI profile interpreta¬
Scale 7 minus Scale 6 > 10 points tion systems have been developed, one by
Gynther and colleagues (Gynther et ah, 1973)
Scale 7 minus Scale 8 > 10 points
and the other by Lachar (1974). Gynther and
Scale 9 < 60 associates developed replicated correlates of
Scales L, F, and K < 70. 14 MMPI codetypes that occurred at least 30
or more times in a sample of 3,400 inpatients
In their revised classification procedure in public mental health facilities. They re¬
(Marks et ah, 1974), a 2-7 codetype is simply ported that 55 to 60 percent of MMPIs for
that: Scales 2 and 7 are the two highest clini¬ white clients could be classified into one of
cal scales. Thus, 12 of their 16 current these codetypes.
codetypes are defined simply by the two high¬ They also produced other interesting
est clinical scales. The other 4 codetypes are findings: no evidence could be found that the
defined by the more complex configural rules correlates of a high-point triad differed sig¬
as in their original system. They also devel¬ nificantly from the codetype; absolute eleva¬
oped 29 codetypes for adolescents, which will tion of the codetype above a T score of 70 did
be discussed in Chapter 8. not affect the obtained correlates; it seemed
Marks and Seeman did not report the that gender may have affected the correlates
percentage of profiles that could be classified within a given codetype; and similar code¬
in their revised system, although they did re¬ types obtained from blacks and whites seemed
port that nearly 75 percent of their profiles to require different interpretations.
could be classified in their original system. It The only rule required for classifying a
would be expected that even more profiles profile within the Gynther system is that the
should be classifiable in their revised system. raw score on the F scale be less than 26.
Whether the simplified criteria for classifying Gynther and colleagues (1973) also provide a
profiles within codetypes significantly alters separate interpretation of profiles in which
the applicability of the system will need to be the raw score on the Fscale equals or exceeds
determined empirically. 26. As Gynther acknowledges, the interpre¬
Although these two profile interpretation tive narratives generated by this system are
systems are specific and rather extensive, exceedingly brief compared to other systems.
some interpretive problems remain. When ei¬ For example, the complete narrative for
ther Gilberstadt’s (Gilberstadt & Duker, a 1-3/3-1 codetype is: “This type of client
1965) or Marks and Seeman’s (1963) original may display an unusual amount of bodily
system is used, surprising variation occurs concern, often in the form of multiple so¬
among the clinical scales obtained by clients matic complaints, that sometimes reach the
with a specific codetype (cf. Sines, 1966). proportions of hypochondriasis. Flowever, it
Moreover, it is commonly reported that should be noted that sometimes real physical
only 15 to 35 percent of codetypes from a problems are the cause of the client’s con¬
given clinical setting will fit into any of the cerns” (Gynther et al., 1973, p. 273).
234 Chapter 6

The very limited number of replicated King and Kelley (1977a, 1977b; Kelley &
correlates in the Gynther system for specific King, 1978, 1979a, 1979b, 1979c) reported
MMPI codetypes should be kept in mind the behavioral correlates of specific MMPI
when interpretation of profiles is discussed in codetypes in a college student outpatient
Chapter 7 so that one can appreciate the sample. The students were almost exclusively
amount of nonvalidated material that may be white, predominantly single, and mostly self-
included. referred. King and Kelley required a mini¬
Lachar (1974) developed an automated mum of five students within each codetype,
MMPI interpretive system in a manner very and they analyzed for gender differences
different from the three systems examined. within a codetype if there were five or more
His system was developed predominantly males and females.
with a young, male, military sample, and Since they have not summarized their re¬
each paragraph in this system was evaluated search into a single source, it is necessary to
by having clinicians familiar with the client consult each of the original articles for the be¬
rate its accuracy. In the description of havioral correlates of that codetype. The
Lachar’s (1974) system, the clinician can codetypes for which behavioral correlates
readily see what rules were used to select a have been reported by King and Kelley, as
specific statement and how frequently clini¬ well as the original article to be consulted,
cians judged the paragraph to be accurate. will be indicated following the descriptions of
For example, if only Scale 9 on the the respective codetypes.
MMPI exceeds a T score of 69 in the client’s So far, MMPI cookbooks have not been
profile, the following paragraph will be used: the panacea that was originally thought. In¬
“Similar individuals are often seen as talk¬ creasing the specificity of a particular code-
ative, distractible, and restless. A low frustra¬ type helps by enhancing the homogeneity of
tion tolerance and an insufficient capacity for the group and increasing the probability of
delay is often accompanied by irritability and finding reliable empirical correlates; how¬
maladaptive hyperactivity of thought and ac¬ ever, it also substantially reduces the number
tion [1/19]” (Lachar, 1974, p. 119). The of profiles that could be classified within a
numbers in brackets indicate that this para¬ codetype.
graph was used in 19 of 1,472 clients and was If the rules for defining codetypes are re¬
judged inaccurate once. laxed so that other profiles can be classified,
Lachar’s system has the unique advan¬ the probability of finding reliable correlates
tage of providing some statement or para¬ decreases because of the heterogeneity of
graph for all MMPI profiles, and it provides profiles within the codetype. Furthermore,
correlates of 28 codetypes. It also attempts to when correlates of a codetype are being as¬
provide at least rudimentary validation of the sessed, it is difficult to identify sufficient
common interpretations made about individ¬ numbers of profiles while controlling for sig¬
ual scales and codetypes. Unfortunately, nificant demographic variables. MMPI cook¬
Lachar’s instructions to his clinicians to books, nevertheless, can assist clinicians who
judge the accuracy of each paragraph may are working in certain settings and with cer¬
have biased his system toward over¬ tain sample characteristics.
generalized (high base rate) statements that It should be emphasized that the corre¬
are accurate but also not discriminating in de¬ lates of a specific codetype found in one pop¬
scribing clients. The reader interested in this ulation or setting may not be found in a new
area of research should consult Meehl (1956) population or setting. Hence, the generaliza¬
or Greene (1977, 1978b). tion of the correlates of a codetype to new
 Codetypes 235

groups or environmental settings needs to be with each clinical scale have different preva¬
made cautiously until the necessary research lence rates. Second, the specific setting in
has been conducted. Confident application which the MMPI-2 or MMPI are adminis¬
of these interpretive systems to other popula¬ tered affects which codetypes are likely to be
tions and settings requires empirical research, seen. It should not come as a surprise that
which is sorely needed. codetypes emphasizing Scales 1 (Hypochon¬
In the interim the clinician needs to be fa¬ driasis), 2 (Depression), and 3 (Hysteria)
miliar with MMPI codetypes in order to (1) occur frequently in medical settings, while
understand and validate cookbooks when Scales 4 (Psychopathic Deviate), 8 (Schizo¬
they are available for use, (2) modify and phrenia), and 9 (Hypomania) occur fre¬
adapt the cookbook descriptions to fit the quently in psychiatric settings.
specific client in question, and (3) interpret Finally, the linear T scores that were
meaningfully those profiles that do not fit used with the original Minnesota normative
into any interpretive system. group are not equivalent from scale to scale,
In a nutshell, if there is an empirically as was discussed in Chapter 2; the transition
derived cookbook that is appropriate for a to uniform T scores on the MMPI-2 also has
specific client and setting, the clinician changed the relationships among the clinical
should use it. In the absence of such informa¬ scales as will be seen below.
tion, the clinician will need to do the best job The information on the frequency of
possible with whatever information is avail¬ MMPI codetypes will be reported first, fol¬
able. The clinician is currently in no real dan¬ lowed by similar information on the MMPI-
ger of being replaced by a cookbook or even 2.
a computer, but discussion of this topic will
be reserved for the next chapter.
MMPI
One critical issue for which there are no
data at the present time is whether these Tables 6-1 and 6-2 provide the frequency
cookbooks or correlates generated on the with which each MMPI codetype occurred in
MMPI may be generalized to the MMPI-2. a subset of a large sample (N =21,000) of
Clinical impressions suggest that well-defined psychiatric inpatients and outpatients col¬
MMPI codetypes are very similar on the lected by Hedlund and Won Cho (1979) in the
MMPI-2. However, research that assesses 1970s. Approximately 7,500 of these clients
whether or not the correlates of specific were administered the 399-item Form R and
MMPI-2 codetypes are similar to their MMPI they were excluded from further analyses.
counterparts is needed. Until such empirical Using criteria specified by Nichols, Greene,
data are available, clinicians will need to use and Schmolck (1989) to assess consistency of
the correlates of MMPI codetypes carefully item endorsement, an additional 1,874 clients
in the interpretation of the MMPI-2. were excluded because of inconsistent item
endorsement and 1,014 clients were excluded
because they omitted more than 30 items,
FREQUENCIES OF CODETYPES
which resulted in a final sample of 10,423 cli¬
The clinician needs to be aware of the relative ents.
frequency with which MMPI-2 and MMPI The careful reader might note that the
codetypes are encountered in specific set¬ sample sizes in Tables 6-1 and 6-2 are 6,152
tings. Each codetype does not occur equally and 2,575, respectively, or a total of 8,727 cli¬
often for several reasons. First, the specific ents—and wonder what happened to the
forms of psychopathology that are associated other 1,696 clients. These 1,696 (16.3 per-
236 Chapter 6

TABLE 6-1 Frequency of MMPI Codetypes in Male Psychiatric Inpatients and Outpatients
(Hedlund & Won Cho, 1979)

Highest Second Highest Clinical Scale

Clinical
Scale 1 2 3 4 5 6 7 8 9 0 Total

1 1.1% 2.5% 1.7% 0.8% 0.1 % 0.2% 0.2% 0.8% 0.4% 0.0% 7.6%
2 2.4 3.0 1.2 3.4 0.3 0.7 3.5 2 4 0.1 0.7 1 7.8
3 0.7 0.3 0.3 0.3 0.0 0.0 0.1 0.1 0.0 0.0 1.7
4 1.1 4.3 1.2 9.1 1.2 2.3 1.7 3.5 4.0 0.2 28.4
5 0.1 0.3 0.2 0.4 1.1 0.2 0.2 0.3 0.4 0.0 3.2
6 0 1 0,3 0.1 0.8 0.2 0.8 0.2 1.7 0.3 0.1 4.5
7 0.2 1.2 0.0 0.6 0.1 0.2 0.4 1.3 0.2 0.1 4.4
8 1.8 3.1 0.2 3.8 0.3 5.5 4.3 0.6 2.1 0.1 21 .6
9 0.2 0.1 0.2 2.4 0.4 0.8 0.3 1.5 4.1 0.0 10.1
0 0.0 0.3 0.0 0.1 0.0 0.0 0.1 0.0 0.0 0.3 0.8
Total 7.7 1 5.4 4.9 21.7 3.7 10.6 10.9 1 2.1 1 1.6 1.4 100.0
(N = 6,1 52)

Note: When the highest and second highest clinical scales are identical, the codetype is a Spike profile. For
example, there are 1.1% Spike 1 profiles in these male psychiatric patients.

cent) clients had no MMPI clinical scale Several conclusions can be drawn
greater than a T score of 69 and consequently quickly even from a cursory review of Tables
are not classifiable in a specific codetype. 6-1 and 6-2. First, it is apparent that all
Thus, the most frequent codetype in these two MMPI codetypes did not occur equally often.
tables is a Within-Normal-Limit (WNL) code¬ Some codetypes are very common (Spike 4,
type. Duckworth and Barley (1988) have pro¬ 8-6, Spike 9, 4-9, 8-7, etc.), whereas other
vided an extensive review of WNL codetypes codetypes are very rare (1-0, 3-0, 5-0, 9-0,
that should be reviewed by clinicians. 0-1, 0-3, etc.).

TABLE 6-2 Frequency of MMPI Codetypes in Female Psychiatric Inpatients and Outpatients
(Hedlund & Won Cho, 1 979)

Highest Second Highest Clinical Scale

Clinical ———-——-— ---—-—--——-—-
Scale 1 234567890 Total

1 0.4% 0.7% 2.4% 0.2% 0.1 % 0.2% 0.1 % 0.4% 0.1 % 0.0% 4.5
2 1 .2 1.7 2.2 2.5 0.0 0.9 2.9 2.4 0.0 1.5 1 5.2
3 1 .4 0.9 1 .0 0.9 0.0 0.4 0.3 0.5 0.2 0.1 5.5
4 0.9 2.8 2.3 7.4 0.4 3.7 0.7 4.4 3.3 0.5 26.3
5 0.0 0.1 0.0 0.2 1.7 0.0 0.0 0.1 0.2 0.0 2.3
6 0.2 0.3 0.3 2.8 0.0 2.1 0.1 4.0 1.3 0.3 1 1 .3
7 0.0 0.9 0.2 0.2 0.0 0.2 0.4 1.1 0.1 0.2 3.3
8 0.7 2.4 0.6 4.4 0.1 5.6 3.1 0.7 1.5 0.5 1 9.7
9 0.2 0.2 0.2 1.7 0.2 1.1 0.1 2.0 3.6 0.0 9.3
0 0.1 0.7 0.1 0.2 0.1 0.1 0.1 0.1 0.0 1.4 2.8
Total 5.2 10.5 9.1 20.3 2.6 14.3 7.7 1 5.6 10.3 4.5 100.0
(N = 2,575)

Note: When the highest and second highest clinical scales are identical, the codetype is a Spike profile. For
example, there are 0.4% Spike 1 profiles in these female psychiatric patients.
 Codetypes 237

Second, Scale 4 is the most frequent examined more closely to determine whether
highest (males: 28.4 percent; females: 26.3 specific subgroups are apparent that could
percent) and second highest clinical scale enhance treatment interventions and out¬
(males: 21.7 percent; females: 20.3 percent). comes. These subgroups within frequent code¬
If the Spike 4 codetypes are subtracted from types will not be reported here because of lim¬
these two numbers, since they are counted ited space.
twice, 41.0 percent of the male (28.4 percent Finally, there are few gender differences
+ 21.7 percent — 9.1 percent) and 39.2 per¬ in the frequency of codetypes, although some
cent of the female (26.3 percent + 20.3 per¬ codetypes are more frequent in men (e.g.,
cent — 7.4 percent) clients had Scale 4 as 1-2, 2-1, 4-2) and some are more frequent in
their highest or second highest clinical scale. women (e.g., 2-3, 6-4, 6-8).
It is no wonder that clinicians get the impres¬ The clinician also needs to be aware that
sion that Scale 4 is being interpreted in every the setting in which the MMPI is adminis¬
profile. The frequent occurrence of code¬ tered will affect the frequency with which
types that include Scale 4 indicates that such specific codetypes are found. Tables 6-3 and
codetypes are an excellent place to begin to 6-4 provide the frequency with which each
identify common subgroups who could have MMPI codetype occurred in a large sample
different treatment interventions and out¬ of male and female medical outpatients who
comes. were referred for a psychiatric evaluation at
Third, Scale 0 is the least frequent high¬ the Mayo Clinic (Colligan & Offord, 1986). It
est (0.8 percent) and second highest (1.4 per¬ must be noted that these clients are a subset
cent) clinical scale in men. Scale 0 also is in¬ of general medical clients since their physi¬
frequently the highest or second highest cian referred them for a psychiatric evalua¬
clinical scale in women, although Scale 5 is tion.
the highest or second highest clinical scale It is readily apparent that Scales 1, 2,
even less frequently. This relatively infre¬ and 3 are much more likely to be elevated in
quent occurrence of Scale 0 as one of the two these clients than in psychiatric clients. Ap¬
highest clinical scales no doubt reflects that T proximately 70 percent (men, 69.5; women,
scores on Scale 0 tend to have a restricted 72.7) of the profiles in these medical outpa¬
range when compared to the other clinical tients referred for psychiatric evaluations had
scales. their highest MMPI clinical scale among the
Fourth, the frequency of the various neurotic triad (Scales 1, 2, and 5), compared
codetypes tends to correspond to the amount to approximately 25 percent (men, 27.1;
of clinical literature that is available. For ex¬ women, 25.2) of the profiles of psychiatric
ample, little interpretive information is avail¬ clients.
able on 1-0/0-1, 3-0/0-3, and 6-0/0-6 The most frequent codetypes in these fe¬
codetypes, as can been seen below, which oc¬ male medical outpatients were 1-3 (15.5 per¬
curred infrequently in either male or female cent), 3-1 (12.8 percent), and Spiked (8.5 per¬
clients. Conversely, Spike 4 codetypes occur cent), whereas the most frequent codetypes in
frequently in male and female clients and female psychiatric clients were Spike 4 (7.4
have a large body of interpretive informa¬ percent), 8-6 (5.6 percent), 4-8 (4.4 percent),
tion. and 8-4 (4.4 percent). The most frequent
It would be instructive for every clinician codetypes in male medical outpatients were
to construct tables such as these in his or her 1-3 (12.8 percent), 2-1 (7.9 percent), 1-2 (6.8
own setting so that frequent codetypes can be percent), 3-1 (5.1 percent), and Spike 2 (4.7
identified. Such frequent codetypes could be percent), whereas the most frequent code-
238 Chapter 6

TABLE 6-3 Frequency of MMPI Codetypes in Male Medical Outpatient Referrals

for a Psychiatric Evaluation at the Mayo Clinic (Colligan & Offord, 1986)

Highest Second Highest Clinical Scale

Clinical
Scale 7 2 3 4 5 6 7 8 9 0 Total

7 3.7% 6.8% 12.8% 0.6% 0.4% 0.2% 0.7% 1.1% 0.4% 0.3% 26.9%
2 7.9 4.7 4.1 2.0 1.3 0.8 7.4 2.8 0.2 1.2 32.4
3 5.1 1.1 2.1 0.6 0.5 0.0 0.3 0.3 0.1 0.0 10.2
4 0.4 1 .4 0.8 2.6 0.6 0.4 0.5 0.7 0.7 0.0 8.1
5 0.3 0.6 0.3 0.4 2.6 0.3 0.2 0.3 0.3 0.0 5.2
6 0.1 0.2 0.1 0.1 0.1 0.6 0.1 0.4 0.1 0.0 1 .6
7 0.3 1.9 0.1 0.3 0.2 0.1 0.6 0.9 0.1 0.1 4.5
8 0.6 1.5 0.2 0.8 0.1 0.9 1.4 0.4 0.7 0.0 6.6
9 0.3 0.1 0.1 0.6 0.2 0.1 0.1 0.6 2.0 0.0 4.2
0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.3 0.4
Total 18.6 1 8.4 20.6 7.9 5.9 3.4 1 1.3 7.5 4.7 1.7 100.0
(N = 3,614)

Note: When the highest and second highest clinical scales are identical, the codetype is a Spike profile. For
example, there are 3.7% Spike 1 profiles in these male medical outpatients who were referred for a psychiatric
evaluation.

types in male psychiatric clients were Spike 4 of the setting in which the MMPI is adminis¬
(9.1 percent), 8-6 (5.5 percent), 4-2 (4.3 per¬ tered on the frequency with which the various
cent), and Spike 9 (4.1 percent). codetypes are encountered.
It is evident both that there is no overlap Clinicians also need to be aware of the
in the most frequent codetypes between these frequency with which low points among the
two settings and that gender has only minimal clinical scales are encountered in frequently
impact within a setting. These data should occurring codetypes. Codetype interpretation
help the clinician realize the potential effect of the MMPI-2 or the MMPI emphasizes the

TABLE 6-4 Frequency of MMPI Codetypes in Female Medical Outpatient Referrals

for a Psychiatric Evaluation at the Mayo Clinic (Colligan & Offord, 1986)

Highest Second Highest Clinical Scale

unmcai
Scale 7 2 3 4 5 6 7 8 9 0 Total

7 2.8% 2.6% 1 5.5% 0.3% 0.1 % 0.2% 0.1 % 0.5% 0.3% 0.3% 22.6%
2 3.5 3.3 5.2 1.8 0.1 1.0 4.1 1.5 0.1 2.1 22.6
3 1 2.8 3.2 8.5 1.1 0.0 0.5 0.3 0.4 0.5 0.0 27.5
4 0.4 1.6 1 .2 2.3 0.0 1.1 0.4 1.0 0.8 0.2 8.9
5 0.0 0.0 0.0 0.0 0.4 0.0 0.0 0.0 0.1 0.0 0.6
6 0.1 0.5 0.3 0.4 0.0 0.9 0.2 0.5 0.2 0.1 3.1
7 0.1 1.0 0.3 0.1 0.0 0.1 0.3 0.6 0.0 0.1 2.7
8 0.5 1.3 0.4 0.9 0.0 0.8 1.0 0.2 0.3 0.0 5.4
9 0.4 0.0 0.3 0.4 0.1 0.3 0.0 0.5 2.0 0.0 4.0
0 0.1 0.7 0.0 0.1 0.0 0.1 0.2 0.0 0.0 1.4 2.7
Total 20.7 14.2 31.8 7.4 0.7 5.0 6.6 5.3 4.3 4.2 100.0
(N = 4,792)

Note: When the highest and second highest clinical scales are identical, the codetype is a Spike profile. For
example, there are 2.8% Spike 7 profiles in these female medical outpatients who were referred for a psychiat¬
ric evaluation.
 Codetypes 239

high point(s) among the clinical scales and clinical scale in both male and female psychi¬
consequently less attention is paid to low- atric clients. However, Scale 9 is much less
point scales. Since the low-point scale already often the low point when Scale 4 (Psycho¬
is available in the standard profile, clinicians pathic Deviate) is the highest clinical scale;
can make use of this information without any instead Scale 0 (Social Introversion) is a fre¬
additional work. quent low point.
For example, a low point on Scale 9 (Hy- As might be expected, Scale 5 is the most
pomania) in conjunction with a high point on common low point in female psychiatric cli¬
Scale 2 (Depression) should alert the clinician ents. Scale 0 is the most common low point in
to the presence of significant depressive male psychiatric clients. It also is interesting
symptoms. Hathaway and Meehl (1951, pp. to note that Scale 7 is rarely a low point with
xxvii-xxix) provided data on the frequency any codetype in either males or females. The
with which low points occur with each high- clinician could consult Table 6-5 with every
point scale, but they did not report low points profile to determine whether a person’s low-
for specific codetypes. point scale is one that occurs commonly, and
Table 6-5 provides the frequency of low incorporate that information into the profile
points on the MMPI clinical scales for the 27 interpretation.
most frequent codetypes. Scale 9 is a com¬ As the clinician amasses data in his or
mon low point when Scale 2 is the highest her specific setting, it is highly recommended
TABLE 6-5 Frequency of Low-Point Scales for 27 Frequent MMPI Codetypes by Gender

Low-Point Scale

Codetype 1 2 3 4 5 6 7 8 9 0

1-2
Male 2.0% 2.0% 27.8% 9.3% 4.6% 6.6% 26.5% 21.2%
Female — — 0.0 5.3 52.6 10.5 5.3 0.0 26.3 0.0
1-3
Male — 1.0% — 0.0 18.3 2.9 2.9 1.9 1 7.3 55.8
Female — 1.6 — 4.9 42.6 6.6 1.6 0.0 31.2 1 1.5
Spike 2
Male 10.1 % — 2.7 3.2 14.4 14.9 1.6 1 7.6 28.7 6.9
Female 7.0 — 2.3 1 1.6 30.2 9.3 0.0 4.7 32.6 2.3
2-1
Male — — 0.0 4.7 22.3 1 2.2 0.0 9.5 39.2 1 2.2
Female — — 0.0 0.0 38.7 6.5 3.2 0.0 51.6 0.0
2-3
Male 0.0 — — 0.0 1 2.0 8.0 1.3 1.3 40.0 37.3
Female 0.0 — — 0.0 63.2 0.0 0.0 0.0 33.3 3.5
2-4
Male 1 3.3 — 1.4 — 16.7 8.6 0.0 6.2 33.8 20.0
Female 1.6 — 0.0 — 71.4 4.8 1.6 0.0 20.6 0.0
2-7
Male 6.4 — 4.1 2.3 19.7 8.3 0.0 1.4 52.3 5.5
Female 0.0 — 0.0 1.4 64.9 1.4 0.0 0.0 32.4 0.0
2-3
Male 7.4 — 3.4 1.3 28.2 4.0 0.7 — 43.6 1 1.4
Female 6.6 — 3.3 1.6 63.9 1.6 0.0 — 21.3 1.6
Spike 4
Male 9.5 2.1 3.6 — 16.9 9.1 4.6 4.3 8.0 41.9
Female 10.9 6.3 3.7 — 34.0 6.8 6.8 3.1 6.8 21.5

continued
TABLE 6-5 continued

Low-Point Scale

Codetype 1 2 3 4 5 6 7 8 9 0

4-2
Male 16.4% — 2.7% — 20.5% 4.9% 2.3% 5.7% 16.4% 31.2%
Female 9.9 — 0.0 — 60.6 0.0 0.0 0.0 18.3 1 1.3
4-3
Male 0.0 0.0% — — 5.6 4.2 0.0 1.4 12.7 76.1
Female 1.7 1.7 — — 51.7 1.7 1.7 0.0 6.9 34.5
4-6
Male 26.2 3.6 3.6 — 16.3 — 4.3 1.4 2.1 42.6
Female 23.2 6.3 2.1 — 48.4 — 4.2 0.0 7.4 8.4
4-7
Male 26.9 1.0 2.9 — 19.2 5.8 — 1.0 7.7 35.6
Female 1 7.7 0.0 0.0 — 47.1 0.0 — 0.0 23.5 1 1.8
4-8
Male 18.7 0.5 6.5 — 20.6 3.3 2.8 — 7.0 40.7
Female 14.3 1.8 3.6 — 57.1 0.9 0.9 — 8.9 1 2.5
4-9
Male 1 5.1 5.7 5.7 — 10.2 3.7 2.5 0.4 — 56.7
Female 1 1.6 7.0 2.3 — 36.1 0.0 3.5 0.0 —
39.5
Spike 6
Male 31.2 4.2 1 2.5 0.0% 10.4 — 8.3 4.2 8.3 20.8
Female 16.7 0.0 18.5 3.7 33.3 — 3.7 1.9 1 1.1 1 1.1
6-4
Male 1 7.7 5.9 9.8 — 13.7 — 5.9 2.0 7.8 37.3
Female 8.5 1.4 5.6 — 59.2 — 4.2 0.0 9.9 1 1.3
6-8
Male 20.2 5.8 12.5 1.0 20.2 — 2.9 — 1 1.5 26.0
Female 10.8 5.9 1 3.7 1.0 59.8 —
1.0 —
1.0 6.9
8-1
Male — 0.0 0.0 0.9 41.4 4.5 0.9 —
9.9 42.3
Female — 0.0 0.0 0.0 89.5 0.0 0.0 —
5.3 5.3
8-2
Male 10.4 — 8.9 1.6 30.7 3.7 0.0 — 31.8 13.0
Female 4.9 — 3.3 0.0 70.5 1.6 0.0 — 19.7 0.0
8-4
Male 1 7.2 1.7 10.8 — 23.3 1.7 0.0 — 8.2 37.1
Female 9.7 3.5 0.9 — 64.0 0.9 0.0 —
6.1 14.9
8-6
Male 1 1.6 1.5 1 1.3 0.9 37.7 —
0.6 —
6.5 30.0
Female 4.2 4.2 9.8 5.6 68.5 —
0.7 —
2.1 4.9
8-7
Male 1 1.5 3.4 9.5 1.9 32.4 2.3 — — 13.4 25.6
Female 8.8 0.0 5.0 1.3 78.8 0.0 — —
5.0 1.3
8-9
Male 10.3 7.1 20.6 0.0 17.5 0.8 0.0 — —
43.7
Female 2.6 1 5.8 21.1 5.3 42.1 0.0 0.0 — —
1 3.2
Spike 9
Male 13.3 13.7 13.3 1.6 9.8 6.3 2.4 2.8 —
36.9
Female 1 9.4 23.7 12.9 0.0 16.1 2.2 4.3 2.2 —
19.4
9-4
Male 1 2.2 1 1.5 8.1 — 10.1 4.1 2.7 0.7 —
50.7
Female 0.0 20.5 6.8 — 25.0 2.3 0.0 0.0 —
45.5
9-8
Male 7.5 20.4 14.0 0.0 10.8 5.4 1.1 — —
40.9
Female 1 1.5 26.9 13.5 1.9 30.8 0.0 0.0 — —
1 5.4
Total
Male 1 1.6 3.0 6.3 0.8 20.6 5.2 1.9 2.7 1 5.4 32.5
Female 8.7 5.0 5.1 1.4 51.8 1.9 1.9 0.6 1 1.3 1 2.4

240
 Codetypes 241

that codetype frequency and low-point fre¬ tients. Scales 4 (Psychopathic Deviate) and 8
quency tables be constructed so that the ef¬ (Schizophrenia) occurred less frequently and
fects of this specific setting can be ascer¬ Scale 6 (Paranoia) more frequently as the
tained. Sample sizes as small as several highest clinical scale on the MMPI-2 in both
hundred clients are sufficient to start provid¬ the male (Tables 6-1 and 6-6) and female
ing reasonable estimates of the relative fre¬ (Tables 6-2 and 6-7) patients, although these
quencies of the codetypes. changes were slightly smaller in the female
patients. There was little change in the fre¬
quency of the second highest clinical scale
MMPI-2 other than for Scale 4 to be slightly more
Tables 6-6 and 6-7 provide the frequency common in both the male and female pa¬
with which each MMPI-2 codetype occurred tients.
in the sample of psychiatric inpatients and There also are a number of differences
outpatients collected by Hedlund and Won when specific codetypes are examined in
Cho (1979) that were described earlier. These these psychiatric patients. Some codetypes in
data were developed by dropping the 13 the male patients (2-0, 3-1, 3-2, 6-4, 6-8,
MMPI items that were not retained on the 6-9,and Spike 0) occurred twice as often on
MMPI-2, and rescoring the data on uniform the MMPI-2, whereas others (2-4, 2-8, 4-7,
T scores. This procedure allows for a com¬ 4-8, 8-2, and 8-4) occurred less than half as
parison between the frequency of MMPI-2 often (Tables 6-1 and 6-6). There were some¬
and MMPI codetypes in the same sample of what fewer changes in specific codetypes in
patients by contrasting Table 6-1 with Table the female patients. Several codetypes (2-6,
6-6 and Table 6-2 with Table 6-7. 3-1, 6-2) were twice as frequent and a number
There are a number of differences be¬ of codetypes (Spike 4, 4-8, 8-4, 8-9) were less
tween the frequencies with which the various than half as frequent in the female patients
codetypes occurred on the MMPI and the (Tables 6-2 and 6-7).
MMPI-2 in this sample of psychiatric pa¬ A simple rule that describes the changes

TABLE 6-6 Frequency of MMPI-2 Codetypes in Male Psychiatric Inpatients and Outpatients
(Hedlund & Won Cho, 1979)

Highest Second Highest Clinical Scale

Clinical ■
Scale 1 2 3 4 5 6 7 8 9 0 Total

1 1.7% 2.0% 3.0% 0.4% 0.0% 0.4% 0.4% 0.8% 0.4% 0.1 % 9.2%
2 1.9 2.8 1.3 1.5 0.1 0.8 2.5 0.9 0.1 2.0 14.0
3 2.3 1.0 0.7 0.7 0.1 0.2 0.2 0.2 0.2 0.0 5.5
4 0.9 2.5 1.2 7.8 0.2 2.3 1 .0 1 .8 2.3 0.4 20.3
5 0.0 0.1 0.1 0.2 0.9 0.1 0.1 0.1 0.2 0.0 1.8
6 0.6 1.1 0.4 2.6 0.2 1.5 1.1 6.5 1 .4 0.5 1 5.8
7 0.2 1.7 0.2 0.5 0.0 0.4 0.5 1.7 0.2 0.6 6.0
8 0.7 1.1 0.4 1 .0 0.1 5.0 2.2 0.4 1 .2 0.3 1 2.4
9 0.5 0.2 0.2 1.8 0.2 1.7 0.4 1.5 4.7 0.1 1 1 .2
0 0.2 1.1 0.0 0.2 0.1 0.2 0.2 0.1 0.1 1 .7 3.9
Total 8.9 1 3.7 7.4 16.5 1.8 1 2.7 8.7 1 3.9 10.7 5.6 100.0
(N = 5,663)

Note: When the highest and second highest clinical scales are identical, the codetype is a Spike profile. For
example, there are 1 .7% Spike 1 profiles in these male psychiatric patients.
242 Chapter 6

TABLE 6-7 Frequency of MMPI-2 Codetypes in Female Psychiatric Inpatients and Outpatients
(Hedlund & Won Cho, 1979)

Highest Second Highest Clinical Scale

Clinical '
Scale 1 2 3 4 5 6 7 8 9 0 Total

1 0.8% 1.3% 2.8% 0.5% 0.3% 0.3% 0.1 % 0.5% 0.3% 0.0% 6.9%
2 2.2 2.1 3.5 2.2 0.3 1.8 5.4 2.1 0.2 1.6 21.3
3 3.3 1.4 1.1 1.1 0.0 0.5 0.5 0.3 0.2 0.0 8.3
4 0.6 2.1 1 .4 3.5 0.5 2.3 0.6 1.6 1.8 0.3 14.6
5 0.3 0.3 0.0 0.4 3.2 0.1 0.1 0.2 0.5 0.2 5.0
6 1.0 1.4 0.6 2.7 0.3 1.6 0.7 5.3 2.1 0.3 1 6.0
7 0.2 1.2 0.5 0.1 0.0 0.3 0.5 1.5 0.2 0.3 4.8
8 0.9 1.8 0.2 1 .4 0.2 3.1 2.4 0.7 0.6 0.2 1 1.4
9 0.3 0.2 0.1 1.5 0.5 1.1 0.3 1.8 3.1 0.0 8.8
0 0.1 0.9 0.0 0.1 0.1 0.0 0.2 0.4 0.0 1.0 2.8
Total 9.6 1 2.6 10.3 1 3.3 5.4 1 1.1 10.7 14.3 8.9 3.9 100.0
(N = 2,687)

Note: When the highest and second highest clinical scales are identical, the codetype is a Spike profile. For
example, there are 0.8% Spike 7 profiles in these female psychiatric patients.

on the MMPI-2 in these patients is that referred for a psychiatric evaluation at the
codetypes involving Scales 3 (Hysteria) and 6 Mayo Clinic (Colligan & Offord, 1986).
(Paranoia) have increased in frequency and Again, it should be noted that these MMPI
codetypes involving Scales 4 (Psychopathic data were rescored to simulate MMPI-2 data
Deviate) and 8 (Schizophrenia) have de¬ as described earlier.
creased in frequency. The pattern of changes from the MMPI
Tables 6-8 and 6-9 provides the fre¬ to the MMPI-2 was slightly different in these
quency with which each MMPI-2 codetype medical patients. Scales 2 (Depression), 4
occurred in the medical outpatients who were (Psychopathic Deviate), and 8 (Schizophre-

TABLE 6-8 Frequency of MMPI-2 Codetypes in Male Medical Outpatient Referrals

for a Psychiatric Evaluation at the Mayo Clinic (Colligan & Offord, 1986)

Highest Second Highest Clinical Scale

Clinical
Scale 1 2 3 4 5 6 7 8 9 0 Total

1 5.0% 5.2% 1 3.6% 0.3% 0.2% 0.4% 0.4% 0.5% 0.4% 0.5% 26.4%
2 4.8 4.1 3.5 1.0 0.5 1 .4 4.8 0.9 0.2 2.7 23.8
3 1 2.1 3.9 3.3 0.6 0.3 0.3 0.6 0.2 0.1 0.0 21.4
4 0.2 0.9 0.4 1.4 0.2 0.6 0.2 0.5 0.3 0.1 4.7
5 0.1 0.1 0.1 0.0 1.2 0.2 0.0 0.1 0.2 0.0 2.0
6 0.3 0.8 0.4 0.5 0.1 1.0 0.5 0.9 0.3 0.1 4.8
7 0.4 2.4 0.4 0.3 0.1 0.3 0.6 1.2 0.1 0.2 5.9
8 0.3 0.6 0.1 0.3 0.0 0.7 0.8 0.3 0.4 0.1 3.5
9 0.3 0.2 0.2 0.5 0.1 0.2 0.2 0.4 2.0 0.0 4.0
0 0.4 1 .2 0.0 0.0 0.1 0.0 0.2 0.0 0.0 1.6 3.6
Total 23.7 1 9.2 22.0 4.9 2.7 5.1 8.4 4.9 3.8 5.3 100.0
(N = 3,41 1)

Note: When the highest and second highest clinical scales are identical, the codetype is a Spike profile. For
example, there are 5.0% Spike 1 profiles in these male medical outpatients who were referred for a psychiatric
evaluation.
 Codetypes 243

TABLE 6-9 Frequency of MMPI-2 Codetypes in Female Medical Outpatient Referrals

for a Psychiatric Evaluation at the Mayo Clinic (Colligan & Offord, 1986)

Highest Second Highest Clinical Scale

Clinical
Scale 1 2 3 4 5 6 7 8 9 0 Total

1 4.8% 3.8% 1 7.4% 0.2% 0.5% 0.3% 0.2% 0.4% 0.5% 0.3% 28.3%
2 5.0 3.2 5.4 1.1 0.2 1.2 5.6 1.3 0.1 1.8 25.0
3 1 7.8 4.2 3.8 0.5 0.1 0.4 0.7 0.2 0.2 0.0 27.9
4 0.2 0.9 0.3 0.8 0.1 0.3 0.2 0.2 0.3 0.1 3.3
5 0.2 0.2 0.0 0.0 1.2 0.1 0.1 0.0 0.1 0.1 1.9
6 0.4 0.7 0.3 0.5 0.0 0.5 0.2 0.7 0.1 0.0 3.3
7 0.3 1.3 0.2 0.2 0.0 0.2 0.5 0.4 0.0 0.1 3.2
8 0.2 0.3 0.2 0.2 0.0 0.6 0.5 0.2 0.1 0.0 2.3
9 0.5 0.1 0.3 0.3 0.1 0.2 0.1 0.3 1.1 0.0 3.1
0 0.1 0.6 0.0 0.0 0.0 0.1 0.1 0.0 0.0 0.9 1.9
Total 29.4 1 5.2 27.8 3.7 2.4 3.8 8.1 3.7 2.5 3.4 100.0
(N = 5,331)

Note: When the highest and second highest clinical scales are identical, the codetype is a Spike profile. For
example, there are 4.8% Spike 1 profiles in these female medical outpatients who were referred for a psychiat¬
ric evaluation.

nia) occurred less frequently and Scales 3 the MMPI-2, whereas a number of codetypes
(Hysteria) and 6 (Paranoia) occurred more (Spike 3, 3-4, Spike 4, 4-3, 4-6, 4-8, 8-2, 8- 7)
frequently as the highest clinical scale in the occurred less frequently (Tables 6-4 and 6-
male patients. The only changes seen in the 9).
female patients were for Scales 4 and 8 to The same rule of thumb noted above de¬
occur less frequently as the highest clinical scribes the changes on the MMPI-2 in the
scale; otherwise, the frequency with which male medical patients: codetypes involving
the highest clinical scale occurred on the Scale 3 (Hysteria) have increased in fre¬
MMPI and MMPI-2 in these women was quency and codetypes involving Scales 4
very similar. (Psychopathic Deviate) and 8 (Schizophre¬
There were similar changes in the sec¬ nia) have decreased in frequency. There does
ond highest clinical scale in both the male not appear to be a simple rule that describes
and female patients: Scale 1 (Hypochondri¬ in the changes in the frequency with which
asis) increased in frequency and Scale 4 de¬ specific codetypes are seen in the female med¬
creased in frequency. Scale 0 (Social Intro¬ ical patients. This latter point suggests that
version) also increased in frequency as the changes from the MMPI to the MMPI-2 will
second highest clinical scale in the male pa¬ be affected by the setting in which the test is
tients. administered.
The changes within specific codetypes in Table 6-10 provides the frequency of
the male medical patients were very similar to low points on the MMPI-2 clinical scales for
what was seen in the male psychiatric pa¬ the 27 most frequent codetypes. There are
tients. Several codetypes (2-0, 3-1, 3-2, 0-2, three major differences between the fre¬
and Spike 0) occurred twice as often on the quency of low points on the MMPI and the
MMPI-2, whereas others (1-8, 2-4, 2-8, 8-2, MMPI-2: (1) Scale 5 (Masculinity-Feminin¬
and Spike 5) occurred less than half as ity) is much less likely to be a low point on the
often (Tables 6-3 and 6-8). However, these MMPI-2 in women; (2) Scale 5 is almost as
female medical patients only had Spike 5 twice as likely to be a low point on the
codetypes occurring twice as frequently on MMPI-2 in men; and (3) Scale 0 (Social In-
244 Chapter 6

TABLE 6-10 Frequency of Low-Point Scales for 27 Frequent MMPI-2 Codetypes by Gender

Low-Point Scale

Codetype 1 2 3 4 5 6 7 8 9 0

Spike 1
Male — 1.0% 0.0% 2.0% 45.9% 13.3% 10.2% 6.1 % 9.1 % 1 2.2%
Female — 0.0 0.0 1 9.1 1 9.1 9.5 23.8 1 9.1 4.8 4.8
1-2
Male — — 1.8 4.5 49.1 7.1 2.7 4.5 23.2 7.1
Female — — 0.0 1 1.4 34.3 14.3 2.9 5.7 28.6 2.9
1-3
Male — 1.2 — 3.6 46.1 6.0 3.6 1.8 1 9.8 18.0
Female — 1.3 — 12.0 28.0 14.3 2.9 5.7 28.6 2.9
Spike 2
Male 3.7% — 1.9 1.2 34.8 1 1.1 1.9 1 9.2 22.4 3.7
Female 5.3 — 8.8 1 5.8 12.3 10.5 0.0 10.5 35.1 1.8
2-1
Male — — 0.9 0.9 39.6 1 3.2 0.9 6.6 36.8 0.9
Female — — 3.5 10.3 34.5 6.9 1.7 3.5 39.7 0.0
2-3
Male 0.0 — — 0.0 32.0 8.0 0.0 5.3 48.0 6.7
Female 0.0 — — 2.2 49.5 4.3 1.1 1.1 37.6 4.3
2-4
Male 2.3 — 2.4 — 33.3 7.1 0.0 9.5 40.4 4.8
Female 5.2 — 3.5 — 43.1 6.9 0.0 3.5 34.5 3.5
2-7
Male 4.9 — 4.2 3.5 31.9 4.9 0.0 0.7 49.3 0.7
Female 2.1 — 2.1 1.4 52.1 6.9 0.0 0.0 33.6 2.1
2-8
Male 2.0 — 2.0 2.0 39.2 7.8 0.0 — 43.1 3.9
Female 3.6 — 7.1 1.8 50.0 10.7 0.0 — 23.2 3.6
2-0
Male 2.6 — 8.7 1.7 23.5 8.7 0.9 9.6 44.4 —

Female 2.3 — 6.8 4.6 36.4 6.8 0.0 0.0 43.2 —

3-1
Male — 0.0 — 0.0 40.6 3.9 1.6 6.3 1 9.5 28.1
Female — 0.0 — 3.4 51.1 9.1 3.4 2.3 1 7.1 1 3.6
Spike 4
Male 7.6 1.2 5.5 — 37.2 9.7 5.3 5.8 6.4 21.4
Female 5.3 4.3 1 1.7 — 24.5 8.5 7.5 7.5 4.3 26.6
4-2
Male 8.3 — 4.9 — 41.7 6.3 3.5 6.3 20.1 9.0
Female 8.8 — 1.8 — 47.4 7.0 3.5 1.8 1 9.3 10.5
4-6
Male 1 6.7 1.5 8.3 — 34.1 — 6.1 1.5 10.6 21.2
Female 16.4 1.6 1 1.5 — 54.1 — 3.3 0.0 1.6 1 1.5
4-8
Male 6.9 0.5 16.8 — 43.6 4.0 2.0 — 10.9 1 5.8
Female 9.1 2.3 13.6 — 54.6 2.3 0.0 —
4.6 1 3.6
4-9
Male 1 2.3 3.1 10.8 — 30.0 7.7 1.5 0.0 —
34.6
Female 10.4 4.2 4.2 — 25.0 6.3 2.1 0.0 — 47.9
Spike 6
Male 1 5.5 3.6 23.8 1 .2 21.4 — 6.0 8.3 7.1 1 3.1
Female 14.3 0.0 1 6.7 0.0 31.0 —
9.5 0.0 1 1.9 1 6.7
6-4
Male 18.5 0.7 8.9 — 38.4 — 2.7 0.7 7.5 22.6
Female 1 5.3 2.8 1 2.5 37.5 2.8 0.0 11.1 1 8.1

continued
 Codetypes 245

TABLE 6-10 continued

Low-Point Scale

Codetype 1 2 3 4 5 6 7 8 9 0

6-8
Male 4.9% 1.9% 16.5% 1.6% 53.7% — 0.8% —
1 1.9% 8.7%
Female 5.0 2.1 19.2 5.7 51.8 — 1.4 —
2.8 1 2.1
8-2
Male 0.0 — 13.9 9.2 44.6 6.2% 0.0 —
24.6 1.5
Female 0.0 — 4.3 0.0 72.3 2.1 0.0 —
19.2 2.1
8-4
Male 6.9 1.7 1 7.2 — 39.7 1.7 0.0 —
1 5.5 1 7.2
Female 0.0 0.0 10.8 — 54.1 0.0 0.0 —
8.1 27.0
8-6
Male 4.6 2.1 14.0 1.1 57.5 — 0.0 —
10.9 9.8
Female 2.4 3.6 19.3 3.6 54.2 —
0.0 —
6.0 10.8
8-7
Male 3.2 1.6 12.0 3.2 48.8 0.8 — — 26.4 4.0
Female 1.6 1.6 1 7.2 3.1 67.2 1.6 — —
3.1 4.7
8-9
Male 4.6 0.0 30.3 1.5 47.0 1.5 0.0 — — 1 5.2
Female 0.0 1 1.8 1 1.8 1 7.7 35.3 0.0 0.0 — —
23.5
Spike 9
Male 8.6 1 1.2 20.5 0.8 23.5 5.6 5.2 1.1% — 23.5
Female 1 3.4 1 1.0 18.3 0.0 18.3 6.1 4.9 2.4 — 25.6
9-4
Male 6.9 7.8 10.8 — 31.4 8.8 3.9 0.7 — 30.4
Female 2.6 5.1 1 2.8 — 30.8 2.6 0.0 0.0 — 46.2
9-8
Male 1.2 10.8 27.7 0.0 27.7 6.0 2.4 — — 24.1
Female 2.1 16.7 33.3 4.2 22.9 4.2 0.0 — — 16.7
Total
Male 6.7 2.0 9.8 1.5 37.0 5.2 2.1 3.4 1 7.2 1 5.2
Female 4.6 2.2 10.6 3.7 39.4 5.6 3.2 1.9 1 5.7 1 3.1

troversion) is much less likely to be a low CONCORDANCE BETWEEN MMPI-2

point on the MMPI-2 in men. AND MMPI CODETYPES
Low points on Scale 5 of the MMPI-2
now occur almost equally often in men and One of the critical issues with the advent of
women, whereas low points on Scale 5 of the the MMPI-2 is the concordance between the
MMPI occurred over twice as often in MMPI-2 and MMPI codetypes (i.e., how fre¬
women. Other than these three changes, it quently the MMPI-2 and MMPI codetype
appears that the same low points on the would be similar or identical for a specific cli¬
MMPI-2 tend to be associated with specific ent). Two general comments need to be made
codetypes as was found on the MMPI. about codetype concordance before examin¬
Again, clinicians are encouraged to consult ing the data on this issue. First, if the ratio¬
Table 6-10 for every profile to determine nale for the revising the MMPI is valid in that
whether a person’s low-point scale is one the items and norms on the MMPI do not ac¬
that occurs commonly, and incorporate curately reflect our contemporary society (see
that information into the profile interpre¬ Chapter 1), then it makes little sense to expect
tation. the MMPI-2 and the MMPI to have perfect
246 Chapter 6

concordance. In fact, perfect concordance ple. For instance, there were 92 men with
would suggest that there is little reason to re¬ Spike 1 codetypes on the MMPI-2 in this
vise the MMPI. sample of psychiatric patients, and 64.1 per¬
Second, it is not clear whether the cent of them also had a Spike 1 codetype on
MMPI or the MMPI-2 should serve as the the MMPI.
“gold” standard against which the other test The concordance rate within each gen¬
is evaluated (i.e., if the MMPI-2 and MMPI der is reported twice for each codetype: first
codetype do not agree for a specific client is with the requirement that the two highest
the MMPI-2 or the MMPI inaccurate?). It is clinical scales be in the same order, and sec¬
typically assumed that the MMPI should ond allowing the two highest clinical scales to
serve as the standard, but it is equally plausi¬ be in either order. For example, there were
ble that the MMPI-2 could be the standard. 218 men with 1-2/2-1 codetypes on the
There are two different methods for de¬ MMPI-2. When it was required that these
veloping data that can be used for assessing men have the same codetype on the MMPI-2
concordance between the MMPI-2 and the and the MMPI (i.e., if the man had a 1-2
MMPI. First, MMPI data can be quickly codetype on the MMPI-2 then he had to have
transformed to simulate the MMPI-2 by a 1-2 codetype on the MMPI), the concor¬
dropping the 13 items that were not retained dance rate was 67.0 percent. When Scales 1
on the MMPI-2 and converting the raw and 2 were allowed to be in either order as the
scores into the appropriate T scores. This two highest clinical scales, the concordance
procedure assumes that the changes made at rate in these men was 81.7 percent.
the item level on the MMPI-2 will not have a The average concordance rate across all
systematic effect on the data. It is also possi¬ MMPI-2 codetypes was around 50 percent
ble to score both the MMPI-2 and the MMPI when the two highest clinical scales had to be
if Form AX, the form used to collect the in the same order, and around 65 percent
restandardization data, was administered when the two highest clinical scales could be
since it contains all of the items on the MMPI in either order. Several codetypes (2-4/4-2,
and the MMPI-2. 2-8/8-2, 4-8/8-4, and 4-9/9-4) had very high
Second, the MMPI and the MMPI-2 can concordance rates in men, whereas a differ¬
be administered to the same clients with some ent set of codetypes (Spike 2, Spike 3, Spike
interval of time between the two test adminis¬ 4, 4-8/8-4, 4-9/9-4, Spike 7, and Spike 0) had
trations (cf. Flonaker, 1990). This latter very high concordance rates in women. There
method has been used infrequently, because also were a number of codetypes that had
it requires a significant amount of time and very low concordance rates: 1-6/6-1 and
effort on the part of the clinician. It also 2-6/6-2 codetypes in men and women, and
tends to confound test-retest reliability 2-0/0-2, 4-0/0-4, and 6-9Z9-6 codetypes in
changes with the differences between the men.
MMPI and the MMPI-2. Butcher, Dahlstrom, Graham, and Tell-
Table 6-11 provides the concordance be¬ egen (1989) provided information on the con¬
tween specific MMPI-2 and MMPI codetypes cordance between the MMPI-2 and the
in the sample of psychiatric inpatients and MMPI using Form AX in a sample of 423
outpatients collected by Hedlund and Won psychiatric patients. They reported concor¬
Cho (1979). The left-hand column indicates dance rates for specific codetypes that were
the specific MMPI-2 codetype and the next very similar to those seen in Table 6-11,
columns report the concordance rates on the which would suggest that the simulation of
MMPI by gender and then for the entire sam¬ MMPI-2 data provided in Table 6-11 is rea-
 Codetypes 247

TABLE 6-11 Concordance between MMPI-2 and MMPI Codetypes

MMPI Codetype

Men Women Total

Two Highest Scales Two Highest Scales Two Highest Scales

MMPI-2 Same Either Same Either Same Either

Codetype N Order Order N Order Order Order Order

Spike 7 92 64.1 _ 20 60.0 _ 63.4 _

7-2/2-7 218 67.0 81.7 93 50.6 50.6 62.1 72.4
1-3/3-1 295 34.2 47.8 163 48.5 57.7 39.3 51.3
7-4/4-7 71 49.3 64.8 28 21.4 53.6 41.4 61.6
1-6/6-1 60 20.0 26.7 37 21.6 21.6 20.6 24.8
1-8/8-1 83 50.6 84.3 36 38.9 50.0 47.1 73.9
Spike 2 1 28 53.1 — 38 81.6 — 59.6 —

2-3Z3-2 1 34 28.4 53.7 1 30 50.8 56.9 39.4 55.3

2-414-2 228 81.6 92.1 1 1 5 47.8 61.7 70.3 81.9
2-616-2 107 21.5 43.9 86 31.4 32.6 25.9 38.9
2-7/7-2 242 63.2 78.9 1 77 46.9 52.0 56.3 67.5
2-8Z8-2 1 16 85.3 94.8 103 38.8 63.1 63.4 79.9
2-0/0-2 210 1 1.9 21.9 86 40.7 51.2 20.3 30.4
Spike 3 37 48.7 — 29 82.8 — 63.7 —

3-4Z4-3 107 34.6 56.1 67 35.8 67.2 35.1 60.4

Spike 4 41 6 68.5 — 88 100.0 — 74.0 —

4-6Z6-4 278 30.9 56.5 1 33 60.2 85.7 40.4 65.9

4-7/7-4 84 65.5 76.2 19 36.8 42.1 60.2 69.9
4-8Z8-4 1 59 83.6 97.5 81 92.6 98.8 86.6 97.9
4-9Z9-4 232 75.0 89.7 87 75.9 96.6 75.2 91.6
4-0/0-4 51 2.0 1 5.7 16 37.5 56.3 10.5 25.4
Spike 5 47 53.2 — 75 73.3 — 65.6 —

Spike 6 80 50.0 — 36 75.0 — 57.8 —

6-8/8-6 654 23.7 58.0 224 52.2 82.6 31.0 64.3

6-979-6 1 74 29.9 37.9 87 54.0 66.7 37.9 47.5
Spike 7 1 1 45.8 — 12 100.0 — 74.1 —

7-8/8-7 221 58.8 81.4 105 54.3 74.3 57.4 79.1

Spike 8 22 36.4 — 28 46.4 — 42.0 —

8-9Z9-8 149 60.4 83.9 65 75.4 87.7 65.0 85.1

Spike 9 256 63.7 — 70 70.7 — 65.2 —

Spike 0 97 48.5 — 25 92.6 — 57.5 —

Mean 48.7 64.0 57.9 62.3 51.9 66.4

Weighted
Mean 49.2 65.3 56.1 65.8 51.4 69.2
N 5059 3873 2359 1 938 7418 581 1

sonably accurate. Thus, clinicians could use and the MMPI using a test-retest format in a
the data in Table 6-11 as a good approxima¬ sample of 55 psychiatric patients. This con¬
tion of the concordance to be expected be¬ cordance rate was similar to the concordance
tween the MMPI-2 and the MMPI in psychi¬ if the MMPI had been readministered (40
atric samples until larger psychiatric samples percent) or the MMPI-2 had been readminis¬
are available on the MMPI-2. tered (35 percent). Basically, Honaker found
Honaker (1990) reported codetype con¬ low concordance rates when either the
cordance of 32 percent between the MMPI-2 MMPI-2 or the MMPI was readministered
248 Chapter 6

regardless of which test was administered ini¬ reported in the left-hand column and the con¬
tially. cordance rates are reported for the MMPI-2.
Table 6-12 provides similar information A quick perusal of Tables 6-11 and 6-12 will
as Table 6-11 on the concordance between reveal that concordance rates for specific
specific MMPI-2 and MMPI codetypes in the codetypes can vary drastically, depending on
sample of psychiatric inpatients and outpa¬ whether the MMPI-2 or the MMPI is used as
tients collected by Hedlund and Won Cho the criterion.
(1979), except that now MMPI codetypes are For example, 4-8/8-4 codetypes on the

TABLE 6-12 Concordance between MMPI and MMPI-2 Codetypes

MMPI-2 Codetype

Men Women Total

Two Highest Scales Two Highest Scales Two Highest Scales

MMPI Same Either Same Either Same Either

Codetype N Order Order N Order Order Order Order

Spike 1 66 89.4 _ 23 52.2 _ 79.8 —

1-2/2-1 299 48.8 59.5 50 78.0 82.0 53.0 62.7

1-3/3-1 144 70.1 97.9 97 81.4 96.9 74.6 97.5
1-4/4-1 1 12 31.3 41.1 28 21.4 53.6 29.3 43.6
1-6/6-1 16 75.0 100.0 10 80.0 80.0 76.9 92.3
1-8/8-1 160 26.3 43.8 28 50.0 64.3 29.8 46.9
Spike 2 141 48.2 — 76 40.8 — 45.6 —

2-3Z3-2 93 40.9 77.4 79 83.5 93.7 60.5 67.9

2-4Z4-2 458 40.6 45.9 1 34 33.6 53.0 39.0 66.9
2-6Z6-2 100 23.0 71.0 60 45.0 48.3 31.3 59.0
2-7/7-2 292 52.4 65.4 98 84.7 93.9 60.5 47.9
2-8Z8-2 338 29.3 32.5 1 22 32.8 53.3 30.2 74.5
2-0/0-2 56 44.6 82.1 55 63.6 80.0 54.0 39.6
Spike 3 37 48.7 — 29 82.8 — 63.7 —

3-4/4-3 88 42.0 65.9 80 30.0 56.3 36.3 61.9

Spike 4 315 90.5 — 1 72 51.2 — 76.6 —

4-6Z6-4 192 44.8 81.8 166 48.7 62.7 46.6 71.9

4-7/7-4 1 34 41 .0 47.8 21 33.3 38.1 40.0 43.5
4-8Z8-4 438 30.4 35.4 226 33.2 35.4 31.4 35.4
4-9Z9-4 373 46.6 55.8 130 50.8 64.6 47.7 60.4
4-0/0-4 14 35.7 85.7 18 33.3 50.0 34.4 68.5
Spike 5 30 83.3 — 72 76.4 — 78.4 —

Spike 6 49 81.6 — 55 49.1 — 64.4 —

6-8/8-6 440 35.2 86.1 246 47.6 75.2 39.6 79.8

6-9Z9-6 69 75.4 95.7 63 74.6 92.1 75.0 93.9
Spike 7 18 61.1 — 19 63.2 — 62.2 —

7-8/8-7 341 38.1 53.1 109 52.3 71.6 41.5 63.8

Spike 8 12 66.7 — 28 46.4 — 52.5 —

8-9Z9-8 216 41.7 57.9 90 54.4 63.3 45.4 61.0

Spike 9 1 77 92.1 — 99 70.7 — 84.4 —

Spike 0 47 100.0 — 50 50.0 — 74.2 —

Mean 54.0 65.8 54.7 67.1 53.5 63.8

Weighted
Mean 47.5 58.5 52.4 65.9 49.1 62.7
N 5265 4373 2533 1910 7798 6283
 Codetypes 249

MMPI have a concordance rate of 35 percent rates for “well-defined” codetypes when the
on the MMPI-2 in men and women (Table MMPI-2 is used as the criterion, and Table
6-12), whereas 4-8/8-4 codetypes on the 6-14 provides similar information when the
MMPI-2 have a concordance rate over 90 MMPI is used as the criterion. Two conclu¬
percent on the MMPI in men and women sions are apparent when Tables 6-13 and
(Table 6-11). Thus, 4-8/8-4 codetypes on the 6-14 are contrasted with Tables 6-11 and
MMPI-2 should be a very homogeneous sub¬ 6-12, respectively: (1) concordance rates
set of 4-8/8-4 codetypes on the MMPI, and it have increased dramatically and now average
would be expected that 4-8/8-4 codetypes on around 80 percent in men and over 90 percent
the MMPI-2 should have more reliable corre¬ in women, and (2) almost three-fourths of the
lates since the codetype is more homoge¬ profiles do not meet this criterion of a well-
neous. defined codetype.
The opposite pattern for concordance Graham (1990a) reported similar con¬
rates between the MMPI-2 and the MMPI cordance rates for well-defined codetypes in
also can be found. For example, 1-3/3-J the Butcher and associates’ (1989) sample of
codetypes on the MMPI have a concordance psychiatric patients: 82 percent for male psy¬
rate over 90 percent on the MMPI-2 in men chiatric patients and 97 percent for the fe¬
and women (Table 6-12), whereas 1-3/3-1 male patients. Graham did not indicate how
codetypes on the MMPI-2 have a concor¬ many profiles did not meet the criterion to be
dance rate around 50 percent on the MMPI in well-defined. Again, it seems that the simula¬
men and women (Table 6-11). Thus, 1-3/3-1 tion of MMPI-2 data based on the MMPIs
codetypes on the MMPI-2 are a more hetero¬ collected by Hedlund and Won Cho (1979) in
geneous subset of 1-3/3-J codetypes on the their psychiatric patients is reasonably accu¬
MMPI, and it would be expected that 1-3/3-J rate and can be used by clinicians until such
codetypes on the MMPI-2 should have less information is available on large samples of
reliable correlates since the codetype is more psychiatric patients with the MMPI-2.
heterogeneous. This discussion of concordance rates be¬
Despite these differences in concordance tween the MMPI-2 and the MMPI leaves un¬
rates between specific codetypes depending answered whether a specific codetype on the
on whether the MMPI-2 or the MMPI is used MMPI-2 will have the same correlates as has
as the criterion, the average concordance rate been found with that codetype on the MMPI.
is around 50 percent if the two highest clinical Harrell (1990) reported symptom correlates
scales are required to be in the same order, of MMPI-2 and MMPI clinical scales to be
and around 60 percent if the two highest clin¬ very similar in 55 psychiatric patients with
ical scales can be in either order. most correlations in the .40 to .60 range.
Another way of examining the concor¬ Research is needed that reports the em¬
dance between the MMPI-2 and the MMPI is pirical correlates of specific MMPI-2 code¬
to require that the specific codetype be “well- types regardless of the concordance between
defined” (i.e., that there be at least a 5 to 10 the MMPI-2 and the MMPI. In fact, clini¬
T point difference between the scales in the cians should be discouraged from reporting
codetype and the next highest clinical scale). information on the concordance rates be¬
Such a requirement is thought to produce a tween the MMPI-2 and the MMPI since it
codetype that would be relatively stable over could impede the investigation of the empiri¬
time if the client were to retake the MMPI-2 cal correlates of specific MMPI-2 codetypes.
or the MMPI. Until such research is available, clinicians can
Table 6-13 provides the concordance use concordance rates as another piece of in-
250 Chapter 6

TABLE 6-13 Concordance between MMPI-2 and MMPI Codetypes

for "Well-Defined" Codetypes

MMPI Codetype

Men Women Total

Two Highest Scales Two Highest Scales Two Highest Scales

MMPI-2 Same Either Same Either Same Either

Codetype N Order Order N Order Order Order Order

Spike 1 20 80.0 _ 1 100.0 _ 81.0 —

1 -2/2-1 24 100.0 100.0 8 75.0 75.0 93.8 93.8

1 -3/3-1 24 83.3 91.7 1 7 94.1 94.1 87.8 92.7
Spike 2 53 71.7 — 27 100.0 — 81.3 —

2-3Z3-2 6 66.7 83.3 14 92.9 92.9 85.0 90.0

2-4Z4-2 36 100.0 100.0 10 90.0 90.0 97.8 97.8
2-7/7-2 34 100.0 100.0 25 76.0 76.0 89.8 89.8
2-0/0-2 70 21.4 21.4 29 93.1 93.1 42.4 42.4
Spike 3 5 100.0 — 10 100.0 — 100.0 —

3-4Z4-3 5 80.0 100.0 5 60.0 80.0 70.0 90.0

Spike 4 200 70.0 — 56 100.0 — 76.5 —

4-6Z6-4 20 80.0 90.0 14 100.0 100.0 88.2 94.1

4-8Z8-4 19 100.0 100.0 12 100.0 100.0 100.0 100.0
4-9Z9-4 49 93.9 93.9 20 100.0 100.0 95.7 95.7
4-0/0-4 13 7.7 7.7 7 85.7 85.7 35.0 35.0
Spike 5 20 70.0 — 36 100.0 — 89.3 —

Spike 6 23 78.3 — 22 90.9 — 84.4 —

6-8/8-6 1 1 1 25.2 86.5 35 80.0 94.3 38.4 88.4

6-9Z9-6 18 72.2 72.2 10 100.0 100.0 82.1 82.1
Spike 7 5 80.0 — 5 100.0 — 90.0 —

7-8/8-7 26 100.0 100.0 14 92.7 92.7 97.5 97.5

8-9Z9-8 21 95.2 100.0 12 100.0 100.0 97.0 100.0
Spike 9 130 81.5 — 51 94.1 — 85.1 —

Mean 76.4 83.1 92.4 91.6 82.1 86.0

Weighted
Mean 69.4 80.0 93.4 92.2 77.1 84.0
N 932 476 440 232 1 372 708

formation in evaluating how to interpret an group mean profiles (cf. Lichenstein &
MMPI-2 codetype. Bryan, 1966; Pauker, 1966; Warman &
Hannum, 1965), which leaves unanswered
whether individual clients’ codetypes have re¬
CODETYPE STABILITY
mained unchanged. Clearly, there would be at
There is little empirical data that indicate least some cause for concern if a client ob¬
how consistently clients will obtain the same tained a 4-9/9-4 codetype on one occasion and
codetype on two successive administrations upon a second administration of the MMPI a
of the MMPI or the MMPI-2. The research few months later in another setting ob¬
on the stability of the MMPI historically fo¬ tained a 2-7/7-2 codetype.
cused either upon the individual validity and Graham, Smith, and Schwartz (1986)
clinical scales (these coefficients were re¬ have provided the only empirical data on the
ported in Chapters 3 and 4, respectively) or stability of MMPI codetypes for a large sam-
 Codetypes 251

TABLE 6-14 Concordance between MMPI and MMPI-2 Codetypes

for "Well-Defined" Codetypes

MMPI-2 Codetype

Men Women Total

Two Highest Scales Two Highest Scales Two Highest Scales

MMPI Same Either Same Either Same Either

Codetype N Order Order N Order Order Order Order

Spike 1 19 84.2 _ 3 33.3 _ 77.3 _

1-2/2-1 28 85.7 85.7 5 100.0 100.0 87.9 87.9
1-3/3-1 22 90.9 100.0 16 100.0 100.0 94.7 100.0
Spike 2 64 59.4 — 30 90.0 — 69.1 —

2-3Z3-2 6 66.7 66.7 13 100.0 100.0 89.5 89.5

2-4Z4-2 63 30.2 30.2 9 100.0 100.0 38.9 38.9
2-7/7-2 53 64.2 64.2 19 100.0 100.0 73.6 73.6
2-8Z8-2 29 24.1 24.1 3 33.3 33.3 25.0 25.0
2-0/0-2 1 5 100.0 100.0 34 79.4 79.4 85.7 85.7
Spike 3 5 100.0 — 1 1 90.9 — 93.8 —

3-4Z4-3 6 66.7 66.7 4 75.0 75.0 70.0 70.0

Spike 4 145 96.6 — 63 88.9 — 94.2 —

4-6Z6-4 20 80.0 90.0 16 87.5 87.5 83.3 88.9

4-8Z8-4 38 50.0 50.0 13 92.3 92.3 60.8 60.8
4-9Z9-4 92 50.0 50.0 23 87.0 87.0 57.4 57.4
Spike 5 1 5 93.3 — 38 94.7 — 94.3 —

Spike 6 21 85.7 — 24 83.3 — 84.4 —

6-8/8-6 105 26.7 91.4 34 82.4 97.1 40.3 92.8

6-9Z9-6 13 100.0 100.0 10 100.0 100.0 100.0 100.0
Spike 7 4 100.0 — 6 83.3 — 90.0 —

7-8/8-7 31 83.9 83.9 13 100.0 100.0 88.6 88.6

Spike 8 1 100.0 — 3 66.7 — 75.0 —

8-9Z9-8 27 63.0 77.8 12 100.0 100.0 74.4 84.6

Spike 9 107 99.1 — 49 98.0 — 98.7 —

Mean 75.0 72.1 86.1 90.1 77.0 76.0

Weighted
Mean 68.3 67.2 90.2 92.4 75.4 74.2
N 929 548 451 224 1380 772

pie (TV = 405) of psychiatric inpatients. They of neurotic, psychotic, and characterologic,
reported 42.7 percent, 44.0 percent, and 27.7 58.1 percent remained in the same category
percent agreement across an average interval when retested.
of approximately three months for high- These data on codetype stability suggest
point, low-point, and two-point codetypes, several important conclusions. First, clini¬
respectively. Only seven specific codetypes cians should be cautious about making long¬
(2-3/3-2, 2-4/4-2, 4-8/8-4, 4-9/9-4, 6-8/8-6, term predictions from a single administration
7-8/8-7, and 8-979-8) occurred frequently of the MMPI-2 or the MMPI. Second, it is
enough to assess their stability; the agreement not clear whether the shifts that do occur in
for these seven codetypes ranged from 26.4 codetypes across time reflect meaningful
percent (6-8/8-6) to 41.4 percent (4-979-4). If changes in the patients’ behaviors, psycho¬
the patients were classified into the categories metric instability of the MMPI or MMPI-2,
252 Chapter 6

or some combination of both of these fac¬ noses), 17 percent as neurotic (depression),

tors. Finally, research is needed to provide and 9 percent as acute organic brain syn¬
additional information on this issue. drome.
Gilberstadt and Duker (1965) reported
that the diagnosis for their prototypic 8-9
RELATIONSHIP BETWEEN
codetype was schizophrenic reaction, cata¬
CODETYPES AND PSYCHIATRIC
tonic type with alternative diagnoses of
DIAGNOSES
“schizo-manic” psychosis and paranoid
Clinicians may be prone to believe that spe¬ schizophrenia. Hathaway and Meehl found
cific diagnoses are associated with certain that 3 (37.5 percent) of their 8 patients with
codetypes. Unfortunately, there is more het¬ 9-8 codetypes were diagnosed as manic-de¬
erogeneity in psychiatric diagnoses within a pressive, manic type. Only 5 patients had
given codetype than clinicians might expect. 8-9 codetypes and none was diagnosed as
For example, Marks and colleagues (1974) manic-depressive, manic type. Thus, it ap¬
reported that 68 percent of adult patients pears that bipolar disorder, manic diagno¬
with a 6-878-6 codetype were diagnosed as ses may be common in 8-9/9-8 codetypes
being psychotic (schizophrenic or paranoid but such diagnoses occur less than one-half
were the most frequent diagnoses), 18 per¬ the time at the best and probably closer to
cent as personality disordered (paranoid), one-quarter.
and 14 percent as chronic organic brain syn¬ Greene (1988a) has summarized the fre¬
drome. quency with which various codetypes were
Gilberstadt and Duker (1965) reported found in specific diagnostic groups, and he
that the diagnosis for their 8-6 codetype was found the expected relationship between
paranoid schizophrenia; since they reported codetypes and diagnostic groups. For example,
prototypic codetypes, they did not provide 2-7/7-2 codetypes occurred frequently in de¬
information on the frequency with which 8-6 pressed patients, 6-878-6 codetypes in schizo¬
codetypes were diagnosed as being schizo¬ phrenic patients, 8-979-8 codetypes in manic
phrenic. Flathaway and Meehl (1951) found patients, and so on. However, the most fre¬
that 16 (46 percent) of their 35 patients with quent codetypes in these groups occurred less
6-8/8-6 codetypes had some form of schizo¬ than 20 percent of the time. It also appeared
phrenic diagnosis. Their patients also were that as the sample sizes increased, the vari¬
twice as likely to be diagnosed as schizophre¬ ability in performance within a specific diag¬
nic when they had an 8-6 codetype as com¬ nostic group increased rather than decreased.
pared with a 6-8 codetype. It appears that pa¬ The clinician should consult the individ¬
tients with a 6-8/8-6 codetype have a ual chapters in Greene (1988b) to review
significant probability of receiving a schizo¬ MMPI performance within specific diagnos¬
phrenic diagnosis although a number of other tic groups, namely, chronic pain (Prokop,
diagnoses also can be encountered. 1988), schizophrenia (Walters, 1988b), mood
Similar data can be provided for 8-9/9-8 disorders (Nichols, 1988), personality disor¬
codetypes so that the clinician can see that ders (Morey & Smith, 1988), substance
heterogeneity in psychiatric diagnoses is char¬ abuse/dependence (Greene & Garvin, 1988),
acteristic of all codetypes. Marks and associ¬ Post-Traumatic Stress Disorder (Penk,
ates (1974) found that 70 percent of adult pa¬ Keane, Robinowitz, Fowler, Bell, & Finkel-
tients with an 8-9/9-8 codetype were stein, 1988), neuropsychological dysfunction
diagnosed as being psychotic (schizophre¬ (Farr & Martin, 1988), and child abuse and
nic or mixed were the most common diag¬ sexual abuse (Friedrich, 1988).
 Codetypes 253

PROTOTYPIC SCORES psychiatric patients (Hedlund & Won Cho,

FOR CODETYPES 1976) and medical outpatients referred for a
psychiatric evaluation (Colligan & Offord,
Although the primary focus of this chapter 1986).
will be on the correlates of each codetype, it is Despite the different frequencies with
important for the clinician to realize that the which these two codetypes are encountered in
relative elevation of the other clinical scales these settings (see Tables 6-6 to 6-9), their
as well as the elevation of a number of the scores on the MMPI-2 standard validity and
supplementary scales can drastically alter the clinical scales are very similar. In fact, the
potential interpretation. prototypic scores for the same codetype in
Profiles 6-1 and 6-2 provide examples these two settings are so similar that it would
of two individuals both of whom have a be redundant to report both sets of scores.
4-9/9-4 codetype. Even a quick perusal of the Consequently, Appendix D contains the pro¬
two profiles reveals that these individuals totypic scores for each codetype based on
would behave very differently in a clinical in¬ psychiatric patients (Hedlund & Won Cho,
terview and they would have different rea¬ 1979).
sons for being referred for treatment. The Persons who are being screened for per¬
first individual (Profile 6-1) is very likely to sonnel selection would be expected to have
have been referred for behavioral difficulties very different patterns of scores on the stan¬
and/or criminal activities, whereas the other dard validity and clinical scales, and are gen¬
individual (Profile 6-2) is more likely to have erally unlikely to elevate most scales. Re¬
interpersonal difficulties, probably with a search is needed to determine whether
spouse or other family member. prototypic scores for a specific codetype in a
Profiles 6-1 and 6-2 illustrate the impor¬ personnel setting are similar to those in a psy¬
tance of assessing how well the obtained pro¬ chiatric setting despite the relative differences
file matches the “prototypic” scores for that in elevation that would be expected.
specific codetype. Appendix D contains ta¬ Since persons in a personnel setting
bles with the prototypic scores for every code¬ would be expected to minimize or underre¬
type discussed in this chapter. Each table pro¬ port any type of psychological problem if
vides the prototypic scores for all of the they are trying to qualify for a position, they
MMPI-2 and MMPI standard validity and generally will not elevate any clinical scale on
clinical scales as well as several of the supple¬ the MMPI-2 to a T score of 65 or higher. In
mentary scales that are scored commonly in those rare instances in which they do produce
psychiatric samples. Examples of these tables a standard codetype, it would be interesting
are provided in this Chapter for five frequently to determine whether the pattern of scores on
occurring codetypes: Spike 1, 2-4/4-2, 2-7-7-2, the standard validity and clinical scales are
4-9/9-4, and 6-8/8-6. The use of prototypic similar to those found in psychiatric settings.
scores also will be illustrated in Chapter 7 when Broughton (1984) and Horowitz,
specific MMPI-2 profiles are interpreted. Wright, Lowenstein, and Parad (1981) have
There are little data to address the issue described the use of prototypes as a means of
of whether prototypic scores for specific assessing personality or psychopathologic
MMPI-2 and MMPI codetypes would be sim¬ constructs. These references can provide a
ilar in different psychiatric settings even if starting point for the clinician who is inter¬
their frequency of occurrence is not the same. ested in a more conceptual understanding of
Profiles 6-3 and 6-4 indicate the similarities this approach to the assessment of psychopa¬
between 2-4Z4-2 and 6-8/8-6 codetypes in thology.
 254 Chapter 6

PROFILE 6-1 Prototypic Scores for a 4-9Z9-4 Codetype

Name_

~,''T i '_ r im\- Address_

Mtk ffrMtut/ily /^mv/A’/y J "
Occupation Date Tested

Profile for Basic Scales Education _ Age_Marital Status

Minnesota \IiiItiphasic Persona I it \ ln\cntor\-2
Copxritihi > h\ (III Rlt.l \ IS()I III! I Nl\I RSITY Oh MINNI SOIA
llM2. I lM> i renew oil ll)7(h. | V) s this Profile form I9N9. Referred By
Ml rights reserxed Distributed c\cltisi\cl> bx NATIONAl COMPl HR SYSThMS. INC
unde license from I he l niversiI\ of Minnesota
MMPI-2 Code
MMPI-2 and Minnesota Mulliphasie Personality Inxentorx-2" are trademarks owned bx
I he l mxersiix of Minnesota Printed in the l niled Stales of America. Scorer's Initials
Hs-SK D Hy Pd- 4K Ml Pa PMK Sc-tK Ma«-2K Si
TorTcL F K 1 2 3 4 5 6 7 8 9 OTorTc

• .

■b 12 ■
. lo ■; r
' 14
'
M •
'■5

.7 13 10 5
2-1 12 10 S
. 12 9 '
17 11 9 .
2i i; - •

20 ’0 s
19 10 3 .
13 9 :
17 9
16 3 ' :
lf> 8 8 3
- 6 3
13 ' 5 3
I?
1!
8
6
5
4
.’
2
Prototypic Scores for

10 5 4 2 _4-9/9-4 Codetypes
3 4 2
: 4 3 1
6 3 2
'

5 3 2
4 2 2 1
3 2 1 i
2 1 l n
' l 0 0
0 0 0 0

Raw Score
NATIONAL
? Raw Score_ COMPUTER
k to be Added._ SYSTEMS

Raw Score with K_ 2^001

 Codetypes 255

PROFILE 6-2 Nonprototypic Scores for a 4-9Z9-4 Codetype

Name_

SM///,,s /,/ lfu/Of’/n ij/i' Address_

I f?t\wut/i(y /(tti’/fAvy -J
Occupation Date Tested

Profile lor Basic Scales

Education_ Age_Marital Status
Mmnosou MuliiplKisic IVisonaliu lmciuor\-2
< op\ right ' In INI RKilMSOI INI l\|\I RSI I Y Oh MINNESOTA
llM: llM> i renewed llPO). This Profile horm I9N9 Referred By_
\ll nehts reseined Disinbtiiccl e\eliisi\eI\ b\ NMIONAl (OMIHITR SYS11MS. INC
uiklei license I'nvjja- me l imersiu of Minnesota.
MMPI-2 Code
and M in tiesola Mullipinisic Personalnt ln\enior\ O" arc trademarks owned b\
I lie L imersiu ok Minnesota. Printed in the t iiileel States ok America.
Scorer’s Initials
as-ilk D Hr Pa-IK Ml Pa PI-IK Sc-IK Ma+2K Si
TorTcL F K 1 2 3 4 5 6 7 8 9 OTorTc

•
. i IS 12 6

- ■

A- 13 10 ■

2b 13 10
- i? 10 s
2i 1? 9 s
;i 9
■
. 3

?0 10 3
19 10 3 2
t? 9 4
y 7 3
16 3 6 3

IS 3 6 3
14 6 3
'
13
1?
11
6
6
5
5
4
3
2
2
Non-prototypic Scores for
in 5 4 2
4-9/9-4 Codetypes
9 S 4 2
3 J 3 2
4 3 i
6 3 2 l

■ 3 2 i
- 2 2 l
3 ? 1 i
? i ! 0
i l 0 0
0 0 0

Raw Score
NATIONAL
? Raw Score_ COMPUTER
k to be Added_ SYSTEMS

Raw Score with K_ 24001

 256 Chapter 6

PROFILE 6-3 Prototypic 2-4Z4-2 Codetypes for Missouri Psychiatric Patients and Mayo Medi¬
cal Outpatients Referred for a Psychiatric Evaluation
Name_
s K I l.irluvv.iv ,muI I ( Mi kmlrv

HintH’.ivttt Address_
fcr.H'na/rfif /metiAvy -J"
Occupation Date Tested.

Profile for Basic Scales

Education_Age_Marital Status
Minnesoi.t Mtiliipluisie Personalitv ln\entor\-2
( op\ i lAiln ‘ hv INI Rlt.lMSOl llll l NI \ I RSITV OK MINNI SOI A
mi. PMa i renewod1 h)7<>). I9S9. I Ins Profile form IW. Referred By_
Ml iil’Iun reserved Distribuned evclusivelv hv NATION \l COM PI TIR SYSTEMS. INC
under license from I he l mversitv of Minnesota.
MMPI-2 Code_
"MMIM-2 and "Mmnesoia Mtiliiplvasic Personalitv Inventors-2" are trademarks owned hv
I-iie l mversitv of Minnesota. Primed in the l nited States of Ameriea.
Scorer's Initials_
Hs-5K 0 Hy Pd*4K Ml Pa PMK Sc* IK Ma-2K Si
TorTcL F K 1 2 3 4 5 6 7 8 9 0 T or Tc

- •

' '2
IS ’2 '
28 ’

’3 Prototypic Scores for 2-4/4-2 Codetypes

25 13 10
'2 10
-
23 '2 9
11 9
5
Medical Outpatient Referrals
2?
21
6 4

(Coiligan & Offord, 1986)

20 10 8 J
19 10 8 4
13 9 4
9 ‘ 3
16 3 6 3

Psychiatric Patients
15 8 6 3
14 7 6 3
13
12
'
6
5
•'
3
2
(Hedlund & Won Cho, 1979)
il 6 4 2

10 5 4 2
» •
8 4 3 2
4 3 1
r. 3 2

5 3 2 i

3 2 l i
2 i 1 0
1 i 0 0
0 0 0 0

T or Tc L F K Hs*5K
D Hy Pd»4K Ml Pa PMK Sc»IK Ma».2K Si T or Tc
1234567890

Raw Score
NATIONAL
? Raw Score_ COMPUTER
k to be Added_ SYSTEMS

Raw Score with K_ 24001

 Codetypes 257

PROFILE 6-4 Prototypic 6-8Z8-6 Codetypes for Missouri Psychiatric Patients and Mayo Medi¬
cal Outpatients Referred for a Psychiatric Evaluation
Name_

MMPI-l
s K I l.trl».i\v.i\ .ind 11 Mi Klink*\

Address_
‘’sofur/rty fmi'tiAry J
B—M—— Occupation Date Tested
Profile for Basic Scales
Education_Age_Marital Status
Minnesota Mtiltiphasic Personally lmcnior\-2
Cop\right > b> III! RI til NTS Oh INI CMN I KSIIMOh MINNESOTA
PHI AM' i renew.Oil #70). p)N9. this Profile Form IW. Referred B>_
Ml iiilhis reserved. Distributed e\clusivel\ b\ NATIONAL COMPUTER SYSTEMS. INC
under license from (he l imersilx of Minnesota.
MMPI-2 Code_
AIMI’I-:" .1 lid "Minnesota Mtiltiphasie Personalty ln\entor\-2" are trademarks owned b\
I lie l im eisiis of Minnesota. Primed in the l nited States of America.
Scorer’s Initials_
Hs-SK D Hy Pd- 4K Ml Pa Pi* IK Sc* IK Ma-2K Si
TorTcL F K 1 2 3 4 5 6 7 8 9 OTorTc

FMcitorii oi k
■ 5 •

b
29 '5 12 ■
23 - ■
;■ ': 11 ■
" 13 10 5

23
2-
13 10
12 10 5 Prototypic Scores for 6-8/8-6 Codetypes
23 12 9 5
11 9 4
2i 11 8 i

20 in 3
Medical Outpatient Referrals
'9 10 3 :
13 9
9
:
3
(Colligan & Offord, 1986)
16 ; 6 3

15 3 6 3
14 7 6 3
13 7 5 3
12
11
6
6
:
4
2
2
Psychiatric Patients
10
0
5 4 2 (Hedlund & Won Cho, 1979)
5 4 2
8 4 3 2
7 4 3 1 » ■#
6 3 2
1
5 3 2 1
4 2 2 i
3 2 i i
2 l i 0
i i 0 0
0 0 0 0

Raw Score
NATIONAL
? Raw Score_ COMPUTER
K to be Added_ SYSTEMS

Raw Score with K_ 24001

258 Chapter 6

Several general comments need to be to act out and is reporting more emotional
made about these prototypic scores that are distress as a consequence of his or her behav¬
provided for each codetype: ior.
An intriguing research question that has
1. The percentage of clients who produced not even been addressed yet is whether clients
inconsistent patterns of item endorse¬ who deviate significantly from the prototypic
ment will be reported for each codetype scores for a codetype are less likely to mani¬
but prototypic scores were based only on fest its correlates. It also would be interesting
those clients who endorsed the items to see if profiles that deviate from the proto¬
consistently. type for the codetype are more similar to the
pattern of scores for another codetype.
2. The percentage of clients within each
Greene and Brown have developed a
codetype who scored below the 25th per¬
computer interpretive program for the
centile or above the 75th percentile on
MMPI (1988) and the MMPI-2 (1990) that
the total T score difference on the Wie¬
also provide information on how well a
ner and Harmon (Wiener, 1948) Obvi¬
client’s scores on the standard validity and
ous and Subtle subscales (see Table
clinical scales match his or her codetype and
3-45, p. 100, for the MMPI-2; Table
all other codetypes.
3-46, p. 101, for the MMPI) for all psy¬
Finally, it would be very appropriate for
chiatric clients is provided as an index of
clinicians to develop their own prototypic
the accuracy of item endorsement; how¬
scores for codetypes that occur frequently in
ever, these clients were included in com¬
their setting. Once these prototypes have
puting the prototypic scores.
been identified within a specific setting, the
3. All scores are reported as T scores with clinician also could begin to see if subgroups
the appropriate clinical scales K-cor¬ of profiles were occurring within the code¬
rected unless indicated otherwise. Raw type, and determine whether they were clini¬
scores are provided for the following cal correlates for these subgroups.
scales: Cannot Say (?), total T score dif¬
ference on the Wiener and Harmon
(Wiener, 1948) Obvious and Subtle sub¬ CODETYPES
scales, total number of Lachar and Any correlate of a codetype discussed in this
Wrobel (1979) critical items endorsed, chapter or in any interpretive system is a
and MacAndrew alcoholism scale (Mac- probabilistic statement that may or may not
Andrew, 1965). apply to a specific client. Each statement
should be understood as applying to most cli¬
As a clinical rule of thumb, any score ents or typical of clients with such a codetype.
that varies by more than one standard devia¬ Consequently, most of these qualifiers have
tion (approximately 10 T points) from the been omitted in the following pages.
prototypic score for that scale should be eval¬ It also will be assumed that the clinician
uated as to whether and how it might modify has assessed the consistency and accuracy of
the standard interpretation of that codetype. item endorsement (see Chapter 3), so dis¬
For example, the prototypic T scores on claimers will not be presented with each code¬
Scales 2 and 0 for an MMPI-2 4-9/9-4 code¬ type as appropriate. For example, the clini¬
type are 53.2 and 45.4 (see Table 6-18), re¬ cian might recall that 43.0 percent of 6-8/8-6
spectively. If a client has T scores of 65 and codetypes on the MMPI and 28.7 percent on
60 for these two scales, the client is less likely the MMPI-2 are a result of inconsistent pat-
 Codetypes 259

terns of item endorsement (see Table 3-47, Their problems tend to be chronically in¬
p. 103, or Tables 6-19 or G-44). grained and they have become adjusted to
A number of the codetypes occur infre¬ them. They frequently are psychotic or mani¬
quently and consequently there is little infor¬ fest a severe characterologic disorder. They
mation available for interpretation, as was have little motivation to consider change.
noted above. In such situations it is fre¬ Duckworth and Barley (1988) and Kelley
quently helpful to note the third highest clini¬ and King (1978) provide interpretive infor¬
cal scale, particularly when Scales 5 (Masculin¬ mation on WNL codetypes.
ity-Femininity) and/or 0 (Social Introversion)
are among the two highest scales. If a client
has a 1-5/5-1 codetype and Scale 3 (Hysteria) Spike 1
is the third highest scale, it probably would be Clients who elevate only Scale 1 (Hypochon¬
instructive to consider that the client has a 1-3 driasis) will present a long history of vague
codetype with a high Scale 5. physical symptoms and ailments. Their exag¬
Anytime Scales 5 and/or 0 are among gerated complaints will reflect primarily a
the two highest clinical scales, it probably somatization process even if they also have
would be a good idea to examine which clini¬ some objective physical conditions. Their
cal scale(s) are next highest and to consider complaints may be used to control and ma¬
how the interpretation of the codetype might nipulate others. They do not report any type
be changed by considering these scales. Re¬ of emotional distress despite their physical
search is needed to determine when the third symptoms.
highest clinical scale can and should be sub¬ They are not psychologically minded
stituted for one of the two highest scales to and they invest little effort in understanding
improve the accuracy of the clinical interpre¬ psychological problems in themselves or oth¬
tation. ers. They have difficulty seeing how their
physical symptoms could be related to their
Within-Normal-Umit (WNL) psychological functioning.
This pattern represents a stable, chronic
Clients who do not elevate any clinical scale mode of adjustment that is difficult to mod¬
above a T score of 65 on the MMPI-2 or a T ify.
score of 70 on the MMPI are quite common Tables 6-15 and D-l provide the proto-
in psychiatric settings. In fact, a WNL code¬ typic scores for Spike 1 codetypes.
type is the most frequently occurring code¬
type in most, if not all, psychiatric settings
7 -2/2-1
(cf. Hathaway & Meehl, 1951; Hedlund &
WonCho, 1979). Clients with 1-2/2-1 codetypes present them¬
These clients describe themselves as selves as concerned about their physical func¬
being happy, healthy, and contented. They tioning. General physical symptoms are seen
see their relationships as being satisfying. It is with manifestations of a somatization or psy-
very important to determine whether this chophysiologic reaction. Even when or if
self-description is consistent with the rea¬ they have real physical symptoms, clients ex¬
son^) for which the clients are being evalu¬ aggerate their severity. These clients often
ated. These clients are not reporting any type of complain of nausea, vomiting, weakness, in¬
emotional distress either as a result of the be¬ somnia, and fatigue rather than classical de¬
haviors or symptoms that led them to be evalu¬ pressive features. Dizziness, chest and back
ated or of the process of being evaluated. pains, and tachycardia may be reported.
260 Chapter 6

TABLE 6-15 Prototypic Scores for Spike 1 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 76 119
Age 46.5 14.1 45.4 14.3
Men 86.8% 82.4%
Women 13.2 17.6

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 2.1%
Total (Obvious-Subtle)3 12.9 54.1 24.6 53.3
Critical itemsb 24.4 8.9 23.5 8.2
Overreportedc 1.3% 1.7%
Underreported0 34.2% 33.6%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.3 5.9 4.2 5.6
L 55.4 9.3 59.1 12.2
F 56.3 6.1 51.8 7.7
K 54.7 8.8 50.4 9.8
KHs) 74.0 4.7 69.3 4.0
2(D) 61.0 6.7 57.3 6.3
3(Hy) 62.8 4.4 56.3 5.6
4(Pd) 60.5 6.5 53.4 7.6
5(Mf) 51.9 8.4 43.6 8.8
6 (Pa) 55.1 6.6 49.0 7.8
7(Pt) 53.6 6.7 47.5 7.0
8(Sc) 55.8 7.3 49.4 6.8
9(Ma) 56.2 7.7 48.7 6.7
0(Si) 52.8 7.4 49.8 6.9

Supplementary Scales
M SD M SD
A 48.2 8.3 49.6 8.5
R 64.8 12.3 54.3 10.1
MAC/M AC-Rd
men 21A 4.5 27.0 4.5
women 23.6 2.6 22.8 3.6

Codetype Concordance
Men Women Men Women
MMPI-2 Spike 1 88.1 % 52.2% MMPI Spike 1 64.1% 60.0%
1-2/2-1 21.7 Spike 4 30.0
1-3/3-1 13.0

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Codetypes 261

They lack insight into their somatic than Scale 2 by 10 T points, this profile is a
symptoms and behavior, often refusing to ac¬ classic conversion “V” (see Chapter 4). In
knowledge that their symptoms are related to general, the higher the elevation of the con¬
emotional conflict and are used as a means of version “V,” the more rigid are the client’s
avoiding their psychological problems. The defenses. A conversion “V” above a T score
somatic symptoms are focused around the al¬ of 80 suggests that many of the client’s ef¬
imentary system, particularly on abdominal forts are ineffectively directed toward trying
pain and backaches. Their symptoms are to ward off anxiety, particularly if the/7 scale
vague, nonspecific, and difficult to isolate also is significantly elevated.
medically. These clients return to their physi¬ The absolute elevation of the conversion
cians repeatedly with limited change in their “V” is not related to whether a psychological
physical condition. or an organic diagnosis is likely to be made
They think in a very concrete manner (Schwartz & Krupp, 1971), although persons
and tend to focus extensively on their physi¬ above age 40 are more likely to receive an or¬
cal symptoms. They see their judgment as ganic diagnosis and younger females are
being poor, particularly when compared to more likely to receive a psychological diagno¬
how they used to function. Sustained atten¬ sis (Schwartz, Osborne, & Krupp, 1972).
tion and concentration are difficult for them. Hence, elevations on Scales 1 and 3 cannot be
They are unwilling to seek psychological used reliably to distinguish functional disor¬
counseling since they do not see their prob¬ ders from actual physical disease. In either
lems as psychological in nature. More fre¬ case the client is using somatic symptoms to
quently they seek another medical opinion avoid thinking or dealing with psychological
when a physician suggests that their problems problems.
could reflect psychological factors. Any med¬ These clients are converting their psycho¬
ical intervention should be as conservative as logical problems into physical symptoms that
possible because of the somatization fea¬ localize the difficulty outside of themselves.
tures. Gynther and colleagues (1973) reported They report a wide variety of physical symp¬
that these clients were frequently alcoholics, toms such as gastrointestinal difficulties,
but no other researcher has found this code¬ chest and neck pains, hay fever and/or
type to be characteristic of alcoholics (cf. asthma, and balance and coordination diffi¬
Clopton, 1978; Greene & Garvin, 1988). culties. They describe themselves as being
Gilberstadt and Duker (1965), Gynther tired, inefficient, and lethargic. They see
and colleagues (1973), and Marks and col¬ themselves as being treated unfairly by life.
leagues (1974) provide interpretive informa¬ Depression and worry are not overtly ex¬
tion on 1-2/2-1 codetypes. Table D-2 pro¬ pressed, no matter how concerned the cli¬
vides the prototypic scores. ents are about poor physical functioning.
Suicidal ideation is quite unusual. Narcis¬
sistic and dependent features are more
1-3
likely to be seen.
Clients with 1-3 codetypes are found fre¬ These clients lack insight into their own
quently in both normal and psychiatric popu¬ behavior and are very resistant to interpreta¬
lations. (Information on 3-1 codetypes is tions that there could be psychological in¬
provided in a later section of this chapter.) volvement in the physical complaints. Even
When Scales 1 and 3 are greater than a T when their complaints seem bizarre, they are
score of 65 on the MMPI-2 or a T score of 70 unlikely to be psychotic. Others are likely to
on the MMPI and Scales 1 and 3 are greater experience these clients’ physical symptoms
262 Chapter 6

as being used in a manipulative or passive-ag¬ consequently quite resistant to change or in¬

gressive manner. tervention.
They are very passive, conventional indi¬ Gynther and colleagues (1973) provide
viduals who invest little energy in under¬ further interpretive information on this code¬
standing themselves or others. They value type, and Table D-4 provides the prototypic
being seen as logical and without psychologi¬ scores.
cal problems.
Clients with a 1-3 codetype are more 1-5/5-1
likely to show more somatization features
Scale 1 generally is not elevated with any clin¬
than the histrionic features characteristic of
ical scale other than Scales 2 and 3. Conse¬
clients with a 3-1 codetype. Their physical
quently, there is little information on this and
complaints are usually more nonspecific and
the five subsequent codetypes.
vague and likely to involve backaches, gas¬
When a 1-5/5-1 codetype does occur, it
trointestinal complaints, and so on.
usually is seen in men. Lachar (1974) has
Clients with a 1-3-9 profile have been de¬
commented that Scale 5 points out the passiv¬
scribed as having a chronic brain syndrome
ity and the fussy, complaining attitude of
with trauma and personality disorder (Gil-
these men. They report vague physical symp¬
berstadt & Duker, 1965). Neither Schwartz
toms that are used to avoid personal responsi¬
(1969) nor Golden, Sweet, and Osmon (1979)
bilities. They focus their dissatisfaction with
were able to replicate this codetype as being
life on their physical problems and com¬
characteristic of brain-damaged persons.
plaints. In response to stress, these clients will
Additional interpretive information on
exhibit an increase in the frequency and/or
1-3 codetypes is available in Gilberstadt and
severity of their physical symptoms, and be¬
Duker (1965), Gynther and associates (1973),
come irritable and easily annoyed. They gen¬
Marks and associates (1974), and Prokop
erally are not depressed, anxious, tired, or fa¬
(1988). Table D-3 provides the prototypic
tigued.
scores for 1-3 codetypes.
Interpretive information on this code¬
type is found in Tanner (1990). Table D-5
1-4/4-1 provides the prototypic scores for 1-5/5-1
codetypes.
Clients with 1-4/4-1 codetypes are relatively
rare. Their primary features appear to reflect
1-6/6-1
a somatization process rather than psychopa¬
thy. The elevation on Scale 4 seems to empha¬ Clients with 1-6/6-1 codetypes also appear to
size the pessimistic, whiny, nagging qualities manifest some form of somatization disorder
of Scale 7. They report general pain and dis¬ and display the hostile qualities tapped by
comfort with few specific symptoms. They Scale 6. They are unaware of these hostile
do not report any type of emotional distress qualities or will attribute them to the way oth¬
such as depression or anxiety as a result of ers treat them. Their interpersonal relation¬
their physical symptoms. ships are quite conflicted because of their
Their interpersonal and familial rela¬ propensity to blame and accuse others of any
tionships are characterized by turmoil and problems that might arise. They may be delu¬
chronic complaining. They often exhibit poor sional or obsessive concerning their physical
social skills, particularly with the opposite symptoms, especially when Scale 8 is ele¬
gender. These clients’ focus on somatic vated, even though it is unusual for them to
symptoms is essentially chronic in nature and report actual psychotic behaviors.
 Codetypes 263

Their personality structure is very resis¬ 1-9/9-1

tant to intervention or change.
Clients with 1-9/9-1 codetypes would be ex¬
Table D-6 provides the prototypic
pected to have multiple somatic complaints
scores for 1-6/6-1 codetypes.
that may be exhibited with a high energy level
or agitation.
1 -7/7-1
They describe themselves as happy, care¬
Clients with 1-7/7-1 codetypes exhibit a wide free, and self-reliant regardless of what oth¬
variety of physical complaints that reflect ers think. They are described as agitated,
constant tension and anxiety as well as gen¬ angry, and difficult to interact with.
eral pain and discomfort. Their somatization Table D-9 provides the prototypic scores
features are resistant to intervention or for 1-9/9-1 codetypes.
change. Intellectualization is quite common.
They often feel insecure, inhibited, inferior,
1-0/0-1
and guilty. They are bothered by obsessive
thoughts, and are concerned that they may be Clients with 1-0/0-1 codetypes are quite rare
losing their mind. They also manifest depres¬ so there is little information available. It
sive features and may report sleep difficul¬ seems likely that they would also elevate
ties. Scale 8 and represent a variant of a 1-8/8-1
They are shy, introverted, and retiring in codetype. They report a very limited number
interpersonal relationships. They worry of physical symptoms. They do not indicate
about their interpersonal skills or lack any form of psychological discomfort or
thereof. emotional distress.
Table D-7 provides the prototypic They are shy, introverted, and socially
scores for 1-7/7-1 codetypes. withdrawn. They are very conventional, law-
abiding individuals primarily out of fear of
1-8/8-1 being caught.
Table D-10 provides the prototypic
Clients with 1-8/8-1 codetypes are likely to
scores for 1-0/0-1 codetypes.
present somatic complaints of a bizarre na¬
ture. They actually may have somatic delu¬
sions. The somatic complaints also may
Spike 2
represent defenses against the emergence of
actual psychotic material. They are easily dis¬ Clients who elevate only Scale 2 (Depression)
tracted and confused. They report concentra¬ are likely to be experiencing a mild reactive
tion and memory difficulties. depression even if they deny depressive feel¬
They are socially inept and inadequate, ings. If Scale 9 is below a T score of 45, par¬
particularly when Scale 0 also is elevated. ticularly in persons under the age of 30, the
They often have poor work histories and a probability of significant depression is even
nomadic lifestyle. They feel alienated, iso¬ higher. Suicidal ideation and plans should be
lated, and different from other people. They routinely evaluated in these clients, although
distrust others. Other people see them as odd, suicide cannot be predicted by the MMPI-2
strange, or bizarre. alone (see Chapter 5). Although clinicians
Interpretive information on this code¬ should routinely check the client’s responses
type is found in Gilberstadt and Duker to items 150, 506, 520, and 524 on the MMPI-
(1965). Table D-8 provides the prototypic 2 since their content refers directly to suicidal
scores for 1-8/8-1 codetypes. ideation and/or suicide attempts, these items
264 Chapter 6

are especially relevant to understanding cli¬ addition to the chronic nature of their adjust¬
ents who have this codetype. ment, make response to any form of psycho¬
These clients will have the general char¬ logical intervention very poor unless motiva¬
acteristics of depressed individuals: feelings tion for long-term psychotherapy can be
of inadequacy, lack of self-confidence, self- discovered.
deprecation, pessimism about the future, and Gynther and associates (1973) furnish
strong guilt feelings. They are good psycho¬ additional information on the 2-3/3-2 code¬
therapy candidates, and they will show signif¬ type, and Table D-l2 provides the prototypic
icant improvement within a relatively short scores.
period of time.
They are socially reserved and intro¬
2-414-2
verted. They tend to withdraw and isolate
themselves when in conflict or under stress, Clients may achieve 2-4/4-2 codetypes for a
which serves to exacerbate their depressive myriad of reasons. One critical factor, which
symptomatology. They are very conventional should be assessed through an interview with
people who are unlikely to get in trouble be¬ the client or knowledge of the client’s reason
cause of their behavior. They are rarely psy¬ for taking the MMPI-2, is whether Scale 2 is
chotic. being elevated by internal (intrapsychic)
Kelley and King (1979a) provide interpre¬ and/or external (situational) causes. Exam¬
tive information on Spike 2 codetypes and ples of the latter are psychopathic individuals
Table D-l 1 provides the prototypic scores. who have been caught in some illicit or illegal
activity and who are being evaluated as a con¬
sequence of their behavior.
2-3/32
The depression in these persons repre¬
Clients with 2-3/3-2 codetypes evidence sig¬ sents the constraints being placed on their be¬
nificant depression as well as lowered activity havior; their depression, or possibly boredom
levels, apathy, and helplessness. The presence at being externally constrained, will alleviate
of significant depressive symptomatology itself quickly once they manage to extricate
can be the result of their ineffective use of themselves from their present situation. The
histrionic mechanisms and defenses. They are presence of even this situational depression in
characteristically described as being over¬ these persons suggests a better prognosis than
controlled and having difficulty in expressing persons in similar circumstances who achieve
their feelings, being immature, inadequate, a Spike 4 or a 4-9/9-4 codetype.
and dependent. These psychopathic clients with a 2-4/4-2
They have grown accustomed to their codetype can be understood best by examining
chronic problems, and they continue to func¬ the correlates of a Spike 4 codetype. They will
tion at a lowered level of efficiency for pro¬ display excellent intellectual insight into their
longed periods of time. Physical complaints behavior, make a positive impression of their
are likely, often with a histrionic quality. earnestness on the clinician, and vehemently
Women are likely to be sexually dysfunc¬ protest that they will change their behavior.
tional; this may represent their control of sex¬ Despite their “sincere” intentions, recur¬
ual feelings or an expression of their unhappi¬ rences of acting out are very likely, followed
ness with their spouse. by the same protestations to do better when
These clients usually lack insight into caught again.
their own behavior and are reluctant to seek Another subgroup of clients with 2-4/4-2
psychological counseling. These factors, in codetypes is more likely to be chronically de-
 Codetypes 265

pressed and unhappy without evidence of an- 2-5/5-2 •

tisocial acting out. These clients are display¬

There are almost no published data on 2-5/5-2
ing hostility and resentment, which often
codetypes. It would be anticipated that males
result from marital conflict, familial difficul¬
with liberal arts college educations who are
ties, or similar situations that make them feel
undergoing situational depression would pro¬
trapped and hopeless.
duce this codetype. The depressive symptom¬
They are immature, dependent, and ego¬
atology tends to be very mild and chronic in
centric and often vacillate between pitying
nature. As the situational depression is re¬
themselves and blaming others for their diffi¬
solved, Scale 2 should decrease in elevation
culties. These behaviors are chronic in nature
and Scale 9 should increase. Their depression
and difficult to resolve through psychological
tends to be more serious when Scale 2 is more
interventions. Involvement of the other mem¬
elevated than Scale 5.
bers of the family or the spouse in the thera¬
They are very passive, dependent, intro¬
peutic interaction is important if meaningful
verted, and shy although they have adequate
behavior change is to occur.
social skills. They are unlikely to get into
Clients with 2-4/4-2 codetypes are fre¬
trouble because of their behavior. They have
quently identified as alcoholics (cf. Clopton,
poor interpersonal and marital relationships.
1978a; Graham & Strenger, 1988; Greene &
King and Kelley (1977b) found that male
Garvin, 1988). Hodo and Fowler (1976) re¬
college students who were psychiatric outpa¬
ported that 21 percent of 1,009 white male al¬
tients with this codetype were anxious, disori¬
coholics had this codetype and that this was
ented, and withdrawn. These students dated
the most frequent codetype among these
infrequently, had somatic complaints, and
males. Greene and Garvin (1988), in their
had a physical history for their complaints
summary of MMPI research on substance
and difficulties.
abuse samples found that 15.3 percent of
King and Kelley (1977b) and Tanner
male and 14.8 percent of female alcoholics
(1990) present more interpretive information
had a 2-4/4-2 codetype. Gynther and col¬
on 2-5/5-2 codetypes, and Table D-14 pro¬
leagues (1973) also found that alcoholism was
vides the prototypic scores.
a replicated correlate of this codetype in their
psychiatric sample. These clients will evi¬
2-6/6-1
dence depressive features, familial conflict,
and vocational problems characteristic of al¬ Clients with 2-6/6-2 codetypes are angry, de¬
coholics. pressed individuals with so much anger that it
In determining which of the above de¬ is directed both against themselves and oth¬
scriptions fits a specific client, useful infor¬ ers. Unlike most depressed clients who are
mation can be obtained by analyzing the con¬ unable to express their anger, clients with this
tent scales; a number of the supplementary codetype usually are openly hostile and re¬
scales such as Dominance, MacAndrew Alco¬ sentful toward others. As a result, these clients
holism, and so on (see Chapter 5); the reasons often have poor interpersonal relationships and
why the client is taking the MMPI-2; and the may be rejected by significant others. They are
other clinical scales. very sensitive to criticism and prone to over¬
Further interpretive information on this interpret the most innocuous comments.
codetype may be found in Gynther and asso¬ This configuration represents a chronic
ciates (1973) and Kelley and King (1979a). pattern of adjustment which is difficult to
Tables 6-16 and D-13 provide the prototypic alter. Their behavior generally invites others
scores for 2-4/4-2 codetypes. to reject and avoid them. If Scale 6 is mark-
266 Chapter 6

TABLE 6-16 Prototypic Scores for 2-474-2 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 606 343
Age 38.1 12.5 36.9 1 2.4
Men 78.1% 66.5%
Women 21.9 33.5

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 6.3% 3.2%
Total (Obvious-Subtle)3 38.6 59.9 68.3 62.3
Critical items'3 32.9 12.1 34.4 1 1.9
Overreportedc 5.9% 16.3%
Underreported0 22.8% 13.7%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.2 5.5 3.4 4.7
L 50.9 8.0 52.5 10.5
F 62.2 9.0 64.9 13.9
K 51.5 8.8 45.3 9.3
1(Hs) 61.1 10.8 58.1 10.3
2(D) 81.8 9.1 76.1 8.4
3(Hy) 65.7 8.8 60.1 10.1
4(Pd) 83.1 8.3 77.6 7.9
5(Mf) 56.3 10.1 48.1 10.1
6(Pa) 63.5 9.6 59.3 1 1.0
7{Pt) 68.5 9.6 63.8 10.3
8(Sc) 65.8 1 1.6 60.6 1 1.7
9(Ma) 57.8 9.7 51.3 9.0
O(Si) 60.0 9.7 58.9 9.4

Supplementary Scales
M SD M SD
A 60.1 10.8 62.7 10.8
R 68.1 13.1 57.4 1 1.5
MAC/MAC-Rd
men 26.9 4.7 26.8 4.9
women 22.4 4.5 22.7 4.4

Codetype Concordance
Men Women Men Women
MMPI-2 2-474-2 45.9% 53.0% MMPI 2-474-2 92.1 % 61.7%
2-7/7-2 1 5.7 Spike 4 1 5.6
Spike 4 10.4 4-878-4 13.0

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Codetypes 267

edly elevated and is higher than Scale 2, the quently, it often is possible to examine the
possibility of a psychotic reaction of a para¬ third highest clinical scale to augment the in¬
noid type should be carefully evaluated. terpretation. Three scales (3, 4, and 8) often
Kelley and King (1979a) provide inter¬ are elevated with Scales 2 and 7, thus produc¬
pretive information on 2-6/6-2 codetypes, ing three high-point triads—2-7-377-2-3,
and Table D-15 provides the prototypic 2-7-477-2-4, and 2-7-877-2-8.
scores. Clients with 2-7-377-2-3 codetypes,
which is the least frequent of the three tri¬
ads, are likely to be docile, passive individ¬
2-7/7-2
uals who are most comfortable in very de¬
Clients with 2-7/7-2 codetypes are very com¬ pendent interpersonal relationships. They
mon in most types of psychiatric settings. are adept at inspiring others to take care of
These clients are anxious, tense, depressed, them and to protect them from their cruel
and constant worriers; this is apparent both fate. They may even persuade the clinician
to themselves and to others. They are guilt- to try to save them. That is, these clients are
ridden, intrapunitive individuals who can be¬ very successful at tapping any tendency a cli¬
come obsessively preoccupied with their nician may have toward rescuing or saving a
personal deficiencies despite frequent evi¬ client. Intervention into this chronic behav¬
dence of their personal achievements. Any ior pattern is fraught with problems even
problem is likely to be attributed to their per¬ for the experienced clinician, who must
sonal inadequacies. provide appropriate support and empathy
Suicidal ideation and plans should be as¬ while inducing significant behavior change
sessed carefully. Clinicians should routinely by confronting or interpreting the client’s
evaluate the client’s responses to items 150, behavior.
506, 520, and 524 on the MMPI-2 in all cli¬ Clients with 2-7-477-2-4 codetypes are
ents, and particularly in this codetype, since characterized by chronic, deeply ingrained
the item content reflects suicidal ideation depressive features in conjunction with exten¬
and/or suicide attempts. sive feelings of inadequacy and guilt. They
Clients often will report cardiovascular are self-deprecating and try to make others
symptoms that reflect their chronic state of feel superior by focusing on their weaknesses
tension and anxiety, insomnia, and decreased and inadequacies. They refuse to recognize
appetite. Because of their willingness to ex¬ their extensive dependency on others. The
amine their own behavior, often in infinite clinical extreme of the behavior shown by
detail, they are excellent candidates for psy¬ these clients is a psychotic depressive reac¬
chotherapy. They have a natural introspec¬ tion, although few clients with this codetype
tive orientation as well as the motivation to are diagnosed as such.
change, which augurs well for any psycho¬ They report financial difficulties, mari¬
therapeutic intervention. tal problerfis, and problems with alcohol. Be¬
If Scale 7 is extremely elevated (T score cause of the chronic, deep-seated nature of
> 85) and particularly if Scale 7 also is higher their problems and their reluctance to expose
than Scale 2, they may be so agitated and themselves to anxiety, prognosis for signifi¬
worried that these symptoms should be ad¬ cant behavior change is poor. Treatment
dressed through methods such as psycho- aimed at alleviating the depressive features
pharmacologic treatment or situational inter¬ may be most beneficial for short-term goals.
vention before treatment is initiated. Clients with 2-7-877-2-8 codetypes often
Since 2-777-2 codetypes occur so fre¬ appear in psychiatric settings. These clients
268 Chapter 6

have multiple neurotic symptoms that are of ence on the Wiener and Harmon (Wiener,
a chronic nature. Their major symptoms in¬ 1948) Obvious and Subtle subscales; the total
clude depression, nervousness, and obses¬ number of the Lachar and Wrobel (1979)
sions. They are ruminatively introspective critical items endorsed; Scales F, 6 (Para¬
and evidence excessive indecision, doubts, noia), and 0 (Social Introversion); and the
and worry. They complain of difficulties in Wiggins (1966) Content scales of Depression,
concentration and thinking. There is a real Organic Symptoms, and Psychoticism. Cli¬
question of whether these multiple neurotic ents with 8-2 codetypes appear to be report¬
symptoms actually mask a thought disorder. ing more psychological distress and to have a
Hence, careful evaluation for a thought dis¬ slightly higher probability of being psychotic.
order is indicated. This pattern of scores is consistent with the
They may have suicidal ruminations; differences in correlates reported below for
these need to be evaluated carefully. Again, these two codetypes.
clinicians are reminded to review the client’s Clients with 2-8 codetypes are experi¬
responses to items 150, 506, 520, and 524 on encing severe depression with associated
the MMPI-2. Such clients are withdrawn and anxiety and agitation. They frequently fear
socially introverted, and these characteristics loss of control. Their depression and agita¬
exacerbate their obsessive and ruminative be¬ tion are usually sufficient to produce con¬
haviors. Psychopharmacologic intervention fusion, forgetfulness, and difficulties in
is indicated frequently. Psychotherapeutic in¬ concentration and attention. They often dis¬
terventions should initially be directed to¬ play obsessive ruminations. Evaluation for a
ward solving the client’s immediate problems thought disorder may be appropriate. So¬
and should avoid any introspective type of matic complaints are common, such as dif¬
self-analysis. ficulty in sleeping or fatigue. These clients
Additional interpretive information on tend to withdraw and isolate themselves
2-7/7-2 codetypes and their triads may be from interpersonal relationships and activi¬
found in Gilberstadt and Duker (1965), ties, which will exacerbate their symptom¬
Gynther and colleagues (1973), Kelley and atology.
King (1979c, 1980), and Marks and associates Suicidal ideation is a prominent feature
(1974). Tables 6-17 and D-16 provide the and suicide attempts are quite frequent; sui¬
prototypic scores. cide potential should be evaluated carefully
as well as the client’s responses to items 150,
506, 520, and 524 on the MMPI-2. This code¬
2 8/8 2
-

type represents a chronic level of adjustment

Since there are a number of differences in the of marginal quality, so prognosis for inter¬
observed correlates of 2-8 and 8-2 codetypes, vention and change is poor.
each will be described separately. These dif¬ Clients with 8-2 codetypes are more
ferent correlates are somewhat surprising likely to evidence actual schizophrenic fea¬
since the prototypes for 2-8 and 8-2 codetypes tures in addition to the behaviors just de¬
are very similar and consequently only one scribed. Auditory and/or visual hallucina¬
prototype is provided for the 2-8/8-2 code- tions and systematized delusions may be
type. present. A careful evaluation for a thought
Clients with 8-2 codetypes scored one- disorder should be made. Somatic symptoms
half standard deviation higher or more on the of a bizarre nature may be seen.
following scales in Table D-17 than clients These clients also are depressed, iso¬
with 2-8 codetypes: the total T score differ¬ lated, and withdrawn. Suicidal ruminations
 Codetypes 269

TABLE 6-17 Prototypic Scores for 2-7/7-2 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 391 418
Age 40.3 12.7 38.8 12.8
Men 75.2% 58.2%
Women 24.8 41.8

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 10.1% 4.5%
Total (Obvious-Subtle)3 82.1 50.6 111.1 55.2
Critical items'3 38.3 10.5 40.3 10.9
Overreportedc 21.4% 34.8%
Underreported0 4.7% 0.7%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.6 6.1 4.3 5.9
L 49.8 7.3 51.8 10.1
F 62.9 8.4 66.8 14.3
K 48.9 7.2 42.7 7.8
1(Hs) 65.9 1 1.5 63.5 11.1
2(D) 89.8 1 1.3 84.3 9.7
3(Hy) 68.0 9.7 64.5 1 1.5
4(Pd) 70.5 9.8 63.8 10.2
5(Mf) 57.7 10.4 48.9 10.4
6(Pa) 66.1 9.4 63.0 11.1
7(Pt) 85.2 9.2 82.0 9.2
8(Sc) 74.4 1 1.2 69.6 1 1.0
9(Ma) 54.4 9.7 49.1 8.8
O(Si) 68.9 8.2 67.8 9.0

Supplementary Scales
M SD M SD
A 69.9 8.1 72.8 8.1
R 71.7 13.0 60.6 1 1.3
MAC/MAC-Rd
men 24.5 4.6 24.5 4.8
women 19.6 3.5 20.0 3.7

Codetype Concordance
Men Women Men Women
MMPI-2 2-7/7-2 65.4% 93.9% MMPI 2-7/7-2 78.9% 52.0%
2-8/8-2 12.0 10.7
2-4/4-2 10.2

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
270 Chapter 6

and suicide attempts are quite frequent; sui¬ junction with their socially introverted fea¬
cide potential should be evaluated carefully tures. They usually display feelings of inade¬
as well as the client’s responses to items 150, quacy, shyness, and isolation in social
506, 520, and 524 on the MMPI-2. This is a settings; these characteristics reflect their ac¬
chronic pattern of adjustment that usually re¬ tual ineptitude in and lack of social skills.
sults in psychiatric hospitalization. They generally are passive, conventional in¬
Gilberstadt and Duker (1965), Gynther dividuals who are unlikely to display either
and colleagues (1973), Kelley and King aggressive or delinquent behaviors.
(1979b, 1980), and Marks and colleagues Although they are unhappy and wor¬
(1974) provide additional interpretive infor¬ ried, their depression is chronically ingrained
mation on 2-8/8-2 codetypes. Table D-17 and they have little motivation to change. Es¬
provides the prototypic scores. sentially they have adjusted to their depres¬
sive state, and engaging them in psychothera¬
2-9/S-2 peutic endeavors may be difficult. Directly
addressing their lack of social skills through
Clients with 2-9/9-2 codetypes demonstrate
assertiveness training, role playing, or similar
an interesting example of a clinical scale con¬
approaches may be beneficial in helping them
figuration that might not be anticipated: cli¬
form meaningful interpersonal relationships.
ents who simultaneously evidence significant
Interpretive information on 2-0/0-2
depression and hypomanic tendencies. It is
codetypes is found in Kelley and King
frequently stated that 2-9Z9-2 codetypes are
(1979a), and Table D-19 provides the proto¬
characteristic of brain-damaged individuals
typic scores.
(cf. Lachar, 1974), but subsequent research
has not been able to replicate these findings
(see Farr & Martin [1988] for a review of this Spike 3
research).
Clients who elevate only Scale 3 (Hysteria)
Clients with a 2-9/9-2 codetype are likely
are characteristically socially conventional
to embody a hypomanic process that is no
individuals who emphasize their harmony
longer sufficient to obscure their depressive
with other people and express an almost un¬
features, at least on the MMPI. They may
assailable optimism. Even when faced with
evidence a severe, agitated depression, al¬
overwhelming failure in their life, they main¬
though the depressive features may be
tain that everything is going fine. They are
masked by their activity level. These clients
unwilling to express any type of feelings of
are narcissistically absorbed in their rumina¬
anger. They believe that they think clearly
tions and may evidence a psychotic process.
and logically, while others perceive them to
If a bipolar disorder is identified, psycho-
be flighty and unfocused.
pharmacologic intervention is indicated.
They are typically immature and egocen¬
Otherwise, psychological intervention is
tric, and display other histrionic features.
probably best directed toward the depressive
When under stress, they develop physical
features.
complaints that have obvious secondary gain
Table D-18 provides the prototypic
characteristics. They are extremely resistant
scores for 2-9/9-2 codetypes.
to entertaining the idea that psychological
factors could in any way be involved in their
2-0/0-2
current problems. Prognosis for significant
Clients with 2-0/0-2 codetypes often show behavior change is unlikely unless they can be
mild but chronic levels of depression in con¬ involved in long-term psychotherapy. Gener-
 Codetypes 271

ally, they will terminate any professional con¬ sequently, only one prototype is provided for
tact once they have weathered their current the 3-4/4-3 codetype.
crisis. These clients tend to be defensive,
Table D-20 provides the prototypic guarded, and unwilling to acknowledge psy¬
scores for Spike 3 codetypes. chological problems even when they are
readily apparent to others. They report little
3-1 psychological or emotional distress and de¬
scribe themselves as less depressed and anx¬
Clients with a 3-1 codetype characteristically
ious than do most psychiatric patients. Be¬
develop physical symptoms when under
havioral problems are more likely to occur in
stress. When the stress is alleviated, the phys¬
this codetype than in any other codetype that
ical complaints will disappear, only to reap¬
includes Scale 3.
pear when the client again experiences stress.
The relationship between Scales 3 and 4
There is rather obvious secondary gain to
serves as an index of whether persons will
their complaints. When physical symptoms
control and inhibit their socially unaccept¬
do appear, they are relatively restricted and
able impulses, particularly aggression and
specific both in location and nature, fre¬
hostility. Clients with a 3-4 codetype are im¬
quently involving pain in the extremities or
mature, egocentric individuals who discharge
the head.
their hostile feelings indirectly. Overtly, they
Some anxiety and nervousness usually
appear to be quiet and conforming. They
occur and are not controlled by their somatic
may become involved with acting-out indi¬
preoccupations. When emotional reactions
viduals and satisfy their own hostile tenden¬
occur, they are of short duration and ex¬
cies in a vicarious fashion.
pressed by crying and emotional lability.
Males may express fears of being homo¬
When the L and/or K scales also are elevated,
sexual, if Scale 5 also is elevated, but they are
these clients are likely to be extremely defen¬
unlikely to have overtly acted out their homo¬
sive, presenting themselves as exceedingly
sexual impulses. They report marital or fam¬
normal. Such clients are very threatened by
ily problems with little understanding of their
any suggestion of psychological problems.
role in them. These clients generally have
They are very unlikely to be psychotic.
chronic problems that are difficult to change.
Additional interpretive information on
Clients with a 4-3 codetype are character¬
3-1 codetypes is available in Gilberstadt and
ized by poorly controlled anger and hostility
Duker (1965), Gynther and associates (1973),
that is expressed in a cyclic fashion. Often
Marks and associates (1974), and Prokop
these clients are eventually incarcerated for
(1988). Table D-21 provides the prototypic
their violent acts. The presence of violent,
scores.
acting-out behavior in these persons has been
replicated in several studies (Davis & Sines,
3-4/43
1971; Persons & Marks, 1971). Buck and
Clients with 3-4 codetypes have been found Graham (1978) failed to replicate this find¬
to display different correlates than clients ing; this is not surprising, however, since
with 4-3 codetypes, so each of these will be Buck and Graham used a different popula¬
described in turn. These different correlates tion than the other studies.
are somewhat surprising since the prototypes Clients with a 4-3 codetype are quiet,
for 3-4 and 4-3 codetypes are virtually identi¬ withdrawn individuals, and their sudden out¬
cal and do not show even the few differences bursts come as a surprise to others. They dis¬
that were found with 2-8/8-2 codetypes. Con¬ play poor judgment under stress, but their
272 Chapter 6

emotional or violent outbursts may be only and have difficulty understanding why others
minimally related to external stress or provo¬ react to them the way they do.
cation. Their outbursts are not illogical or ir¬ The hostility, egocentricity, and un¬
rational unless Scale 8 also is elevated. The cooperativeness become readily apparent in
cyclic pattern of violent outbursts with inter¬ any relationship of more than a casual na¬
mittent periods of appropriate behavior rep¬ ture. Such individuals are unwilling to con¬
resents a chronic and stable personality disor¬ sider that psychological factors might be in¬
der that is extremely difficult to change. volved in their interpersonal problems. The
Davis and Sines (1971) could not find any possibility of paranoid or psychotic features
lasting effects of any form of psychological should be evaluated, even though these fea¬
intervention or incarceration. tures are relatively unusual. The prognosis
Gilberstadt and Duker (1965) and Kelley for significant behavior change is poor.
and King (1979a) provide additional interpre¬ Table D-24 provides the prototypic
tive information on 3-4/4-3 codetypes, and scores for 3-6/6-3 codetypes.
Table D-22 provides the prototypic scores.
3-7/7-3
3-5Z5-3
Clients with 3-7/7-3 codetypes are tense, anx¬
There is little information on S-5/5-3 ious individuals who develop chronic physi¬
codetypes. College-educated males with his¬ cal ailments in the head or the extremities
trionic features would be expected to achieve resulting from psychological stress and con¬
this codetype. They see themselves as being flicts. They tend to highly ruminative and ob¬
well adjusted, happy, calm, and self-confi¬ sessive. Despite the overt behavioral evidence
dent. Others see them as manipulative, of tension and anxiety, they will deny the ex¬
crafty, immature, and demanding. They do istence of psychological problems and be un¬
not report behavioral problems or getting concerned about their physical ailments.
into trouble because of their behavior. They They are fearful and frequently phobic. They
have many physical problems, and they may feel depressed and have problems sleeping.
be narcissistic. They are seldom self-referred. Their lack of insight into their histrionic
Interpretive information on this code¬ mechanisms makes psychological interven¬
type is found in Tanner (1990). Table D-23 tion slow and arduous at best.
provides the prototypic scores for 3-5Z5-3 Table D-25 provides the prototypic
codetypes. scores for 3-7/7-3 codetypes.

3 6/6-3 3-8Z8-3

Clients with 3-6/6-3 codetypes are seen as Clients with 3-8/8-3 codetypes are seen as
angry, hostile individuals who attempt to strange and peculiar individuals who com¬
deny or rationalize these feelings. They deny plain of difficulties in thinking and concen¬
their hostility and suspiciousness, even tration. They are likely to be experiencing
though these attributes are apparent to every¬ significant psychological distress despite their
one but themselves. Their anger and hostility attempts to deny and repress problems. Psy¬
are frequently directed toward family mem¬ chological stress is converted into physical
bers, often in indirect, passive ways. They symptoms, which may consist of headaches,
often deny and resent any kind of psycholog¬ insomnia, fatigue, or bizarre complaints.
ical interpretation of their problems. They They display histrionic features, such as
perceive their relationships in positive terms immaturity, egocentricity, and dependency,
 Codetypes 273

as well as hostility, tension, and worry. They these two scales should be antithetical, and
may actually be psychotic, and the possibility the infrequent occurrence of this codetype
of a thought disorder should be evaluated seems to support this contention. The rarity
carefully. When psychotic reactions are seen, of occurrence is all the more remarkable since
there are infantile, narcissistic qualities ac¬ other scales that supposedly tap antithetical
companied by behavioral regression. They processes are not infrequently seen, such as
are often emotionally inappropriate, apa¬ 2-9/9-2 and 4-7/7-4.
thetic, and fearful. These clients describe themselves as
Supportive procedures that bolster the being very conventional, law-abiding individ¬
histrionic defenses are frequently sufficient uals who do not get in trouble because of
to weather the current crisis. Any form of in- their behavior. They are very shy, with¬
sight-oriented psychotherapy should be con¬ drawn, and socially reserved. They tend to
sidered carefully because of the possibility of isolate themselves and avoid interacting with
an underlying psychotic process. others. They are not experiencing any degree
Marks and colleagues (1974) furnish ad¬ of psychological or emotional distress.
ditional interpretive information on this Table D-28 provides the prototypic
codetype, and Table D-26 provides the pro¬ scores for 3-0/0-3 codetypes.
totype scores.

Spike 4
3-9Z9-3
Clients who elevate only Scale 4 (Psycho¬
Clients with 3-9/9-3 codetypes are gregari¬
pathic Deviate) may show impulsive behav¬
ous, dramatic, outgoing individuals who
ior, rebelliousness, and poor relationships
readily make their presence known in any so¬
with authority figures. They are likely to be
cial setting. This configuration becomes par¬
seen as egocentric, lacking insight, and shal¬
ticularly pathognomic if Scale 0 (Social
low in their feelings for others. They have a
Introversion) approaches a T score of 30 (see
low tolerance for frustration, and this quality
Scale 0 in Chapter 4). These clients report
combined with poorly controlled anger and
acute attacks of physical symptoms, such as
poor self-control often results in outbursts of
chest pains, cardiovascular problems, and
physical aggression. They form quick, super¬
headaches. The symptoms improve rapidly
ficial relationships, but have difficulty in
with medical intervention only to recur at a
more intimate interpersonal relationships.
later date. These clients are hostile, irritable,
Scale 0 is typically low (see Table 6-5
and emotionally labile.
[MMPI] or Table 6-10 [MMPI-2]), and these
They are unwilling to examine psycho¬
qualities become more pathognomic as Scale
logical factors in their behavior and will for¬
0 approaches a T score of 30. Problems with
get treatment once their physical symptoms
substance abuse are quite frequent, and the
abate.
MacAndrew Alcoholism scale (see Chapter 5)
Kelley and King (1979a) provide addi¬
should be scored for these individuals if it not
tional interpretive information on 3-9/9-3
routinely scored.
codetypes, and Table D-27 provides the pro¬
These clients are rarely self-referred;
totype scores.
rather a social agency usually refers them. As
soon as the social agency no longer requires
3-0/0-3
treatment, the client will discontinue therapy.
There is little information on 3-0/0-3 code¬ The psychopathic qualities found in these cli¬
types. The underlying processes tapped by ents make them poor candidates for psycho-
274 Chapter 6

therapy. They conceive of themselves as tle emotional distress. They think clearly and
being better than other people, externalize rationally and report good insight into their
blame for their problems onto others, and behavior.
have little insight into their own behavior. Males with 4-5/5-4 codetypes are pas¬
These psychopathic qualities tend to be sively unconventional, usually in both their
alleviated by age; maturation is often seen as appearance and their behavior. They are re¬
a viable though long-term treatment alterna¬ belling against social conventions and mores
tive. If they do not change with age, some and delight in defying and challenging any
form of vocational counseling and career form of rule or regulation. Both dominance
guidance that draws on the clients’ “skills and dependence are key issues for them, and
and talents” may be useful. A low Scale 2 (T the supplementary scales for measuring these
score < 50) in combination with a Spike 4 traits should be scored (see Chapter 5).
codetype of even moderate elevation indi¬ Although they have strong needs for de¬
cates an especially low probability of behav¬ pendency, they fear domination by signifi¬
ior change. cant others. They may be concerned about
Several investigators have reported that homoerotic impulses, and a subset of per¬
Spike 4 codetypes occur frequently in college sons with this codetype will be actively ho¬
students and have questioned whether new mosexual. These active homosexuals dis¬
norms are needed (cf. Goodstein, 1954). King play this codetype only if they want to
and Kelley (1977a) found in their college stu¬ acknowledge openly their homosexual status.
dent sample that Spike 4 codetypes were in¬ If they so desire, they can deny the presence
dicative of significant psychopathology, and of homosexual impulses or behaviors with¬
they concluded that new norms were not out being detected by any validity or clini¬
needed in their psychiatric outpatient clinic. cal scale configuration. Hence, the detection
For interpretive information on Spike 4 of homosexual behavior in 4-5/5-4 codetypes
codetypes, consult Gilberstadt and Duker is best accomplished when the clinician di¬
(1965) and King and Kelley (1977a), and see rectly asks the client about such activity. Er¬
Table D-29 for prototypic scores. ickson, Luxenberg, Walbek, and Seely (1987)
reported that 4-5/5-4 codetypes were among
the most frequent in their sample of sex of¬
4-5/B-4
fenders.
Clients with 4-5/5-4 codetypes are almost al¬ Male college students with 4-5/5-4 code¬
ways men because of the infrequency with types are described as experiencing general
which women elevate Scale 5 on the MMPI interpersonal difficulties of a transient nature
above a T score of 70. The one major excep¬ with only slight indications of a significant
tion to this statement is the finding by Sutker, personality disorder (King & Kelley, 1977b).
Allain, and Geyer (1978) that this MMPI Although these students are passive and expe¬
codetype was the most frequent (23 percent) riencing heterosexual adjustment problems,
among women convicted of murder. It does homosexuality is not characteristic of them.
not appear that women are much more likely The probability for significant behavior
to produce this codetype on the MMPI-2 (see change is poor because of the chronic, in¬
Table 6-7). grained nature of their personality features.
These clients are very satisfied with King and Kelley (1977b) and Tanner
themselves and their behavior. They also are (1990) provide interpretive information
very defensive and guarded about revealing about 4-5/5-4 codetypes, and Table D-30
themselves. As a consequence, they report lit¬ provides the prototypic scores.
 Codetypes 275

4-6Z6-4 4-7/7-4

Clients with 4-7/7-4 codetypes are an exam¬

Clients with 4-6/6-4 codetypes are angry, re¬
ple of the apparently paradoxic elevation of
sentful, argumentative individuals who are
two antithetical scales. Since Scale 4 (Psycho¬
difficult to interact with personally or so¬
pathic Deviate) assesses persons’ insensitivity
cially because of these characteristics. They
to and disregard of the social consequences of
usually are able to control the acting out of
their behavior, and Scale 7 (Psychasthenia)
their hostility but do exhibit violent out¬
assesses persons’ excessive concern about and
bursts on occasion. They will externalize
analysis of their own behavior, it does not
blame for their anger. They often express
seem likely that simultaneous elevations on
rather vague emotional and physical com¬
these two scales should occur. Nevertheless,
plaints and they report feeling depressed
they do.
and anxious. They are suspicious of the mo¬
The primary characteristic of these
tives of other people, and the likelihood of
clients is cyclic behavior between these
paranoid features should be examined
two extremes. It is as if Scale 4 character¬
closely.
istics become dominant for a period of time
These clients have a long history of se¬
and the clients act out impulsively with lit¬
vere social maladjustment with poor work
tle regard for social conventions and the
histories. Poor interpersonal relationships as
needs and wishes of other people. Follow¬
well as marital problems are quite common.
ing these periods of acting out, the Scale 7
They are difficult to interact with because of
characteristics become dominant, and the
their hostile attitudes and behaviors. Conse¬
clients will feel guilty, remorseful, and self-
quently, they are poor candidates for any
deprecating about having exhibited such
type of psychological intervention. In
behaviors. They appear to be over¬
women with a 4-6/6-4 codetype it is impor¬
controlled during this phase, but these
tant to know the elevation on Scale 5 to un¬
controls are not sufficient to prevent recur¬
derstand how these behaviors will be exhib¬
rences of acting out.
ited (see page 163).
The clinician will encounter these clients
When Scale 8 is also elevated in this
during their guilty, remorseful phase, and the
4-6Z6-4 codetype, the process is even more
clinician is likely to believe that significant
malignant. In addition to the above charac¬
therapeutic progress is being made. All too
teristics, these clients are even more evasive
often, however, these individuals will again
and defensive about admitting any form of
act out impulsively and return feeling more
psychological problem, and difficulties in
guilty, and so on. Successfully intervening
logic and judgment begin to appear. They
in this behavior cycle is extremely difficult
seethe with anger, which in conjunction with
without a long-term therapeutic relation¬
their sensitivity to criticism and suspicious¬
ship.
ness can lead to unpredictable and irrational
Kelley and King (1979a) provide inter¬
violent outbursts. They are likely to be
pretive information on 4-7/7-4 codetypes,
openly defiant and hostile. Their solution for
and Table D-32 provides the prototypic
behavior change is to have others change to
scores.
meet their expectations.
Additional interpretive information is
4-818-4
available in Marks and associates (1974), and
Table D-31 provides the prototypic scores Clients with 4-8/8-4 codetypes are typically
for 4-6/6-4 codetypes. characterized by a chronic marginal schizoid
276 Chapter 6

adjustment if they are not actually schizo¬ ships are shallow and superficial. They ex¬
phrenic. They have difficulty with close, hibit an enduring tendency to get into trou¬
emotional relationships, distrust others, and ble, usually only in a way that damages their
are socially withdrawn. They are dissatisfied own or their family’s reputation, although anti¬
with their relationships with other people, social and criminal acts are not uncommon.
but their angry, resentful qualities, which This codetype is common in clients with
they have difficulty modulating or express¬ marital problems, illegitimate pregnancies,
ing, only serve to exacerbate their alienation child abuse (Paulson, Afifi, Thomason, &
from others. They see the world as dangerous Chaleff, 1974), alcohol or drug abuse
and other people as rejecting and unreliable. (Greene & Garvin, 1988; Hodo & Fowler,
They are moody and emotionally inappropri¬ 1976; Toper, Kammeier, & Hoffmann, 1973),
ate. Suicide attempts are quite frequent and delinquency, repeated crimes of indecent expo¬
should be evaluated carefully. Clinicians sure (McCreary, 1975), and sex offenders (Er¬
should note the client’s responses to items ickson et ah, 1987; Hall et ah, 1986).
150, 506, 520, and 524 on the MMPI-2. These clients have an inordinate need for
Their behavior is typically unpredictable excitement, and their acting out frequently
and nonconforming at best. They frequently serves these purposes. They temporarily cre¬
get into social and legal difficulties because ate a favorable impression because they are
of their judgment and their problems in logic socially facile and are free from anxiety,
and thinking. A history of criminal activity worry, and guilt. Their judgment, however,
with numerous arrests is common. Their is notably poor, and they do not seem to learn
crimes are often poorly planned and executed from experience. They lack the ability to
and may involve bizarre or violent behaviors. postpone gratification and have difficulty in
This is a frequent codetype among child mo¬ any activity requiring sustained effort.
lesters (Hall, Maiuro, Vitaliano, & Proctor, Sheppard, Smith, and Rosenbaum (1988)
1986; McCreary, 1975), rapists (Armentrout found that 50 percent of their patients within
& Hauer, 1978), rapists and exposers (Rader, a 4-9Z9-4 cluster type did not complete a 30-
1977), sex offenders (Erickson et ah, 1987). day alcohol treatment program.
These clients are chronically maladjusted, Acting out is the primary defense mech¬
which indicates that any form of psychologi¬ anism used by these individuals, although ra¬
cal intervention will be of limited benefit. tionalization also plays an important role.
Additional interpretive information on When Scales 6 and 8 are also elevated and
4-8/8-4 codetypes may be found in Gynther with higher elevations on Scale 9, there is an
and colleagues (1973) and Marks and col¬ increased probability of acting-out episodes
leagues (1974), and Table D-33 provides the that may be quite intense and violent. These
prototypic scores. persons exhibit very stable personality pat¬
terns, and they have poor prognosis for be¬
havior change or even participation in therapy.
4-9Z9-4
Huesmann, Lefkowitz, and Eron (1978)
Clients with 4-9/9-4 codetypes are likely to reported that the sum of the T scores on
display some form of acting-out behavior. Scales F, 4, and 9 was a valid predictor of ag¬
The hypomania seemingly energizes or acti¬ gression in older adolescents. Normal males
vates the behaviors assessed by Scale 4 (Psy¬ (M = 183.3) and females (M = 178.9) scored
chopathic Deviate). These clients are significantly lower than delinquent males (M
overactive, impulsive, irresponsible, and un¬ = 217.4) and females (M = 237.7) on this
trustworthy. Their interpersonal relation¬ index. Huesmann and associates did not re-
 Codetypes 277

port standard deviations by group or gender; ity-Femininity). The paucity of data reflects
they stated that the composite standard devi¬ the fact that Scale 5 was not among the origi¬
ation was about 25. Since there is substantial nal clinical scales, and thus it was not in¬
overlap in these distributions, it remains to be cluded in the early clinical research on the
seen how useful this index will be for predict¬ MMPI. Only King and Kelley (1977b) and
ing aggression in individual clients. Tanner (1990) have examined the behavioral
Additional interpretive information on correlates of codetypes in which Scale 5 is one
4-9/9-4 codetypes is reported by Gilberstadt of the high-point scales.
and Duker (1965), Gynther and colleagues In addition, Scale 5 is a frequently occur¬
(1973), King and Kelley (1977a), and Marks ring high-point only in normal college edu¬
and colleagues (1974). Tables 6-18 and D-34 cated men with liberal arts majors and
provide the prototypic scores. avowed homosexuals who are not trying to
hide their homoerotic behavior. The former
4-0/0-4 group is rarely of interest to most clinical re¬
searchers, and the latter group is only starting
Clients with 4-0/0-4 codetypes are statisti¬ to be recognized and evaluated. The psycho¬
cally rare (see Tables 6-1 to 6-4). In fact, this metric qualities of Scale 5 on the MMPI make
codetype is so unusual that it is not listed in it very difficult for women to have this scale
any of the major MMPI references. Theoret¬ as a high point (it is unusual for most women
ically, this infrequency of occurrence makes to score above a T score of 60).
sense because psychopathic persons who typ¬ Finally, it is the least well-developed and
ically elevate Scale 4 (Psychopathic Deviate) standardized scale of the individual scales on
are unconcerned and lack anxiety about in¬ the MMPI, which also has contributed to its
terpersonal relationships and consequently lack of attention. It remains to be see whether
should score low on Scale 0 (Social Intro¬ Scale 5 will fare any better on the MMPI-2.
version). Hence, at a theoretical level, no These clients describe themselves as self-
specific classification of psychopathology confident, easygoing, and assured. They re¬
would be expected to show this particular late easily to others and will make a good impres¬
codetype. sion on them. They are in good physical health
These clients are experiencing little emo¬ and experiencing few psychological problems.
tional distress and they are generally free King and Kelley (1977b) found that stu¬
from anxiety and guilt. They are in good dent outpatients at a university mental health
physical health. They report problems inter¬ center with Spike 5 codetypes were generally
acting with their family members and persons normal students with no significant psycho¬
in positions of authority. pathology despite their seeking treatment.
They are shy, retiring individuals who More information on Spike 5 codetypes
avoid social interactions. They tend to be is available in King and Kelley (1977b), and
very conventional and do not get into trouble Table D-36 provides the prototypic scores.
because of their behavior.
Table D-35 provides the prototypic
scores for 4-0/0-4 codetypes. 5-6Z6-5

There is little information 5-6/6-5 codetypes.

These clients are reluctant to expose them¬
Spike 5
selves to others and they frequently are de¬
There is little information on the empirical scribed as being guarded, resistant, aloof,
correlates of elevations on Scale 5 (Masculin¬ impulsive, abrasive, irritable, and easily an-
278 Chapter 6

TABLE 6-18 Prototypic Scores for 4-9Z9-4 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 522 318
Age 30.6 12.3 29.9 1 1.8
Men 75.1% 73.0%
Women 24.9 27.0

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 3.0% 2.9%
Total (Obvious-Subtle)3 20.2 56.9 46.6 60.0
Critical itemsb 32.2 1 2.1 33.7 12.4
Overreportedc 2.7% 9.4%
Underreported0 28.3% 20.1 %

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.5 5.9 3.8 5.0
L 48.9 7.0 50.8 9.9
F 63.6 10.5 67.3 16.1
K 51.3 8.0 45.9 8.7
1(Hs) 54.5 9.5 52.6 10.0
2(D) 56.0 9.9 53.2 9.6
3(Hy) 58.0 8.6 52.2 10.0
4(Pd) 81.3 7.3 76.1 7.7
5(Mf) 55.3 9.6 47.2 9.6
6(Pa) 62.4 9.9 57.6 10.9
7(Pt) 59.8 8.7 55.5 9.5
8(Sc) 65.4 9.5 60.1 10.0
9(Ma) 79.3 7.2 75.7 8.0
0(Si) 47.7 7.1 45.4 7.7

Supplementary Scales
M SD M SD
A 54.5 9.9 57.2 10.3
R 52.5 10.2 43.4 8.9
MAC/M AC-Rd
men 31.0 3.8 31.4 3.7
women 28.1 4.1 27.9 3.8

Codetype Concordance
Men Women Men Women
MMPI-2 4-9Z9-4 55.8% 64.6% MMPI 4-9Z9-4 89.7% 96.6%
Spike 9 15.8

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 7 5th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Codetypes 279

gered. They have difficulties with social rela¬ themselves and in their relationships with
tionships, especially heterosexual relation¬ others. They also have many family prob¬
ships. lems. They are shy, reserved individuals who
They have little insight into their own be¬ avoid interpersonal contact.
havior since they describe themselves as being More information is presented in King
happy, self-confident, and as having good and Kelley (1977b) and Tanner (1990). Table
judgment. They have strong religious and po¬ D-39 provides the prototypic scores for
litical beliefs that they express directly. They 5-8/8-5 codetypes.
see themselves as mixing easily and enjoying
being in social situations.
5-979-5
Interpretive information on 5-676-5
codetypes is provided by Tanner (1990), and There is little information on 5-979-5
Table D-37 provides the prototypic scores codetypes. These clients are very comfortable
for 5-6/6-5 codetypes. with themselves and their behavior. They do
not report any form of psychological prob¬
5-7/7-5 lems, emotional distress, or physical symp¬
toms. They like to keep active and become
Clients with 5-7/7-5 codetypes are more
bored easily; hyperactive thoughts and be¬
likely to be depressed than anxious. They
haviors are possible. They describe them¬
have restricted affect and lack spontaneity.
selves as self-confident, easygoing, and
They are bothered by recurring thoughts and
assured. They relate easily to others and
ideas and mild depressive symptoms. They
make a good social impression, but they have
are easily excited. They are in good physical
extremely unstable heterosexual relation¬
health.
ships. These clients prefer to be leaders rather
These clients are shy, bashful, and easily
than followers in any activity. They seldom
embarrassed. They are more sensitive than
are self-referred. They tend to use alcohol to
most people and their feelings are easily hurt.
reduce anxiety with a relatively high inci¬
They have interpersonal and/or heterosexual
dence of violence when drinking.
difficulties.
See Tanner (1990) for additional infor¬
See King and Kelley (1977b) and Tanner
mation. Table D-40 provides the prototypic
(1990) for more information. Table D-38
scores for 5-979-5 codetypes.
provides the prototypic scores for 5-777-5
codetypes.
5-0/0-5
5-878-5
There is little information for 5-070-5
Clients with 5-878-5 codetypes are likely to codetypes. These clients see themselves as ex¬
have a family history of alcohol abuse, men¬ tremely well adjusted. They do not report any
tal illness, and physical abuse. They fre¬ psychological problems, emotional distress,
quently have long psychiatric histories that or physical symptoms; they do not appear to
began in childhood. They report reactive de¬ be defensive or guarded. They describe them¬
pression, paresthesia, and religious preoccu¬ selves as happy and contented with their
pations but have intact thought processes. lives. Clients are socially introverted and tend
They can be described as odd, eccentric indi¬ to withdraw from social contact. They are
viduals who have difficulty making emo¬ easily embarrassed in social situations. They
tional contact with others. They report rarely get into trouble because of their behav¬
numerous conflicts over sexuality both in ior.
280 Chapter 6

Table D-41 provides the prototypic that they do maintain will be tinged with re¬
scores for 5-0/0-5 codetypes. sentment, suspiciousness, and hostility.
Their behavior is frequently unpredict¬
able and socially inappropriate. They may be
Spike 6
preoccupied about and ruminate over ab¬
Clients who elevate only Scale 6 (Paranoia) stract, theoretical issues, religion, and sexual
are relatively rare. Since the item content on themes. General apathy may permeate all of
Scale 6 is obvious, the paranoid symptom¬ their behavior. Behavioral regression, autis¬
atology is usually evident in clients with Spike tic thought processes, inappropriate affect,
6profiles, particularly if the Dominance scale and bizarre associations may be seen. Diffi¬
is higher than the Dependency scale on the culties in concentration and attention, mem¬
MMPI (see Chapter 5). The possibility of a ory deficits, and poor judgment are quite
paranoid process should be evaluated. These common.
clients are suspicious, distrustful, and project These clients are severely and chroni¬
blame for their problems onto others. cally maladjusted even if they are not actu¬
Table D-42 provides the prototypic ally psychotic. They are suspicious and dis¬
scores for Spike 6 codetypes. trustful of others and have poor social skills.
They generally feel apathetic, socially iso¬
6-7/7-6 lated, and withdrawn.
Gilberstadt and Duker (1965), Gynther
Clients with 6-7/7-6 codetypes are anxious,
and colleagues (1973), and Marks and col¬
worried, suspicious individuals. They are also
leagues (1974) present further interpretive in¬
rigid, hypersensitive, and stubborn. They
formation on 6-878-6 codetypes. Tables 6-19
brood and ruminate over both their own and
and D-44 provide the prototypic scores.
others’ behavior. They express hostile feel¬
ings indirectly. They are unlikely to be actu¬
ally paranoid. Their problems tend to be 6-979-6
chronic and characterologic in nature.
Clients with 6-979-6 codetypes are usually en¬
They tend to keep people at a distance
countered only in inpatient settings. They are
and to have poor social judgment. Their hy¬
angry, hostile individuals who may exhibit
persensitivity and tendency to misinterpret
grandiosity and egocentricity. They are also
the statements and behaviors of others often
irritable, excited, and energetic. They may re¬
leads to volatile and distant relationships.
port difficulty thinking and concentrating,
Kelley and King (1979a) provide interpre¬
and exercise poor judgment.
tive information on 6-777-6 codetypes, and
The presence of a psychotic process that
Table D-43 provides the prototypic scores.
is more likely to reflect a mood disorder than
a thought disorder should be considered.
6-878-6
They describe themselves as happy, calm,
Clients with 6-878-6 codetypes are likely to easygoing, and in good physical health. Oth¬
evidence a thought disorder with paranoid ers see them as angry, hostile, and over-
features as in paranoid schizophrenia. Sys¬ responsive to minor stresses and problems; less
tematized delusions may be present. Such in¬ frequently they are seen as tense and anxious.
dividuals express significant personal stress Further information is found in Gynther
through their complaints of tension, worry, and associates (1973) and Marks and associ¬
depression, and so on. They are socially iso¬ ates (1974). Table D-45 provides the proto¬
lated and withdrawn. Any social relationship typic scores for 6-979-6 codetypes.
 Codetypes 281

TABLE 6-19 Prototypic Scores for 6-8Z8-6 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 678 863
Age 30.4 12.2 29.8 12.0
Men 67.9% 74.3%
Women 36.1 25.7

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 43.0% 28.7%
Total (Obvious-Subtle)3 1 59.6 68.3 202.5 70.8
Critical itemsb 64.0 17.2 66.2 16.5
Overreported0 41.1% 84.5%
Underreported0 1.1% 0.5%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 5.1 6.3 4.8 5.9
L 49.4 7.9 50.5 10.5
F 92.5 14.9 105.0 17.3
K 42.8 7.2 36.3 7.3
1(Hs) 70.3 1 5.1 69.4 14.3
2(D) 75.8 14.9 73.2 13.2
3(Hy) 67.1 1 2.1 64.6 1 5.3
4(Pd) 79.4 12.2 73.7 12.9
5(Mf) 60.3 1 1.3 54.0 10.5
6(Pa) 95.3 1 1.4 97.4 1 3.5
7(Pt) 80.7 1 3.4 78.0 13.7
8(Sc) 100.5 1 5.1 96.0 14.5
9(Ma) 75.2 10.9 69.6 12.3
O(Si) 65.5 9.4 65.7 10.2

Supplementary Scales
M SD M SD
A 72.0 9.5 76.1 9.6
R 59.2 14.0 49.9 12.2
MAC/MAC-Rd
men 28.6 4.4 28.0 4.4
women 25.5 4.4 25.8 4.2

Codetype Concordance
Men Women Men Women
MMPI-2 6 8/8-6 86.1 % 75.2% MMPI 6-8Z8-6 58.0% 82.6%
7-8/8-7 1 2.2
4-8Z8-4 1 1.6
2-8/8-2 10.9

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
282 Chapter 6

6-0/0-6 havior. When Scale 7 is higher than Scale 8,

they are still resisting the establishment of seri¬
There is little information on 6-0/0-6
ous thought and behavior disorders. That is,
codetypes. These clients see themselves as
the thoughts and behaviors that they are experi¬
calm, happy, well adjusted, and in good
encing are still upsetting and distressing to them,
physical health. They believe that they think
which is a positive clinical sign. When Scale 8 is
clearly and have good judgment and mem¬
higher than Scale 7, they may be adapting to the
ory. They are not experiencing any form of
presence of serious psychopathology, which
psychological or emotional distress.
makes intervention more difficult.
They are very shy, bashful, and easily
Clients with 7-878-7 codetypes are chron¬
embarrassed in social situations. Their feel¬
ically worried, tense, and agitated. They are
ings are easily hurt and they are unlikely to be
socially uncomfortable and have poor social
assertive or confrontive with others. They
skills and judgment. They have difficulty
have pleasant relationships with others.
forming close interpersonal relationships and
Table D-46 provides the prototypic
usually are withdrawn, isolated, and intro¬
scores for 6-0/0-6codetypes.
verted. These characteristics exacerbate their
obsessive ruminations. They often engage in
Spike 7 sexual fantasies and their sexual adjustment
is poor. Hall and colleagues (1986) found this
Clients who elevate only Scale 7 (Psychasthe-
codetype to occur frequently in men who sex¬
nia) are relatively uncommon because of the
ually assaulted children. They should be eval¬
extensive item overlap between Scale 7 and
uated carefully for the presence of a thought
the other clinical scales, especially Scales 2
disorder; a diagnosis of schizophrenia is very
and 8. Clients with Spike 7codetypes are seen
common.
as mildly tense and anxious. They usually ex¬
Psychopharmacologic interventions are
hibit obsessive-compulsive defenses that are
difficult because of the ingrained nature of
no longer able to handle their problems com¬
their psychological conflicts and because of
pletely. Phobias, compulsions, and obses¬
their difficulty in forming interpersonal rela¬
sions may be seen. They are easily frightened.
tionships.
They are very shy, reserved individuals
See Gilberstadt and Duker (1965),
who become embarrassed easily. Their social
Gynther and colleagues (1973), and Kelley
isolation and fear of social interaction keeps
and King (1980) for further interpretive in¬
them out of behavioral problems.
formation on 7-8/8-7 codetypes. Table D-48
These clients have a very stable personal¬
provides the prototypic scores.
ity structure, which requires long-term psy¬
chological treatment to produce significant
behavior change. They are unlikely to require 7-9/9-7
hospitalization, although their symptoms
Clients with 7-979-7 codetypes are chroni¬
may seriously interfere with their jobs and
cally worried, tense, agitated, and depressed.
general interpersonal relationships.
Their high energy level seems only to enhance
Table D-47 provides the prototypic
their obsessive ruminations. They talk exces¬
scores for Spike 7 codetypes.
sively about unconnected ideas, which may
be difficult to follow. They may alternate be¬
7-8/8- 7
tween periods of impulsive acting out and pe¬
Clients with 7-8/8-7 codetypes share numer¬ riods of guilt and self-deprecation. The
ous clinical features in common, but scale possibility of manic or hypomanic features
order is important in understanding their be¬ should be investigated. They find it difficult
 Codetypes 283

to relax and interrupt their obsessive rumina¬ A subset of clients with Spike 8 codetypes
tions about their fears and problems. manifest serious psychopathology; review of
Their interpersonal relationships are the supplementary scales (see Chapter 5) will
often awkward since their inconsiderate and assist in their identification. Sexual preoccu¬
impulsive behavior makes it difficult to es¬ pation is frequent along with sexual confu¬
tablish contact at more than a superficial sion and bizarre sexual fantasies. These cli¬
level. They are immature and self-centered. ents also are mentally disorganized, have
Psychopharmacologic intervention may vague goals, are indecisive and unhappy,
be necessary because of their excessive level worry a great deal, and suffer from insomnia.
of anxiety. A careful evaluation to rule out the presence
Kelley and King (1979a) provide interpre¬ of a thought disorder is indicated in all cases.
tive information on 7-9/9-1 codetypes, and These clients may be psychotic, with feel¬
Table D-49 provides the prototypic scores. ings of unreality, memory difficulties, and
confused or bizarre thoughts or beliefs. Hal¬
7-0/0-7 lucinations, psychomotor retardation, and
withdrawal are possible. A history of psychi¬
Clients with 7-0/0-7 codetypes are uncom¬ atric hospitalizations is common in these cli¬
mon. They report a mild, general dissatisfac¬ ents. The chronic and characterologic nature
tion with their lives. Their symptoms tend to of the problems represented in a Spike 8
be vague and nonspecific. They are unlikely codetype also does not augur well for thera¬
to be psychotic. peutic change.
They are very shy, reserved individuals Table D-51 provides the prototypic
who become embarrassed easily. They lack scores for Spike 8 codetypes.
self-confidence and are easily threatened or
intimidated by others. Their social isolation
and fear of social interaction keeps them out 8-979-8
of behavioral problems. Clients with 8-979-8 codetypes evidence seri¬
Table D-50 provides the prototypic ous psychopathology, even when these scales
scores for 7-0/0-7codetypes. are only slightly elevated above a T score of
65 on the MMPI-2 or a T score of 70 on the
MMPI. They have a rapid onset of excite¬
Spike 8
ment, confusion, disorientation, and hyper¬
Clients who elevate only Scale 8 (Schizophre¬ activity. They may have difficulty thinking
nia) relate poorly to other people and tend to and concentrating. They excessively engage
escape from their own unacceptable impulses in daydreaming and fantasy, and their reality
into need-fulfillment fantasies. Some origi¬ testing may be marginal. They are depressed,
nal, unusual, or eccentric qualities may be anxious, hostile, and irritable. They are un¬
present in their thinking. At best, these clients predictable and prone to act out unexpect¬
have abstract theoretical or philosophical in¬ edly. Most of these clients have a psychotic
terests and a tendency not to conform, al¬ process either of a bipolar manic or schizo¬
though occasionally an intelligent and phrenic type. Autistic thinking, hallucina¬
creative person will produce a Spike 8 code¬ tions, and delusions are frequent.
type. They are confident in social situations They are fearful of relating to others and
and are not self-conscious. They tend to be focusing on a specific topic or idea, which in¬
aloof from others. They believe that their terferes with their interpersonal relationships
memory, concentration, and judgment are both at home and on the job. Their relation¬
good. ships are marked by distrust, suspicion, and
284 Chapter 6

anger. They are boastful, emotionally labile, problems. They are extroverted, outgoing,
and egocentric. and sociable in their interpersonal relation¬
They are difficult to work with therapeu¬ ships, but these relationships have no real
tically because of their social withdrawal and depth or intimacy. They are rebellious and
fear of relating to others. They also flit so hostile, and they have difficulty controlling
rapidly from topic to topic that addressing a their impulses. They are grandiose, hyperac¬
specific issue in a therapeutic relationship is tive, talkative, and not depressed. They also
very difficult. believe that they have good memory, judg¬
Gilberstadt and Duker (1965), Gynther ment, and concentration.
and associates (1973), and Marks and col¬ These clients are in good physical health
leagues (1974) have additional interpretive in¬ and do not tire easily. They may evidence
formation on 8-9/9-8 codetypes. Table D-52 manic features, which should be carefully in¬
provides the prototypic scores. vestigated. Their thought processes may be¬
come bizarre during acute hypomanic phases,
8 0/0-8 and they may become extremely belligerent if
their grandiose plans are interrupted. If
Clients with 8-0/0-8 codetypes are relatively manic features are evident, a psychopharma-
infrequent despite the common theme of cologic intervention is indicated.
problems with interpersonal relationships Further interpretive information is avail¬
that is characteristic of both scales. They de¬ able in Gilberstadt and Duker (1965) on
scribe themselves as being mildly depressed Spike 9 codetypes. Table D-54 provides the
and anxious and getting little pleasure from prototypic scores.
life. Their family life is conflictual and unsat¬
isfying. They have trouble making decisions
9-0/0-9
and they worry what others may think about
the decisions they have made. They are easily Clients with 9-0/0-9 codetypes are very un¬
frightened and frequently phobic. common. They describe themselves as being
These clients are extremely introverted, happy and well adjusted. They are in good
shy, and bashful. They are withdrawn, so¬ physical health and experience few pains or
cially isolated, and avoid interpersonal rela¬ physical discomforts. They are elated, fre¬
tionships. They are very uncomfortable in quently excited individuals who definitely are
any type of social setting or interaction. They not depressed.
give up easily and avoid conflict whenever These clients tend to be socially shy and
possible. withdrawn, although they have adequate so¬
They are generally nonverbal, which cial skills. They are egocentric, self-confi¬
makes a psychotherapeutic relationship diffi¬ dent, and occasionally grandiose.
cult. The presence of schizoid features should Table D-55 provides the prototypic
be considered. scores for 9-0/0-9 codetypes.
Table D-53 provides the prototypic
scores for 8-0/0-8 codetypes.
Spike 0
Clients who elevate only Scale 0 (Social Intro¬
Spike 9
version) are relatively unusual. They are ex¬
Clients who elevate only Scale 9 (Hypoma- periencing only minor psychological and
nia) are impulsive, acting-out individuals emotional distress that tends to be chronic in
with a history of criminal and interpersonal nature. They are easily frightened and fre-
 Codetypes 285

quently phobic. They are in good physical comfort represents a schizoid adjustment, a
health. They do not get into trouble because neurotic reaction, or simply a lifestyle prefer¬
of their behavior. ence will have to be determined by under¬
Clients are very shy, bashful, introverted standing the client’s history and his or her
individuals who are easily embarrassed in so¬ reason for taking the MMPI-2.
cial situations. They lack self-confidence and Further interpretive information is avail¬
are easily overwhelmed by others. They are able in King and Kelley (1977a) on Spike 0
uncomfortable in any type of interpersonal codetypes. Table D-56 provides the proto¬
relationship. Whether this interpersonal dis¬ type scores.
 CHAPTER 7

Interpreting
the MMPI-2 Profile

The preceding chapters have reviewed the MMPI-2 interpretation. First, the interpre¬
common interpretive statements made about tive process in this chapter will be “blind,”
specific elevations on the individual validity that is, done without any additional sources
and clinical scales and have explored the var¬ of data about the client. This is not to suggest
ious interscale relationships. Both of these that the clinician in actual practice should
sources of information provide the database follow such a procedure, but it is necessary
used in profile interpretation of the MMPI-2. here in order to limit the interpretation to a
Consequently, the reader should be thoroughly finite, known database. Such a procedure en¬
familiar with the preceding chapters and refer sures that readers will know what data are
back to specific sections when appropriate and used in making the interpretation and that
necessary in interpreting individual profiles. they will become aware of what information
The task of this chapter will be to ana¬ can and cannot be obtained from the MMPI-
lyze the interpretive process for MMPI-2 pro¬ 2. There are obviously questions that the cli¬
files. In addition to a discussion of issues in nician cannot answer from the MMPI-2
profile interpretation, two profiles are pre¬ alone, and this blind analysis should sensitize
sented to serve as examples of the interpretive the clinician to the additional data that are
process. Each profile will be interpreted by needed to answer these particular questions.
an individual clinician. This interpretation Second, MMPI-2 data, like most clinical
will then be compared and contrasted with in¬ data, are amenable to more than one inter¬
terpretations produced by three computer- pretation. Thus, the clinician’s task, and the
based interpretive systems. task in the examples provided, is not to find
one and only one “correct” interpretation. It
is, rather, to find an interpretation of the pro¬
PRELIMINARY ISSUES IN
file that is internally and theoretically consis¬
PROFILE INTERPRETATION
tent as well as empirically testable.
Several issues should be discussed prior to un¬ The focus of this chapter, then, is to ex¬
dertaking an analysis of the process of plore how interpretations are made rather

287
288 Chapter 7

than to decide which alternative interpreta¬ scribed in Chapter 3. Once it has been ascer¬
tion is “better.” The reader should keep this tained that the client has endorsed the items
in mind when it appears that only one “cor¬ consistently and accurately, the clinician
rect” interpretation is being made of each should examine the client’s scores on each of
profile in this chapter. It is possible that each the individual validity and clinical scales. For
profile may have an alternative interpreta¬ each scale, the clinician needs to decide
tion, and once the reader understands the whether this score is in the normal or deviant
procedure explained in this chapter, he or she range for this particular client. (This proce¬
may discover a “better” interpretation. This dure will be illustrated for the two sample
process of becoming an informed skeptic profile interpretations later in this chapter.)
when interpreting MMPI-2 profiles is At this point the clinician should note
strongly encouraged. what a score in this range for each scale
Finally, even though the description of means by referring back to the earlier sections
the process of analysis is being limited to the of this text. It is important to remember that
MMPI-2, the same general skills and proce¬ a score in the normal range on any scale may
dures are involved in the interpretation of any be as deviant as a score in the deviant range.
clinical material, including the Rorschach, an For example, if the client has recently com¬
interview, or dreams. Hopefully, the empiri¬ mitted some heinous or bizarre crime and yet
cal nature of the MMPI-2 will make under¬ the MMPI-2 does not indicate the presence of
standing of this basic process of interpreta¬ any guilt, remorse, or depression, these “nor¬
tion easier. (The interested reader should mal” scores should provide valuable infor¬
consult Levy’s [1963] text, Psychological In¬ mation about the client’s personality and po¬
terpretation, for a more abstract discussion tential for significant behavior change.
of this issue.) Each scale should be analyzed sequen¬
tially with the notes made on what each score
means and any hypotheses recorded that
THE INTERPRETIVE PROCESS occur to the clinician while proceeding
Even in a blind analysis, basic demographic through this process. Once familiar with all
data on the client whose profile is being inter¬ of the individual validity and clinical scales,
preted are needed in order to help the clini¬ the clinician should make predictions about
cian determine which reference group to use. ensuing scales on the basis of the score on a
At a minimum, the clinician needs to know particular scale.
the client’s age, gender, education level, so¬ A frequently occurring example is signif¬
cial class, and ethnic group; any of these vari¬ icant elevations on scales/indexes measuring
ables can drastically alter the interpretation underreporting of psychopathology (see
of particular scales or entire profiles. For ex¬ Chapter 3). When this happens, the clinician
ample, a T score of 60 on Scale 9 (Hypoma- should suspect that all of the clinical scales
nia) may be typical for an adolescent and will be somewhat less elevated, since the de¬
extremely unusual for a 60-year-old. Thus, fensiveness tapped by these scales should
only after one knows the basic demographic serve to conceal the pathology measured by
data is one ready to start interpreting the pro¬ the clinical scales. In such a case the clinician
file. should be more aware of any scales that are
The first step in profile interpretation is elevated significantly since they are atypical.
to insure that the client has endorsed the At this stage of the interpretive process
items in a consistent and accurate manner the clinician should be willing to entertain
using the scales and indexes that were de¬ virtually any hypothesis that fits the data. No
 Interpreting the MMPI-2 Profile 289

hypothesis should be summarily dismissed not ignore the data simply because of the base
unless there are absolutely no data to support rate.
it. This does not mean that great inferential Thus, the clinician is advised to use Bar-
leaps can be made from the data, but it is wise num-type statements that have virtually 0
to acknowledge even tenuous hypotheses at percent or 100 percent applicability only
this stage. when they provide the most accurate ap¬
The neophyte clinician tends to alternate praisal of the client, and they should be indi¬
between the two extremes: refusing to make vidualized insofar as possible.
any inferences whatsoever from the data or The next step in the process of interpre¬
making such great inferential leaps that virtu¬ ting an MMPI-2 profile is to examine the data
ally no one would give credence to his or her gathered from the individual validity and
hypothesis. Hopefully, by being aware of clinical scales for consistent and inconsistent
these extremes, the clinician can learn to fol¬ information. Inferences or hypotheses that
low a more moderate course and cautiously are suggested by several scales are important
assert hypotheses about the meaning of to note since they may represent central fea¬
scores on particular scales. tures of the client. It is a rare profile, how¬
A cautionary note concerning base rates ever, that does not also contain inconsistent
(the frequency with which a specific behavior information; the contradictory inferences
occurs) of reference groups also is in order. If that result are important since they will some¬
75 percent of the clients in a specific setting how need to be resolved in interpreting the
are psychotic, the clinician should realize that profile. Occurrences of such inconsistent in¬
any hypothesis stating that the client is not ferences are commonplace, and both sample
psychotic is going against the base rates. profiles interpreted in this chapter contain at
One also must be wary of making “Bar- least one example. Any information from the
num”-type statements (Meehl, 1956), which validity or clinical scales that is inconsistent
are accurate for virtually everyone in the ref¬ with the client’s demographic data also
erence group and hence are meaningless in should be noted.
describing a specific client. Thus, to hypothe¬ The next level of analysis in profile inter¬
size that an acutely psychotic patient is hav¬ pretation involves subgroups of scales or
ing heterosexual difficulties is neither partic¬ interscale relationships. In many respects this
ularly brilliant nor insightful. On the other is the most common type of interpretation,
hand, the accuracy of such statements should since at its least complex level it involves the
not be ignored, and they should be tailored to relationship of only two scales. If the two
make them more specific for the individual. scales examined are the highest or most devi¬
All too often, once clinicians are famil¬ ant clinical scales, a codetype (high-point
iar with base rates and Barnum-type state¬ pair) analysis, the core of any MMPI-2 inter¬
ments, they avoid any statement that could pretive system, is being utilized.
be even vaguely labeled as such. Unfortu¬ In fact, many interpretive systems look
nately, such a reaction is much like throwing no farther than the codetype for any profile.
the baby out with the bath water. The clini¬ For this analysis it is only necessary that two
cian is afraid to make any base rate statement clinical scales or a single clinical scale (a spike
even when it may provide the most accurate profile) be elevated at or above a T score of
appraisal of the situation. For example, if an 65 on the MMPI-2 or a T score of 70 on the
elevated score on Scale 3 (Hysteria) virtually MMPI. Once the codetype for the profile has
excludes the possibility of the person’s being been determined, interpretation begins by se¬
diagnosed as psychotic, the clinician should lecting the appropriate statements from the
290 Chapter 7

description of this particular codetype for the involves the criterion group on which the in¬
reference group from which this person terpretive system is based. Possible criterion
comes (see Chapter 6). groups on which interpretive systems have
In selecting the appropriate statements been developed include male veterans (Gil-
from the description of the codetype, there berstadt, 1970; Gilberstadt & Duker, 1965), ad¬
are several issues to consider. The first is olescents (Marks, Seeman, & Haller, 1974),
whether the order of the two scales in the psychiatric inpatients (Gynther, Altman, &
codetype makes a difference in interpretation Sletten, 1973; Lachar, 1974; Marks et al.,
and whether the reference source being used 1974), a wide variety of diagnostic and patho¬
discriminates order within this codetype. logical groups (Lanyon, 1968), and college
(Codetypes in which scale order is important students (Drake & Oetting, 1959; Kelley &
have been noted explicitly in Chapter 6.) King, 1978, 1979a, 1979b, 1979c, 1980; King
Although order of scales within code¬ & Kelley, 1977a, 1977b).
types seems basic to any interpretive system, Several references in addition to this text
a little simple arithmetic will show how diffi¬ also provide general interpretive data on
cult it is to implement such a system. A code¬ codetypes: Archer (1987), Dahlstrom, Welsh,
type system based on all 10 clinical scales in and Dahlstrom (1972), Duckworth and An¬
any combination involves 45 ([10 x 9]/2 = derson (1986), Friedman, Webb, and Lewak
45) possible codetypes. Controlling for gen¬ (1989), Good and Brantner (1961, 1974),
der (male, female), age (adolescent, adult, Graham (1987, 1990b), and Lachar (1974).
aged), and social class (upper, middle, When the client whose profile is being in¬
lower), with a minimum of 10 persons per terpreted matches one of the above criterion
group, requires a sample of more than 8,000 groups, the clinician should consult the ap¬
(2 x 3 X 3 x 10 x 45 = 8,100) persons. propriate source. An even greater improve¬
Controlling for the order of the scales would ment, however, is when such information is
double the number of persons required. Ob¬ available from the clinician’s own institution
viously, if finer discriminations in any of or setting. It is unlikely that this goal will ever
these groups are of interest, or if other vari¬ be reached for many discrete settings, but a
ables such as ethnic group or reason for refer¬ serious student of the MMPI-2 should collect
ral are of interest, the number of persons re¬ as much information as possible on local cli¬
quired for an interpretive system quickly entele. Since most clinicians will never have
becomes astronomical. access to a sufficient number of clients to
Consequently, the clinician should be evolve a local interpretive system, the best
aware of how each interpretive system was that can be hoped for is the appropriate mod¬
developed and must work within the limita¬ ifications of the most applicable system.
tions of the data each provides. A prime ex¬ A third issue in interpreting codetypes in¬
ample of the importance of scale order is 1-3 volves elevation. Although there are some
and 3-1 codetypes. It is important to know notable exceptions (e.g., Gilberstadt &
which scale is higher in this codetype for ac¬ Duker, 1965), most interpretive systems do
curate interpretation, since they represent not discriminate among scale elevations
two very different groups of individuals; in¬ above a T score of 70 on the MMPI, the usual
terpretive systems that do not discriminate cutoff for inclusion as a codetype. This is not
between these two codetypes will usually con¬ to say that the developers of such systems be¬
tain a single description filled with seemingly lieve that distinguishing among scale eleva¬
contradictory data. tions above a T score of 70 on the MMPI is
A second issue in interpreting codetypes unimportant. Rather, the problem is again
 Interpreting the MMPI-2 Profile 291

one of how many variables it is feasible to in¬ tent features represent areas that need to be
corporate into a system and still have it re¬ examined more carefully.
main of manageable proportions. It is extremely important, at any level of
The clinician, then, is faced with the profile interpretation, for the clinician to be
need to take into consideration the degree of wary of focusing exclusively on any one fea¬
elevation of the codetype and has little empir¬ ture and consequently ignoring, biasing, or
ical direction on how to do so. For some misinterpreting the rest of the data. The clini¬
codetypes (e.g., 4-9/9-4) a difference of 10 to cian has always to be willing to entertain al¬
20 T score points seemingly has little clinical ternate hypotheses or inferences from the test
import, whereas for other pairs (e.g., 7-8/8-7) data. All too often a novice clinician selects
such a difference cannot be ignored. To a cer¬ one characteristic or feature as being impor¬
tain extent, the higher the elevation of the tant and then cannot get away from this per¬
codetype, the more distress the client is expe¬ ceptual set in interpreting the test data. In¬
riencing or reporting and the more ego alien stead, the clinician should acknowledge
or ego dystonic is the psychopathology. For inconsistencies with the inference being made
personality disorders, the higher elevations and consider other inferences that might be
indicate that the pathology is more ingrained better able to fit the data. But even after the
and resistant to change. clinician feels that the best inferences have
Beyond these simple statements it seems been selected from those available and all al¬
that the importance of the degree of elevation ternative inferences have been carefully re¬
of a codetype is a question that still needs to jected, he or she should come back to the
be researched. The degree of elevation of the data at a later time with an open mind.
entire profile has been researched more ex¬ Special mention should be made of the
tensively, probably because it is easier to in¬ necessity to note the low-point clinical scales,
vestigate. (This issue will be taken up when which are essentially the converse of the high-
interpretation of the entire profile is dis¬ point scales. Frequently, clinicians who are
cussed.) Finally, it is being assumed that a T learning to interpret the MMPI-2 focus solely
score of 65 on the MMPI-2 is equivalent to a on the analysis of the codetype or some other
T score of 70 on the MMPI. This assumption feature of the profile and miss other signifi¬
is probably fairly safe since elevation does cant data, usually the low points in the pro¬
not appear to have any systematic effects file. This is unfortunate since there are sev¬
on profile interpretation of the MMPI, but eral scales for which low points are
at a minimum clinicians need to be aware particularly significant (see Chapter 4). Some
that this assumption is being made with the interpreters of the MMPI even argue that fo¬
MMPI-2. cusing on low points is a means of determin¬
Once the appropriate statements have ing the kinds of behavior the client is defend¬
been selected from the description of the ing against and does not want to reveal (cf.
codetype, the clinician can again start gener¬ Duckworth & Anderson, 1986). There is no
ating hypotheses about the type of client who known research to support the tenability of
would produce such a description. The state¬ this inference.
ments drawn from the description of the On several clinical scales, moreover, low
codetype should be compared with each of points do not represent the opposite end of
the individual scale interpretations for consis¬ the dimension measured by high points. At
tencies and inconsistencies. Redundant inter¬ any rate, the clinician should note low points
pretations serve to highlight the important in the profile, keeping in mind that low-point
characteristics of the client, whereas inconsis¬ interpretation is generally based on absolute
292 Chapter 7

rather than relative elevation of the low-point and 3 (Hysteria) (see pages 116-121 and 148—
scale. For example, a low point on Scale 3 151.
(Hysteria) at a T score of 47 does not indicate For this type of analysis the complexity
the same potential problem areas for the cli¬ increases appreciably because the degree of
ent as a low point at a T score of 31 on this relative elevation of each scale within the
scale. The latter score is much more unusual triad must be considered as well as the overall
and hence clinically more important. The elevation of the entire triad. The initial step in
reader also should recall that Tables 6-5 this analysis is a close examination of the be¬
(MMPI) and 6-10 (MMPI-2) provided the havioral and clinical correlates of the specific
frequency with which low points are found in elevation for each scale within the configura¬
commonly occurring codetypes. tion. This information must then be com¬
There is a tendency for clinicians using bined and integrated with that for the config¬
the MMPI-2 to discuss high-point triads in uration as a whole. In carrying out this
addition to codetypes (cf. Duckworth & An¬ second step, it is necessary to match the con¬
derson, 1986; Friedman et al., 1989). This ap¬ figuration as closely as possible to the various
proach recognizes the amount of information examples provided. Obviously, the closer the
not being used by a codetype system and the configuration being analyzed fits one of the
increased specificity offered by discriminat¬ examples, the better or more accurately the
ing among subgroups within any particular empirical and clinical correlates should de¬
codetype. The increase in specificity, how¬ scribe the client in question.
ever, is offset by the difficulty of obtaining a Again, the inferences garnered from the
large enough sample of profiles to make such configurations should be examined for con¬
a system feasible. This approach to the analy¬ sistencies and inconsistencies with all the pre¬
sis of high-point triads is not an analysis of vious information. At this point redundant
scale configuration; rather, it simply follows inferences, all representing cardinal features
the rationale for a high-point pair system of the client, should be appearing. Also, any
with an additional scale. glaring discrepancies with these inferences
Therefore, all of the above discussion of should be carefully noted, since they will
the procedures and problems of high-point need to be dealt with when all the material is
pair analysis is equally applicable to high- finally integrated into one report.
point triads. Some investigators (cf. Gynther The final level of analysis of the MMPI-2
et al., 1973) also have been unable to find involves configurational analysis of the entire
replicable correlates of high-point triads that profile. This level of analysis is typically lim¬
differ from those of the respective codetype. ited to the clinical scales, since the validity
For example, Gynther and associates found scales lend themselves more readily to a triad
that the correlates of a 2-1-3 high-point configurational analysis as described above.
triad did not differ from those for a 2-1 It is clearly the most complex level and fre¬
codetype. quently it is given only passing consideration
The next level of analysis of the MMPI-2 in interpreting the MMPI-2. The primary rea¬
profile involves configuration analysis of son for its neglect is the difficulty involved in
subsets of the validity and clinical scales. attempting to match an entire profile for con¬
At this level of analysis, groups of scales, figuration. Even though there is not an entire
typically triads, are examined at the same interpretive system at this level of analysis,
time. The best example of such configuration there are characteristic profiles for specific
analysis involves Scales L, F, and K, or diagnostic groups and subgroups.
Scales 1 (Hypochondriasis), 2 (Depression), Tanyon’s (1968) text on group profiles
 Interpreting the MMPI-2 Profile 293

provides the most readily available data, but chotic tetrad also decreases in elevation and
he does not furnish any information other none of the clinical scales is elevated.
than simply a mean profile for various diag¬ Thus, without additional information it
nostic groups. That is, he does not give any is difficult to discriminate the profile of a cli¬
empirical or clinical correlates for each pro¬ ent with a chronic psychotic condition from
file. If the clinician can determine how the profile of a normal person. Confirmation
closely the profile being interpreted matches of this point can be easily obtained by using
one of Lanyon’s diagnostic groups, this the Meehl-Dahlstrom (1960) rules for profile
source may be of some help in interpretation. classification, which were intended for use
Even so, the primary features in interpreting with hospitalized patients, on a normal
the configuration of the profile are its eleva¬ person’s MMPL These rules will classify a
tion, slope, and phasicity. Each of these will normal profile as psychotic most of the time.
be examined in turn. (This fact should also be a reminder to use the
The overall elevation of the clinical appropriate criterion group for interpreting
scales gives a fairly accurate representation any MMPI-2 or MMPI profile.) The inter¬
of how much distress the client is experienc¬ ested reader also could compare the mean
ing and how ego alien or ego dystonic is the MMPI profiles for chronic schizophrenics
symptomatology that the client is acknowl¬ and normals in Lanyon’s (1968) text; there
edging. The higher the elevation of the clini¬ are virtually no differences between the two
cal scales, the more the client is saying that he profiles.
or she is hurting. Sometimes, in their “cry for In neurotic disorders there is an eleva¬
help,” clients will even overemphasize their tion in the neurotic triad corresponding to the
symptomatology in order to insure that they onset of the symptomatology. As the neuro¬
will receive assistance sooner. (A careful ex¬ sis becomes more disturbing and upsetting to
amination of the scales to assess overreport¬ the client, there will be a concomitant in¬
ing of psychopathology that were discussed crease in the elevation of the psychotic tetrad,
in Chapter 3 will be helpful in these circum¬ although these scales rarely get much over a T
stances.) In general, however, the higher the score of 75. As the neurotic symptomatology
overall elevation of the clinical scales, the more subsides, there will be a gradual reversal of
distress the client is experiencing and the more this sequence until a within-normal-limits
likely the client is to resemble those individuals (WNL) profile is achieved.
on whom the scales were constructed. The interpretation of the slope of the
If sequential MMPI-2s are examined for MMPI-2 profile is based largely upon the re¬
the same client and that client is either neu¬ lationship between the neurotic triad and the
rotic or psychotic, there is a characteristic psychotic tetrad. Positive slope reflects the
pattern in the overall elevation of the clinical fact that the psychotic tetrad is elevated
scales. In psychotic disorders there is a grad¬ higher than the neurotic triad, and negative
ual or sudden increase in the elevation of the slope indicates the reverse relationship be¬
entire profile that corresponds to the onset of tween these two groups of scales. Positive
the symptomatology. If the condition be¬ slope is generally related to psychological dis¬
comes chronic, there will be a gradual de¬ orders in which the client is experiencing lim¬
crease, first in the neurotic triad (Scales 7, 2, ited impulse control, poor contact with real¬
and 3), while the psychotic tetrad (Scales 6, 7, ity, or even disorientation and confusion. In
8, and 9) remains elevated. Finally, as the short, positive slope is generally related to
psychotic symptomatology becomes inte¬ psychotic disorders, particularly when eleva¬
grated into the client’s personality, the psy¬ tion is considered in conjunction with slope.
294 Chapter 7

Negative slope is more characteristic of Before turning to the two examples of

acute psychological upsets involving anxiety, MMPI-2 profile interpretation, a comment
depression, poor morale, and physical symp¬ about automated interpretation of the
toms without psychotic distortions. The MMPI-2 is necessary. The intent in the up¬
height of the general elevation in the profile coming examples is to compare and contrast
with a negative or positive slope corresponds the process of MMPI-2 profile interpretation
to the magnitude of the discomfort and dis¬ by an individual clinician with those provided
tress the client is experiencing. At lower ele¬ by automated interpretive systems. Thus,
vations, profiles with zero slope (flat profiles) three automated systems were selected to il¬
typically are found in normal individuals, cli¬ lustrate that method of profile interpretation,
ents with a chronic psychotic condition, or but this is not to suggest that these three sys¬
clients with severe, ingrained behavior or per¬ tems are better or worse than other auto¬
sonality disorders. At higher elevations, pro¬ mated systems.
files with zero slope are characteristic of cli¬ For the interested reader, a general de¬
ents with psychological reactions that have scription and review of all automated inter¬
recently come to be called borderline states. pretive systems for the MMPI can be found
In these clients many psychotic symptoms in The Eighth Mental Measurements Year¬
can be detected, although actual psychotic book (Buros, 1978, pp. 938-962). Dahlstrom
disorders are not readily discerned. and colleagues (1972) and Graham (1977)
Phasicity, which is a measure of the have examples of automated interpretation
number of peaks in the profile, is the least com¬ of a single MMPI profile by most of these
monly used index in profile interpretation. systems.
Primarily, this disinterest reflects the fact that
this index is so confounded with elevation and
slope of codetypes that there is little to be EXAMPLES OF PROFILE
gained by considering it. Since most interpre¬ INTERPRETATION
tive systems are based on codetypes, it seems The procedure for interpreting the MMPI-2
that most profiles are best classified as being profile, which was described above, will be il¬
biphasic. There are only two additional com¬ lustrated for two clients. Each of these pro¬
ments that need to be made about phasicity. files was interpreted by an individual
First, the flatter the profile, the more clinician and also by three computerized in¬
likely the person is to be well adjusted, partic¬ terpretive services. For each client back¬
ularly at lower elevations, or the person has ground information and the MMPI-2 profiles
the characteristics discussed above for pro¬ are presented below. Then the clinician’s in¬
files with zero slope. Second, a “saw¬ terpretation of response consistency and ac¬
toothed” profile in which Scales 2 (Depres¬ curacy and of the validity and clinical scales,
sion), 4 (Psychopathic Deviate), 6 subscales, and configurations is given. This
(Paranoia), and 8 (Schizophrenia) are signifi¬ sequence reflects the steps the clinician fol¬
cantly elevated above the other clinical scales lowed in gathering information to interpret
is a particularly malignant profile. Clients the profile.
with this saw-toothed profile are likely to be This information is followed by the
experiencing a very serious psychotic disor¬ clinician’s integration of all these sources of
der. The clinician should note that the flat information into a profile interpretation.
and the saw-toothed profiles represent the ex¬ Next, the three computer interpretations are
tremes of phasicity; most profiles, which are provided. Finally, some general comments
biphasic, fall between these two extremes. are made about the four interpretations of
 Interpreting the MMPI-2 Profile 295

each profile. In addition to the MMPI-2 pro¬ God’s will for him and his family to be re¬
file, the clinician was provided only the basic united, and he insists that they were never ac¬
demographic data on each client, so the clini¬ tually separated. He believes that his family
cian and the computer interpretive systems wants him to return home but he cannot be¬
had similar information on which to make cause of constant interference from the police
the profile interpretation. and judge.
The two examples for profile interpreta¬ There is no history of alcohol or drug
tion were selected so that one example use.
matched the prototype for the codetype as The client believes that there is no reason
closely as possible; the second example had a for him being in the state hospital and he has
number of the clinical and supplementary filed suit to be released immediately. The
scales different from the prototype for the MMPI-2 (see Profiles 7-1, 7-2, and 7-3) was
codetype. Readers can see these differences administered approximately one week after
for themselves by comparing Profile 7-1 for he entered the hospital. The prototypic scores
the first example with Profile 7-4 for the sec¬ for 6-9Z9-6 codetypes can be found in Table
ond, which also have the prototypic scores 7-1.
plotted for the standard validity and clinical
scales. It would be expected that the com¬
puter interpretations would be much more Clinician's Interpretation
accurate when the profile matches the code¬
type then when it does not. The clinician may item Omissions
or may not be able to realize that the second
(See Chapter 3) He endorsed all of the items
profile varies in significant ways from the
? = 3 except three. None of these
prototype and incorporate that information
three items is omitted by
into the interpretation.
most clients. Item omis¬
sions are not a problem for
profile interpretation. The
Interpreting the MMPI-2: Example 1
content of these items is 88:
The client is a 41-year-old, separated, white “I believe women ought to
male who was admitted for the second time to have as much sexual free¬
the state hospital. His previous hospitaliza¬ dom as men”; 258: “I can
tion occurred two years ago following a sepa¬ sleep during the day but
ration from his wife, and resulted from his not at night”; and 259: “I
increasing agitation over trying to maintain am sure 1 am being talked
his relationship with his wife. He had been about.”1
telephoning her as many as 75 times a day,
driving by her residence repeatedly, and
Consistency of item Endorsement
physically abused her once. At that point the
wife obtained a restraining order because the (See Chapter 3) The score on VRINand the
patient was “obsessed with keeping the fam¬ VR1N = 9 absolute difference be-
ily together.” F =12 tween Fand FB are at the
He believes that he is in the state hospital Fr =5 lower end of the marginal
because the judge and police are fabricating range. By the next two
evidence against him. His wife reported con¬ rules for assessing consis¬
tinued harassment and he recently chased his tency of item endorsement
son home from school. He states that it is (see Table 3-45, p. 100),
 296 Chapter 7

PROFILE 7-1. MMPI-2 Standard Validity and Clinical Scales for Example 1
Name Example 1
s K I l.iiluw.i\ .iiuI I < \K kmika

MMPI-2 Huuh'si’Ut
/mt’nfvru -J"
Address

Occupation
2709 Elmwood
Auto Mechanic Date Tested 12/12/89
Profile for Basic Scales
Minnesota Muliipluisic Personal’! i\ Inventi>r\-2
Education 12th Age 41 Marital Status Divorced
( op> right i h\ INI RKil NTS OE lilt UNIVERSITY OK MINNESOTA
PMl 144.1 i renewed 1970). I9N9. This Profile Form I9N9.
Ml ri-elus reverxed Distrihuied evelusixelv h\ NATIONAL COMPUTER SYSTEMS, INC.
Referred by Dr. Harry Davis
under license from The l niversiiv of Minnesota.

"MM PI-'" and "Minnesota Muliipluisic Personalih lmenior\-2" are trademarks owned b\
MMPI-2 Code 6’948-0/25317 F’L-/K
Lhe l imersit\ of Minnesota Printed in the United States of America.

TorTcL F K
Hs* 5K
1
0
2
Hy
3
Pd+ 4K
4
PI-IK
7
Sc* IK
8
Ma*2K
9
Si
0 T or Tc Scorer's Initials HG

K -

6
• IS 12 6
- 14 !1 6
14 11 5
?6 13 10 5 Client • 9

.A 13 10 S
■ 12 10 s
23 12 9 s
22 11 9 4
21 11 8 4
Prototype
20 10 8 4
19 10 8 4
13 9 7 4
’’ 9 7 3
16 8 6 3

IS 8 6 3

13 7 5
b
11 6 4 2

10 5 4 2
9 5 4 ?
8 4 3 ?
7 4 3 1
6 3 2

6 3 2 l
4 2 2 l
3 2 l i
2 i i 0
' i 0 0
0 0 0 0

Raw Score 6 12 13 3 17 18 24 24 18 9 19 24 31
■*
NATIONAL
? Raw Score _3 K to be Added_L 5 13 13 ^
COMPUTER
SYSTEMS

Raw Score with K 10 29 22 32 27 24001

 Interpreting the MMPI -2 Profile 297

PROFILE 7-2. MMPI-2 Content Scales for Example 1

Name Example 1
s K Il.iili.iwm .11uI I t Mi KiiiKa

Hnmi\ickr Address 2709 Elmwood

f?f\h.vur/(fy /mt'nk’ru
Occupation Auto Mechanic Date Tested 12/12/89
Pi ol ik- for ('anient Scales
liniclier. (iiiiham. Williams and Bcn-Poralh ( 1989) Education 12th# Age 41 Marital Status Divorced
Miuik-ni'U Mulnph.isk- Persona Imp ln\enioi\-2
(op\neM • h\ mi RI til \ I S Of III! L NI \ I RSI [ \ Oh MINNl.SO'lA
;iu: POt t renewed P' ui. 1 oso. I Ins Pnv|jk ton* KS9. Referred by Dr. Harry Davis
\'ii nelus rescrxeil I )isi n hilled escluM\el> h\ WIIOWI COM IH MR SYSIIMS. INC
•ander ueense 111'Mi I he l ni\ersit\ o| \lmikMM.i
MMPI-2 .u'lI MihMe'M'ij \luliiphasic Personalin Imenlon-2" are trademarks owned b\ Scorer's Initials HG
I he I imeiMt;. "i Minnesota Pruned in the l imed Slates ol Muenea

Raw
Score Jk 3 5 13 3 8 8 9 11 14 3 17 13 J 10 NATIONAL
COMPUTER
SYSTEMS

24002
 298 Chapter 7

PROFILE 7-3. MMPI-3 Supplementary Scales for Example 1

Name Example 1
jjjjjPjlr Wr 9 ® JSl. ifa/tift/t,/.»/<■ 2709 Elmwood
■4L >4 ;3|i' fll H
Address
H JKtt /mettfory -J"
Occupation Auto Mechanic Date Tested 12/12/89
Profile lor Supplementary Scales
M 1 'V'.'i.i MoliipluMc lViM*nalii\ lmcmor\-2
Education 12th Age 41 Marital Status Divorced
1 , P'.- j--! In lill RU.IMSOI llll l\l\IRMM Oh MINNESOTA
• "'4! renewed l^'nt. 1‘>S4 Mm Piofile form 14N4
\
i' 1
..-vcivcd Dim nbmed c\cluM\el\ In \ \TK)\.-\1 (OMI’l HR SYSTEMS. INC Referred by Dr. Harry Davis
. uu 'uT'C irom I lie l rmersin of Minnesota.

MMI'I ' .rui Minnesota Miilnphasic Personalin Imcntor\-2~ arc trademarks owned h\ Scorer's Initials HG
l:v 1 oci'ii> oi Minnesota. Primed in ihe l niled Stales of America.
TRIM
Si, T/F

T A R Es MAC R O H Do Re Mt GM GF PK PS Si, S.3 fb vrin t/f t

TRIN

Raw
Score 12 18 39 31 8 15 14 14 40 20 9 14 5 8 4 5 9 10 NATIONAL
COMPUTER
SYSTEMS
24004
 Interpreting the MMPI-2 Profile 299

TABLE 7-1 Prototypic Scores for 6-979-6 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 131 260
Age 33.9 13.0 31.2 12.4
Men 51.9% 66.5%
Women 48.1 33.5

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 13.9% 15.0%
Total (Obvious-Subtle)3 74.6 54.0 1 14.9 62.6
Critical items6 42.5 12.3 46.3 14.3
Overreportedc 12.7% 41.8%
Underreported0 12.0% 3.5%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.0 5.5 4.4 6.0
L 49.5 7.4 50.1 9.9
F 73.2 12.3 82.6 19.4
K 43.4 7.1 37.2 7.2
1(Hs) 53.0 1 1.0 52.7 12.5
2(D) 52.0 8.7 51.7 9.6
3(Hy) 52.6 9.7 48.3 1 1.7
4(Pd) 66.7 8.5 62.8 10.2
5(Mf) 57.0 9.1 51.4 9.4
6(Pa) 80.6 8.6 81.2 1 1.3
7 (Ft) 59.6 8.0 58.2 10.6
8(Sc) 70.8 9.4 67.9 1 1.5
9(Ma) 82.5 8.6 80.5 10.3
O(Si) 51.8 8.1 49.5 8.2

Supplementary Scales
M SD M SD
A 61.2 9.1 65.9 9.9
R 48.6 1 1.4 38.9 10.2
M A C/M AC-Rd
men 30.4 4.2 32.0 4.1
women 27.6 4.0 27.9 4.2

Codetype Concordance
Men Women Men Women
MMPI-2 6-979-6 84.2% 88.2% MMPI 6-979-6 38.5% 71.9%
Spike 6 1 5.8 6-878-6 23.1% 12.3%
1-6/6-1 1 1.8

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
300 Chapter 7

the client endorsed the Validity Scale Configuration

items consistently. Profile
(See Figure 3-3, This client is admitting to
interpretation may proceed
dashed line) personal and emotional
since the client has en¬
difficulties, and simulta¬
dorsed the items consis¬
neously trying to defend
tently.
himself against these
problems in an unsophisti¬
Accuracy of Item Endorsement cated manner. This pat¬
tern of ineffective
(See Chapter 3) Both scales/indexes
defenses with the simulta¬
Total (Ob-Sub) = 34 are within the accept-
neous admission of fairly
Crit items = 28 able range. When
severe problems is typical
his scores are com¬
of chronically malad¬
pared to the proto¬
justed clients. He is not
type for a 6-9/9-6
an optimal client for any
codetype (see Table
type of psychological in¬
7-1), he is almost
tervention.
two standard devia¬
tions lower on both
scales/indexes. Al¬ Codetype
though the client has (See Chapter 6) Clients with a 6-9/9-6
endorsed the items codetype are usually en¬
accurately, he is re¬ countered only in inpa¬
porting less distress tient settings. They are
than most clients angry, hostile individuals
with this codetype. who may exhibit grandios¬
Profile interpreta¬ ity and egocentricity.
tion may proceed They are also irritable, ex¬
since the client has cited, and energetic. They
endorsed the items may report difficulty
accurately. thinking and concentrat¬
ing, and exercise poor
judgment.
Validity Scales
The presence of a psy¬
(See Chapter 3) He may have a tendency to chotic process that is
L = 61 resort to denial mecha¬ more likely to reflect a
(raw score = 6) nisms. mood disorder than a
F = 73 He is acknowledging the thought disorder should
(raw score unusual experiences repre¬ be considered. They de¬
= 12) sented in this scale more scribe themselves as
than the typical individual. happy, calm, easygoing,
K = 45 He describes himself as and in good physical
(raw score having adequate personal health. Others see them as
= 13) resources. He has a proper angry, hostile, and over-
balance between self-disclo¬ responsive to minor
sure and self-protection. stresses and problems;
 Interpreting the MMPI-2 Profile 301

less frequently they are Pd, = 65 He may struggle against fa¬

seen as tense and anxious. milial control.
Pd2 = 68 He may resent societal de¬
mands and conventions and
Prototypic Scores parental standards.
5(Mf) = 46 He has traditional masculine
(See Table 7-1) The client was more than interests and activities.
one standard deviation 6 (Pa) = 79 He is likely to be suspicious,
above the mean on Scale hostile, and overly sensitive
L. He was more than one and overtly verbalizes these
standard deviation below qualities. A thought disorder
the mean on Scales F, 7, may be readily apparent.
7, and 9. It will be impor¬ Pa, = 82 He has ideas of external in¬
tant to insure that these fluence. He externalizes
atypical scores for the pro¬ blame for his problems, frus¬
totype are accounted for trations, and failures. He
in the interpretation. may have persecutory ideas.
He projects responsibility for
Clinical Scales his negative feelings.
(See Chapter 4) Pa2 — 62 He may consider himself as
l(Hs) = 42 He may be denying the pres¬ special and different from
ence of vague physical com¬ other people.
plaints. Clearly, he is not 7(Pt) = 41 He is secure and comfortable
reporting any health-related with himself and is emotion¬
problems. ally stable. He is success ori¬
2(D) = 47 He reported a typical num¬ ented, persistent, and capable.
ber of attitudes and behav¬ There is an absence of wor¬
iors that reflect symptomatic ries and a relaxed attitude to¬
depression. Neither his pres¬ ward responsibilities. Again,
ence in the state hospital like with Scale 2, he is not re¬
(even though he believes that porting any type of anxiety or
he is being confined unjustly) distress.
nor his problems with the 8(Sc) = 60 He thinks differently than
judge and police that led him others. He tends to avoid re¬
to be sent to the hospital are ality through fantasy and
seen as being upsetting or dis¬ daydreams.
tressful. His normal score is 9(Ma) = 69 He is overactive, emotionally
not normal! labile, and may experience
3(Hy) = 43 He tends to be caustic, sar¬ flight of ideas. Although his
castic, and socially isolated. mood is typically euphoric,
He has few defenses. He has outbursts of temper may
narrow interests and is so¬ occur. He is impulsive and
cially conforming. may have an inability to
4(Pd) = 64 He has adjusted to an habit¬ delay gratification. Manic,
ual level of interpersonal and narcissistic, and grandiose
social conflict. features may be seen.
302 Chapter 7

0(Si) = 56 He reported a balance be¬ and experiences, paranoid

tween socially extroverted ideation, and hallucinations.
and introverted attitudes and ANG — 50 He does not see himself as
behaviors. being angry, moody, or irri¬
table. He did not endorse
Supplementary Scales any of the items directly re¬
lated to anger.
(See Chapter 5) CYN = 48 He is not cynical.
A = 53 He has a normal amount of ASP = 55 He did not engage in prob¬
anxiety. Again, he is not re¬ lematic behaviors while in
porting any distress. Note school.
the low scores on other first- TP A = 68 He is hard-driving, fast-mov¬
factor scales (Alt = 54; PK ing, and work-oriented, who
= 52; and PS = 54). frequently becomes impa¬
R 56 He shows an appropriate tient, grouchy, irritable, and
willingness to discuss his annoyed. He does not like to
behavior and problems. wait, be interrupted, or be¬
MAC = 31 He has a significant probabil¬ lieve that someone has gotten
(raw score) ity of abusing alcohol or the best of him.
drugs. He is likely to be impul¬ LSE = 62 He has somewhat of a low
sive, have a high energy level, opinion of himself. He lacks
have shallow interpersonal re¬ self-confidence and sees him¬
lationships, and be psychologi¬ self as not as good or capable
cally maladjusted. as others.
SOD = 71 He is very uneasy around oth¬
Content Scales ers and is happier by himself.
(See Chapter 5) He sees himself as shy. He
ANX = 45 He does not report general dislikes parties and other
symptoms of anxiety as has group events.
been noted several times FAM = 71 He reports considerable fa¬
above. milial discord. His family
FRS = 54 He has a typical number of was lacking love, support,
specific fears. and companionship. Fam¬
OBS = 50 He does not report obses¬ ily members are nervous
sional thoughts. and have quick tempers.
DEP = 66 He has depressive mood and They are to be avoided.
thoughts. He feels blue, un¬ WRK = 50 He is as able to work as he
happy, and is likely to ever was, and his current
brood. He is prone to blame problems are not interfering
others for his problems. He with his ability to work.
did not endorse any of the TRT = 64 He dislikes going to doctors
items related to suicidal ide¬ and he believes that he should
ation or attempts. not discuss his personal prob¬
HEA = 44 He does not report any con¬ lems with others. He does not
cerns about his health. believe that anyone under¬
BIZ = 70 He reported strange thoughts stands or cares about him.
 Interpreting the MMPI-2 Profile 303

Profile Interpretation2 may occur (Scale 6). He did not report suici¬
dal ideation or any history of suicide at¬
Test-Taking Behaviors tempts.

The client endorsed the items consistently

(VRIN, F, Fb ) and accurately (Total T score
difference; Total critical items), although he Interpersonal Relationships
is reporting less distress than most clients The client reported a balance between so¬
with this codetype on the MMPI-2. He is ad¬ cially extroverted and introverted attitudes
mitting to personal and emotional difficul¬ and behaviors (Scale 0). He has traditional
ties, and simultaneously trying to defend masculine interests and activities (Scale 5 ).
himself against these problems in an unso¬ He is very uneasy around others and is hap¬
phisticated manner (validity scale configu¬ pier by himself {SOD). The client sees himself
ration). This pattern of ineffective defenses as shy and he avoids others when given the
with the simultaneous admission of fairly se¬ opportunity {SOD). He has shallow interper¬
vere problems is typical of chronically malad¬ sonal relationships {MAC-R) and consider¬
justed clients (validity scale configuration). able familial discord {FAM).

Cognitive Processes
Treatment
He has difficulty thinking and concentrating,
The client is a poor candidate for most forms
and may exercise poor judgment {6-9/9-6
of psychological interventions for a variety of
codetype). He thinks differently than other
reasons. He is experiencing little internalized
people (Scale 8), and he reported strange
distress (Scales 2, 7, A, Mt, PK, PS, ANX),
thoughts and experiences {BIZ). There is a
and he is prone to blame others for his prob¬
strong likelihood of a potentially psychotic
lems (Scale 6, Pa,) and to expect them to
process that needs to be evaluated carefully
change to meet his needs. His problems are
{6-9Z9-6 codetype; BIZ). A review of his
chronic in nature (validity scale configura¬
background and reasons for coming to the
tion), and he sees little hope of changing
state hospital may be sufficient to document
{TRT). He dislikes going to doctors and he
the presence of a psychotic process.
believes that he should not discuss his per¬
sonal problems with anyone {TRT). He also
may be prone to abuse alcohol or other drugs
Mood
{MAC-R), which will only exacerbate all of
The client did not describe himself as being de¬ the problems described above.
pressed (Scale 2), anxious (Scale 7), or emotion¬ Therapy will proceed slowly at best and
ally distressed {A, Mt, PK, PS, ANX) despite should focus on his relationships with mem¬
his presence in the state hospital. Neither his bers of his family {FAM). It will be important
being hospitalized nor the behaviors that led for the therapist not to place any unusual de¬
him to being hospitalized are creating any mands on the client until some trust in the re¬
emotional distress. He does not see himself lationship is developed (Scale 6). Despite all
as being angry {ANG) or cynical (CYN). of his problems, he does not see his work as
Others are likely to see him as angry and hos¬ being affected by his problems (WRK). If his
tile {6-9Z9-6 codetype; TPA) and blaming statement about his work is accurate, it could
them for his problems {Pa,, DEP). Outbursts be used as a point of strength from which to
of temper (Scale 9; TPA) or physical acting-out work on other issues.
304 Chapter 7

Diagnostic Impression special mystical powers or a special “mis¬

sion” in life which others do not understand or
Axis I 296.44 Rule out Bipolar Disor¬
accept. He considers himself in good health and
der, Manic Type, with
does not complain of somatic difficulties.
Mood-Congruent Psy¬
chotic Features Interpersonal Relations
301.13 Rule out Cyclothymia
Axis II 301.00 Rule out Paranoid Person¬ He is overly sensitive and rigid in interper¬
ality Disorder sonal relations. His hypersensitivity, lack of
trust, and inability to compromise are likely
to disrupt or strain relationships. He broods a
Computer Interpretation —The great deal and becomes hostile when he feels
Minnesota ReportrM3 threatened. He also tends to hold grudges and
seeks to “get even” with others for perceived
Profile Validity wrongs. His lack of trust may prevent him
from developing warm, close relationships.
This is a valid MMPI-2 profile. The client’s re¬
He tends to feel insecure in personal relation¬
sponses to the MMPI-2 validity items suggest
that he cooperated with the evaluation enough ships, is hypersensitive to rejection, and may
to provide useful interpretive information. The become jealous at times. He tends to need a
resulting clinical profile is an adequate indica¬ great deal of reassurance.
tion of his present personality functioning. The content of this client’s MMPI-2 re¬
sponses suggests the following additional in¬
Symptomatic Pattern formation concerning his interpersonal rela¬
tions. He views his home situation as
Individuals with this MMPI-2 profile are ex¬ unpleasant and lacking in love and under¬
periencing severe psychological adjustment standing. He feels like leaving home to escape
problems. Extremely sensitive to criticism, a quarrelsome, critical situation, and to be
the client tends to overreact to minor prob¬ free of family domination. His social rela¬
lems with anger or hostility. He trusts no one tionships are likely to be viewed by others
and is constantly on guard to prevent others as problematic. He may visibly by uneasy
from doing him harm or injustice. When he around others, sits alone in group situations,
feels threatened, he may react with aloofness and dislikes engaging in group activities.
and self-righteous indignation, or take a rig¬
idly moralistic stance. He blames others for Behavioral Stability
his problems and rationalizes to avoid per¬
His suspicious and mistrusting behavior is
sonal responsibility. Some individuals with
long-standing. He is not likely to change sig¬
this profile type harbor delusional beliefs or
nificantly, although he may become less in¬
feel that they have a special mission in life of
tensely angry or ‘ ‘clam up” to reduce attention.
which others are unaware.
His interpersonal style is not likely to change
In addition, the following description is
significantly if retested at a later date.
suggested by the content of this client’s re¬
sponses. He views the world as a threatening
Diagnostic Considerations
place, sees himself as having been unjustly
blamed for others’ problems, and feels that he This profile strongly suggests a diagnosis of
is getting a raw deal out of life. He endorsed a Paranoid Personality or Paranoid Disorder.
number of extreme and bizarre thoughts sug¬ His unusual thinking and bizarre ideas need to
gesting the presence of delusions and/or hallu¬ be considered in any diagnostic formulation.
cinations. He apparently believes that he has His extremely high score on the Mac-
 Interpreting the MMPI-2 Profile 305

Andrew Alcoholism Scale—Revised suggests K. Interpretation for each of the individual

great proneness to the development of an ad¬ validity scales is presented below.
dictive disorder. Further evaluation of sub¬
stance use or abuse problems is strongly rec¬
Validity Scales
ommended. However, he seems to deny the
possibility of a substance use or abuse disor¬ ? (raw) = 3
der and would probably be reluctant to dis¬ Scores in this range reflect a relatively
cuss the possibility freely. small number of unanswered items, which in
and of itself should not have an impact on the
Treatment Considerations validity of the profile.

Individuals with this profile tend not to seek L T = 61

psychological help; instead, they are usually L scores in this range are suggestive of in¬
brought to treatment at the insistence of oth¬ dividuals who may be defensive, lack insight,
ers. Because they are likely to be quite and be slightly more conforming and moral¬
guarded, suspicious of the therapist, and op¬ istic than usual. They may have a tendency to
positional or detached in treatment, their ther¬ repress or deny problems and unfavorable
apy sessions will probably be stormy. Treatment traits.
is likely to be terminated prematurely.
If psychological treatment is being con¬ F T =13
sidered it may be profitable for the therapist
F scores in this range, if they are valid,
to explore the client’s treatment motivation
suggest the increasing probability of serious
early in therapy. The item content he en¬
psychological and emotional problems which
dorsed includes some feelings and attitudes
are often characteristic of severe neurosis,
that could be unproductive in psychological
psychosis, or behavioral problems. Scores in
treatment and in implementing self-change.
this range also may occur because individuals
Examination of item content reveals a con¬
have had some difficulty reading or under¬
siderable number of problems with his home
standing the test items (evaluate measures of
life. He feels extremely unhappy and alien¬
consistency of item endorsement), or because
ated from his family. He related that he feels
they have some motivation to overreport psy¬
that his home life is unpleasant and feels pes¬
chopathology (evaluate measures of accuracy
simistic that the situation will improve. Any
of item endorsement).
psychological intervention with him will need
to focus upon his negative family feelings if
K T = 45
treatment progress is to be made.
Scores in this range are typically ob¬
tained by individuals who exhibit an appro¬
Computer Interpretation —MMPI-2 priate balance between self-disclosure and
Adult Interpretive System self-protection. These individuals usually are
psychologically well adjusted and capable of
Profile matches and scores for the client are
dealing with problems in their daily lives.
provided in Table 7-2.
Scores in this range are also indicative of
good ego strength, sufficient personal re¬
sources to deal with problems, a positive self-
Configural Validity Scale Interpretation
image, adaptability, and a wide range of in¬
There is no information available for this terests. Prognosis for psychological intervention
configuration of scores for scales L, F, and is generally good.
306 Chapter 7

TABLE 7-2 MMPI-2 Adult Interpretive System: Profile Matches and Scores for Example 1

Highest
Client Scale Best Fit
Scale Profile Codetype Codetype

Codetype match: 6-9/9-6 6-9/9-6 (4)

Coefficient of Fit: .91 .94
Scores: ? (raw) 3
L 61 50 50
F 73 83 69
K 45 37 41
Hs (1) 42 53 47
D (2) 47 52 49
Hy (3) 43 48 46
Pd (4) 64 63 65
Mf (5) 46 51 48
Pa (6) 79 81 73
Pt (7) 41 58 52
Sc (8) 60 68 58
Ma (9) 69 81 75
Si (0) 56 50 46

Mean Clinical
Elevation: 55 59 58
Ave age-males: 30 30
Ave age-females: 34 36
% of male codetypes: 3.1% 1.4%
% of female codetypes: 3.2% 1.0%
% of males within codetype: 66.7% 74.5%
% of females within codetype ■
33.3% 25.5%

Configural clinical scale interpretation is provided in the report for the following codetype(s):
6-9Z9-6
6-9Z9-6 (4)

Note: Reproduced by special permission of Psychological Assessment Resources, Inc., from

MMPI-2 Adult Interpretive System by Greene, Brown, & PAR. Copyright 1 990. Further reproduc¬
tion is prohibited without permission from PAR, Inc.

Configural Clinical Scale Interpretation process that is more likely to reflect a mood
disorder than a thought disorder.
6-9/9-6 Codetype (High Match) They are distrustful of others and vulner¬
able to perceived threat. They often project
Clinical Presentation their feelings and problems onto others. They
have difficulty expressing their feelings ap¬
These individuals are very active and ener¬ propriately. They may vacillate between
getic. Sometimes they become so agitated and overcontrolling and undercontrolling their
excited that they may report difficulty in emotions. These individuals are sometimes
thinking and concentration. At these times described as being tense, anxious, and irrita¬
they may exhibit indications of a psychotic ble. They often exercise poor judgment, al-
 Interpreting the MMPI-2 Profile 307

though they think that their judgment is very emotions. They often exercise poor judg¬
good. ment, although they think that their judg¬
The self-concept of these individuals is ment is very good. They believe that they are
often grandiose and egocentric. They are cyn¬ as able to work as they ever were. They are in
ical of the abilities of others. good physical health. These individuals are
The exaggerated need for affection ex¬ sometimes described as being tense, anxious,
hibited by these individuals, coupled with and irritable.
their suspiciousness, hypersensitivity and fear The self-concept of these individuals is
of emotional involvement, often results in often grandiose and egocentric. They are cyn¬
unsatisfying and volatile interpersonal rela¬ ical of the abilities of others.
tionships. They tend to be unconcerned The exaggerated need for affection ex¬
about others’ evaluations of them. hibited by these individuals, coupled with
their suspiciousness, hypersensitivity and fear
Treatment of emotional involvement, often results in
unsatisfying and volatile interpersonal rela¬
The prognosis is generally poor with individ¬ tionships. They tend to be unconcerned
uals who obtain this codetype. about others’ evaluations of them. They
found their home life to be unpleasant and
Possible Diagnoses frequently wanted to leave home, if they did
Axis I Rule Out Mood Disorders not do so.
Manic Episode
Hypomanic Episode Treatment
Cyclothymia The prognosis is generally poor with individ¬
Axis II Rule Out Paranoid Personality uals who obtain this codetype.
Disorder
Rule Out Schizoid Personality Possible Diagnoses
Disorder
Rule Out Schizotypal Personality Axis I Rule Out Mood Disorders
Disorder Manic Episode
Hypomanic Episode
Cyclothymia
6-9Z9-6 (4) Codetype (Best Fit)
Axis II Rule Out Paranoid Personality
Disorder
Clinical Presentation
Rule Out Schizoid Personality
These individuals are very active and ener¬ Disorder
getic. Sometimes they become so agitated and Rule Out Schizotypal Personality
excited that they may report difficulty in Disorder
thinking and concentration. At these times
they may exhibit indications of a psychotic
Computer Interpretation —
process that is more likely to reflect a mood
Caldwell Report
disorder than a thought disorder.
They are distrustful of others and vulner¬
Test-Taking Attitude
able to perceived threat. They often project
their feelings and problems onto others. They He made a few atypical responses to the
have difficulty expressing their feelings ap¬ MMPI-2. Otherwise, he was straightforward
propriately. They may vacillate between and not unduly defensive. The basic validity
overcontrolling and undercontrolling their scales were within acceptable limits.
308 Chapter 7

The supplemental validity scales indicate means of expression. He is apt to misinterpret

extensive efforts to “look good” on the the intentions of others and to overreact to
MMPI-2. He showed a moderately high level anything he perceives or suspects to be a
of conscious defensiveness, responding “too threat to his security. He would be quick to
positively” to many of the MMPI-2 items. feel poorly treated and unfairly dealt with.
Despite this he showed little elevation on When he feels “wronged” he would be quite
scale K, suggesting a rather limited to low slow to forgive and forget. Such projections,
level of verbal sophistication. His below-av- evasiveness when challenged, a circumstan¬
erage score on the scale measuring his level of tial stream of thought, and breakdowns of his
currently attained, recently experienced, or reality testing would reflect pervasive para¬
wished for socioeconomic status (Ss) is con¬ noid characteristics.
sistent with his having obtained an unsophis¬ These distortions could prove to be delu¬
ticated low K despite this degree of faking- sional. Difficulties with alcohol would
good. His profile appears marginally valid readily aggravate these problems. Neverthe¬
or, at the least, several of his scales are apt to less, his ego strength tests as well above aver¬
be significantly underelevated. (As with scale age which predicts organized functioning and
L, we cannot tell from the Mp and Sd scales immediate practical self-sufficiency in many
which of his scales are most underelevated.) areas.
There were no indications on the Ds of His profile indicates a general pattern of
any attempt to malinger or exaggerate his hypomanic excitability and overactivity. His
level of disturbance. The scattered atypical plans and expectations could be seen by oth¬
and rarely given responses shown in his eleva¬ ers as unrealistically optimistic. He is likely to
tion on scale F appear, in the absence of any take on multiple activities or commitments as
Ds elevation, to reflect the valid reporting of if needing to distract himself as well as to
some unusual experiences and attitudes on prove his self-worth. He could be particularly
the MMPI-2. The elevation on F also sug¬ conflicted around the importance of taking
gests an internally driven person who may be advantage of all the opportunities that he does
described by such terms as dissatisfied, rest¬ get lest he “lose out” on an important experi¬
less, changeable, or complex. Despite the ence. He may also be seen as stubborn about
mildly elevated F score, his clinical scale doing things “my own way,” as if demanding
scores are not likely to be overelevated; the F validation that his way is “the right way.”
score does not appear to reflect any con¬ Talkative and expansive when things are
sciously self-critical distortion or biasing of going well for him, he is apt to become
his responses. abruptly emotional when under pressure. He
These scores suggest a person who is de¬ tests as irritable and demanding when
fensive in some areas but willing to report crossed, and personal setbacks could easily
somewhat atypical reactions in other areas. break down his controls over his aggressive
The extent of distress that he did report in¬ impulses. Confused and high strung, he ap¬
deed does appear genuine. pears capable of sharp temper outbursts if
not physical assaultiveness. These outbursts
could be the adult equivalent of temper tan¬
Symptoms and Personality trums in his childhood. The profile would not
Characteristics rule out a potential for dangerous violence.
His profile has typically been associated with He tests as notably impulsive and as
episodes of intense hostility and resentment lacking in tolerance for frustrations. His
for which he lacks adaptive and constructive chronic anger may be expressed through re-
 Interpreting the MMPI-2 Profile 309

peated resentments and defiance of authority other handicaps. Unreleased resentments ac¬
figures. Needs for attention and approval are cumulated around the child’s efforts to deal
likely to conflict with his fears of being hurt with feelings of not being appreciated along
and his underlying ambivalence about emo¬ with increasingly fixed projections that served
tional closeness and vulnerability. This could to explain why “they never treated me right.”
focus specifically on sexual behavior and ap¬ In many cases one of the parents had
proach-avoidance conflicts about involve¬ been away from the home or had been an oth¬
ment with women. That is, his positive needs erwise unavailable figure with whom to iden¬
for sexual gratification could particularly tify. Frequently the patient as a child had
conflict with his difficulties around emo¬ been particularly demanding of the mother’s
tional closeness. attention with a lack of substantial emotional
Wanting of female company, he is vul¬ gratifications. This led to a conflict between
nerable to problems around his sexual impul¬ wanting affection and hating domination.
siveness. The ego gratifications around his Chronic patterns of blame and targets of re¬
sexual activities could have become over¬ sentment and unforgiving anger accumulated
emphasized, and thus they could distort the in these family interactions. As adults these
giving and receiving of love. However, he feelings of being unappreciated and of seeing
would be quickly resentful of external con¬ others as unfairly favored over them repeat¬
trols and especially of what he would see as edly led to resentments and projections as
“meddling” by friends or family members. well as a defensive pride. Often these feelings
His overall balance of masculine and femi¬ had erupted in their marriages as acute
nine interests is within the normal range for mother-in-law conflicts along with overreac¬
his age and education. tions to other outside interventions in their
His underlying moral code appears in¬ marriages.
flexible both toward himself and others. In Despite these interpersonal problems,
some cases this was tied to religious beliefs, past periods of good work adjustment are
political convictions, or other principles and likely, although he could have repeatedly
moral standards with an excessive rigidity. overreacted to threats to his career and espe¬
Phases of active involvement in church cially to any clear or imminent job failures.
groups or other organizations would not be
unusual. Employment, community involve¬
ment, church work, and other responsibilities Diagnostic Impression
may provide positive compensations for the The typical diagnoses are of paranoid and
deficient emotional gratifications in his inter¬ schizo-affective schizophrenia and bipolar
personal or family relationships. manic-depressive illness, manic type. Second¬
He would have strong needs to be seen as ary personality disorder diagnoses such as
normal, or at least “as normal as everyone narcissistic personality and passive-aggres¬
else.” His defenses against being seen as “not sive personality, aggressive type, are fairly
normal” would include rationalizations and common in these cases. Secondary diagnoses
such reaction formations as, “It is not my reflecting chronic dependence on alcohol,
anger; I want to be kind and at peace.” drugs, or other chemical agents were frequent
In many cases this pattern has been asso¬ in these cases. It should be reemphasized,
ciated with a “mistreated sibling” life role. however, that his general understatement of
During the childhoods of these patients, sib¬ his problems and his mildly idealized self-pre¬
lings and other family members had been sentation make his profile more ambiguous
openly favored because of physical or various than most.
310 Chapter 7

Treatment Considerations downs or emotional upheavals would be pro-

gnostically unfavorable, even if of a different
The pattern indicates a major risk of a “para¬
nature from his current adjustment. The
noid explosion” if he felt trapped or cor¬
family histories of similar patients usually
nered. Similar patients have often benefitted
suggested genetic contributions, especially
from anti-manic agents. In some cases lith¬
where these involved histories of major mood
ium was combined with an initial use of a
swings in family members.
nonsedating phenothiazine, and the latter
The pattern suggests a great reluctance
was gradually tapered off and discontinued
to open up in psychotherapy, to expose him¬
over a period of weeks.
self psychologically, and to risk his precari¬
While the calming effects usually were
ous pride. Projecting anger onto the thera¬
desirable if not urgent, some of these patients
pist, he would be vigilantly sensitive to what
reacted against such medications in part be¬
he would perceive to be “hostile interpreta¬
cause of their secret fears of being poisoned.
tions” by the therapist, and he could even try
Others resisted such medications because of
to provoke such interpretations as a way of
their fears of becoming depressed and their
testing the therapist. His strong needs to keep
dislike of being “drugged.” He tests as se¬
control over himself could help him to regain
verely addiction prone. His makeup is often
his previous level of practical functioning.
associated with histories of alcoholism or re¬
However, these controls could also put a
lated forms of chronic chemical abuse and
“tight lid” on exploring his emotional feel¬
dependence. His responses suggest asking if
ings.
he has been in trouble with the law. If cur¬
The low level of expressed anxiety and
rently involved, the stress of this could have
depression would also tend to limit his in¬
precipitated or aggravated his symptoms or
volvement in treatment. Furthermore, his
otherwise have led him to make professional
emphasis on action could conflict with intro¬
contact.
spection and with “talking about feelings.”
His responses also suggest a careful re¬
Contacts with family members and other in¬
view of his sexual history as to any repetitive
formants could add considerable perspective
pattern involving (1) increasing sexual ten¬
to the situational variations of his behavior
sion, (2) hasty or ill-judged sexual encoun¬
and the related current stresses.
ters, and (3) subsequent struggles over his
The treatment of many similar cases has
self-justification that block out his negative
begun with reality confrontations around
self-judgments.
current frustrations, including what the pa¬
If not already expressed in the interview,
tient is doing and what he is running around
the therapist may wish to follow up the
or away from facing. He would benefit from
patient’s “true” responses to the following
supportive reality testing when life becomes
items:
threatening and acutely disturbing to him.
Accepting and then managing his many re¬
“There are persons who are trying to
sentments of family members and of his ex-
steal my thoughts and ideas.” wife could be central, even if he were slow
“Someone has control over my mind.” and roundabout in letting go of his denial.
Interviews are likely to focus on helping
Longstanding paranoid features and any him to recognize and verbalize the immediate
fixed and defended projections of his anger hurt-anger sequences in these interactions.
would predict a relatively gradual recovery This could involve a careful balance between
from the current episode. Previous break¬ enabling him to express his intensely angry
 Interpreting the MMPI-2 Profile 311

feelings while also cultivating social ease and use, which would indicate that the MAC-R
more mutually satisfying interpersonal roles. was a false positive in this client.
He may be slow to accept his own anger as a The client was released from the hospital
normal and sometimes desirable reaction lest after a stay of approximately two weeks be¬
he be “consumed” by rage. cause of the threatened litigation. He was re¬
The channeling and directing of his in¬ ferred for outpatient treatment at his local
tense energies and needs for excitement and mental health clinic. Once he was released
stimulation can be beneficial; vigorous physi¬ from the hospital, he filed suit against all of
cal activities were reported as a positive com¬ the people and agencies that he believed were
pensation for many patients with similar pro¬ involved in his hospitalization. He did not
files. Ways in which he felt superior to his keep any of his appointments at the mental
peers as child or now feels emotionally health clinic.
unique and different could lead to an explo¬ The reader has probably already noticed
ration of his interpersonal ambivalences. the different types of interpretations pro¬
A related assertion of personal beliefs, vided by the three computer-based interpre¬
of elaborated explanations of ongoing tive systems. Both The Minnesota Report™
events, and of unchallengeable values may and Caldwell Report are like the evaluations
overprotect against painful criticism and that the individual clinician would write,
threats to his self-esteem. A very gradual ac¬ whereas the MMPI-2 Adult Interpretive Sys¬
ceptance of the intensity and directions of tem provides a briefer narrative that the clini¬
his own angry feelings could help him to ex¬ cian would have to integrate with other mate¬
press them more appropriately and lead to rial.
an eventual relaxing and freeing up of his All three computer-based interpretive
controls. systems appear to use a configural approach
to profile interpretation, and the MMPI-2
Adult Interpretive System provides a quanti¬
General Comments tative estimate of how well the client’s profile
All four interpretations for this first example matches prototypic codetypes. The Caldwell
are in general agreement about the profile’s Report provides the most detailed and dy¬
primary features: the client blames others for namic interpretation of the client’s profile
his problems, which are chronic in nature; he and it suggests a number of issues that might
is angry, hostile, and may become physically be explored in treatment. The interested
assaultive; he has strange and unusual experi¬ reader also can take each statement in these
ences that may reflect a psychotic process; he interpretations and try to deduce the scales
has significant familial problems; he is not and/or indexes that were used as the basis for
motivated for treatment and will be difficult making the statement.
to treat; he has difficulties with substance
abuse; and he has a paranoid disorder or bi¬
Interpreting the MMPI-2: Example 2
polar disorder, manic type.
All four interpretations were incorrect as The client is a 32-year-old married, white fe¬
to the presence of substance abuse according male, who was admitted to the hospital as a
to all available information since the client result of an overdose of medication and her
did not use alcohol or other substances and verbalized statements of further suicidal at¬
he had no history of such use. As noted in tempts. She reported four previous suicidal
Chapter 5, MAC scores are typically elevated gestures by overdose; the most recent attempt
in 6-9/9-6 codetypes regardless of substance was due to her depression over her husband’s
312 Chapter 7

physical disabilities. She also is addicted to Accuracy of Item Endorsement

heroin and needed to be withdrawn from it
(See Chapter 3) Both scales/in-
while she was hospitalized.
Total (Ob-Sub) = 103 dexes are within
The client has been married for five
Crit Items = 43 the acceptable
years to an alcoholic, who is physically dis¬
range. When her
abled. It was the second marriage for both of
scores are com¬
them. She had a five-year relationship prior
pared to the proto¬
to her present marriage that was very cha¬
type for a 2-4/4-2
otic. That man committed suicide several
codetype (see
years after the termination of their relation¬
Table 7-3), she is
ship.
about one standard
She feels trapped in her dysfunctional
deviation above the
marital relationship, but her dependency
mean on both
needs are so great that she cannot bring her¬
scales/indexes.
self to terminate the relationship even though
She has endorsed
she has good skills for employment and some
the items accu¬
insight into the unhealthiness of their rela¬
rately, and she is
tionship. She was not seriously depressed,
experiencing
but rather chronically dysthymic and fearful
slightly more emo¬
of making changes in her life.
tional distress
The MMPI-2 (see Profiles 7-4 to 7-6)
than most clients
was administered approximately one week
with this codetype.
after she entered the hospital. The prototypic
Profile interpreta¬
scores for 2-4/4-2 codetypes can be found in
tion may proceed
Table 7-3.
since she endorsed
the items accu¬
rately.
Clinician's interpretation
Validity Scales
Item Omissions
(See Chapter 3) She may be trying to create
(See Chapter 3) She endorsed all of the L = 38 an extremely pathologic
? = 0 items. Item omissions are (raw score = 1) picture of herself.
not a problem for interpre¬ F =12 She is willing to acknowl¬
tation.
(raw score edge more than the typical
= 10) number of unusual experi¬
ences. She is experiencing a
Consistency of Item Endorsement
mild to moderate degree of
(See Chapter 3) Both indexes are well emotional distress.
VRIN = 4 within the acceptable K = 41 She has a proper balance
F =10 range, which indicates (raw score between self-disclosure and
Fb = 6 that she endorsed the = ID self-protection. She has suf¬
items consistently. Profile ficient personal resources
interpretation may pro¬ to desire and tolerate a
ceed since she endorsed psychological interven¬
the items consistently. tion.
 Interpreting the MMPI-2 Profile 313

PROFILE 7-4. MMPI-2 Standard Validity and Clinical Scales for Example 2
Name Example 2
.nuI 11 \l< Knilcv

MMPI-2 /ifatf /nti’tiA'rt/ -J"

Address

Occupation
3612 New Jersey
Medical Technician Date Tested 11/20/89
Profile for Basic Scales
Education 10th Age 32 Marital Status Married
Minnesota Multiphase Personally ln\entor\-2
(opvrmht ‘ b\ INI RKilMSOh INI l \l\ I RSI I Y Oh MINNESOTA
I SMI I'M' (renewed W?0). I9S9 Iliis Profile Komi W>.
\ll riehiN reserved Distributed e\elusi\el\ b\ N MIONAl COM PI "MR SYSTEMS. INC
Referred by Dr. Harry Davis
under lieense from I lie l niversiy of Minnesota.

"MMPI-: and "Minnesota Multiphasie Personally Inventors-2" are trademarks owned b\

MMPI-2 Code 2*40"7’8-136/95 F’-K:L
I lie l ni\ersiix ol Minnesota Printed in the I nited States of America.

TorTcL F K
Hs- bK
1 2
D Hy
3
Pa- 4K
4
Pi-IK
7
Sc-IK
8
Ma-2K
9
Si
0 T or Tc
Scorer's Initials HG

- -

29 lb 12 ■
14 ;i ■
14 11 -
13 to •

. '3 10 ■
- 12 ’0 ;
12 9 -i
:? 11 9 -
21 11 8 '

to 3 4
19 19 3 4
18 9 4

lb 3 ' 3

15 3 6 3
': ' 6 3
13 7 • 3
.
- - 1
10 5 2
9 5 4 2
- 4 3 2
4 3 '
6 3 2 '

' 3 2 ■
4 2 2 i
3 2 l i
? ' i 0
' ' 0 0
0 0 0 IJ

Raw Score J 10 11 104025323911 28 25 17 56 NATIONAL

? Raw Score 0 4
COMPUTER
k (o be Added 6 11 11 ^ SYSTEMS

Raw Score with K 10 36 39 36 19 24001

 314 Chapter 7

PROFILE 7-5. MMPI-2 Content Scales for Example 2

Name Example 2
: Wf |§j W&jMF %////<’.>,/,/ Htt/ttfr/n Address 3612 New Jersey
H /t?/\uvur/Sfy /msttSvrt/ -J"
Occupation Medical Technician Date Tested 11/20/89
Profile for Content Scales
Butcher, Graham, Williams and Ben-Porath ( 1989) Education 10th Age 32 Marital Status Married
Minnesota Mulliphasic Personality Inventorv-2
Copyright C by THE REGENTS OF THE UNIVERSITY OF MINNESOTA
1942. 1943 (renewed 1970), 1989. This Profile Form 1989.
Referred by Dr. Harry Davis
All rights reserved Distributed exclusively by NATIONAL COMPUTER SYSTEMS, INC.
under license from The University of Minnesota.
"MMPI-2" and "Minnesota Multiphasic Personality Inventory-2” are trademarks owned by
Scorer's Initials HG
The University of Minnesota. Printed in the United States of America.

Raw
Score 15 i 11 21 U J J5 14 15 _4 13 20 13 23 14 NATIONAL
COMPUTER
SYSTEMS

24002
 Interpreting the MMPI-2 Profile 315

PROFILE 7-6. MMPI-2 Supplementary Scales for Example 2

Name Example 2
;|BB| jdiF B %■//;*•.<*•/, r Hit/Wft/nt.u*•
Address 3612 New Jersey
■ /i’/’.Hvur/ity /rt*I'ttAvy J '

Occupation Medical Technician Date Tested 11/20/89

Profile for Supplementary Scales
Minnesota Mtiiuphasic Personalit\ lnveniorv-2 Education 10th Age 32 Marital Status Married
t op\neht In PHI REGENTS OK THE UNIVERSITY OK MINNESOTA
'Ml 1944 (renewed 1970). I9K9. This Profile Eorm 1989.
\i i -Jills resei’xed Distributed e\dusivel> b\ NATIONAL COMPUTER SYSTEMS. INC Referred by Dr. Harry Davis
nuler license from 1 he L niversil\ of Minnesota.

MMPI-2" and "Minnesota Multiphasic Personaliu Inventory-2" are trademarks owned h\ Scorer's Initials HG
liie l m\ersit\ of Minnesota. Printed in the United States of America.
TRIN

Raw
Score 28 21 24 23 13 _9 17 28 19 30 25 29 13 17 12 6 A 19 NATIONAL
COMPUTER
SYSTEMS
24004
316 Chapter 7

TABLE 7-3 Prototypic Scores for 2-474-2 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 606 343
Age 38.1 12.5 36.9 12.4
Men 78.1% 66.5%
Women 21.9 33.5

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 6.3% 3.2%
Total (Obvious-Subtle)3 38.6 59.9 68.3 62.3
Critical items6 32.9 12.1 34.4 1 1.9
Overreportedc 5.9% 16.3%
Underreported0 22.8% 13.7%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.2 5.5 3.4 4.7
L 50.9 8.0 52.5 10.5
F 62.2 9.0 64.9 13.9
K 51.5 8.8 45.3 9.3
KHs) 61.1 10.8 58.1 10.3
2(D) 81.8 9.1 76.1 8.4
3(Hy) 65.7 8.8 60.1 10.1
4(Pd) 83.1 8.3 77.6 7.9
5(Mf) 56.3 10.1 48.1 10.1
6(Pa) 63.5 9.6 59.3 1 1.0
71 Pt) 68.5 9.6 63.8 10.3
8(Sc) 65.8 1 1.6 60.6 1 1.7
9(Ma) 57.8 9.7 51.3 9.0
0(Si) 60.0 9.7 58.9 9.4

Supplementary Scales
M SD M SD
A 60.1 10.8 62.7 10.8
R 68.1 13.1 57.4 11.5
MAC/MAC-Rd
men 26.9 4.7 26.8 4.9
women 22.4 4.5 22.7 4.4

Codetype Concordance
Men Women Men Women
MMPI-2 2-474-2 45.9% 53.0% MMPI 2-474-2 92.1% 61.7%
2-7/7-2 15.7 Spike 4 1 5.6
Spike 4 10.4 4-878-4 13.0

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1 979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Interpreting the MMPI-2 Profile 317

Validity Scale Configuration sis than persons in similar

circumstances who
(See Figure 3-3, This client is admitting to
achieve a Spike 4 or a
solid line) personal and emotional
4-9/9-4 codetype.
problems, is requesting as¬
These psychopathic
sistance with these prob¬
clients with a 2-4Z4-2
lems, and is unsure of her
codetype can be under¬
own capabilities for deal¬
stood best by examining
ing with these problems.
the correlates of a Spike 4
codetype. They will dis¬
Codetype play excellent intellectual
(See Chapter 6) Clients may achieve insight into their behav¬
2-4/4-2 codetypes for a ior, make a positive im¬
myriad of reasons. One pression of their
critical factor, which earnestness on the clini¬
should be assessed cian, and vehemently pro¬
through an interview with test that they will change
the client or knowledge of their behavior. Despite
the client’s reason for tak¬ their “sincere” inten¬
ing the MMPI-2, is tions, recurrences of act¬
whether Scale 2 is being el¬ ing out are very likely,
evated by internal (intra¬ followed by the same pro¬
psychic) and/or external testations to do better
(situational) causes. when caught again.
Examples of the lat¬ Another subgroup of
ter are psychopathic indi¬ clients with 2-4/4-2
viduals who have been codetypes is more likely to
caught in some illicit or il¬ be chronically depressed
legal activity and who are and unhappy without evi¬
being evaluated as a con¬ dence of antisocial acting
sequence of their behav¬ out. These clients are dis¬
ior. The depression in playing hostility and re¬
these persons represents sentment, which often
the constraints being result from marital con¬
placed on their behavior, flict, familial difficulties,
and their depression, or or similar situations that
possibly boredom at make them feel trapped
being externally con¬ and hopeless.
strained, will alleviate it¬ They are immature,
self quickly once they dependent, and egocentric
manage to extricate them¬ and often vacillate be¬
selves from their present tween pitying themselves
situation. The presence of and blaming others for
even this situational de¬ their difficulties. These be¬
pression in these persons haviors are chronic in na¬
forebodes a better progno¬ ture and difficult to
318 Chapter 7

resolve through psycho- 3(Hy) = 56 The client has a typical

logical interventions. In¬ number of attitudes and
volvement of the other behaviors that relate to
members of the family or histrionic dynamics.
the spouse in the thera- 4(Pd ) = 84 She is fighting against
peutic interaction is im¬ something, which is usu¬
portant if meaningful ally some form of conflict
behavior change is to with authority figures.
occur. These conflicts may not
Clients with 2-4/4-2 necessarily be acted out
codetypes are frequently overtly; the rebelliousness
identified as alcoholics. and hostility toward au¬
These clients will evidence thority figures are readily
depressive features, famil¬ apparent even in these
ial conflict, and vocational cases. She is likely to be
problems characteristic of al¬ unreliable, egocentric,
coholics. and irresponsible. She
The prototypic scores may be unable to learn
for a 2-4/4-2 codetype can from experience or to
be seen in Table 7-3. plan ahead. She has a
good social facade and
Prototypic Scores makes a good initial im¬
(See Table 7-3) The client was more than pression, but the psycho¬
one standard deviation pathic features will
above the mean on Scales surface in longer interac¬
2 and 0. She was more than tions or under stress. Psy¬
one standard deviation chological interventions
below the mean on Scale are less effective than mat¬
L. She is significantly uration in achieving
more depressed and so¬ change.
cially introverted than Pdj = 80 She is struggling against
most clients with this familial control.
codetype. Pd2 = 62 She does not resent soci¬
etal demands and conven¬
tions and parental
Clinical Scales standards. Her problems
(See Chapter 4) She has a typical number may involve more of the
l(Hs) = 57 of physical ailments. familial aspects of Scale 4
2(D) = 92 She exhibits a general sad¬ rather than the authority
ness and depressed mood conflict seen in socio-
either about life or herself. pathic individuals.
Pessimism, depression, 5(Mf) = 43 The client is genuinely in¬
and hopelessness are pre- terested in traditional fem¬
vading her life. She is de¬ inine interests and
pressed, withdrawn, guilty, activities. She may be pas¬
and self-deprecating. sive in this role.
 Interpreting the MMPI-2 Profile 319

6{Pa) = 52 She may be very sensitive R =62 She is reluctant to discuss

and suspicious, yet able her behavior and any
to avoid endorsing the ob¬ problems she may have.
vious items on this Scale. She may be merely sup¬
She has the normal degree pressing this material or re¬
of interpersonal sensitiv¬ pressing and denying that
ity. any problems exist. She
7{Pt) = 72 She is worried, tense, and may lack insight into her
indecisive. Agitation may own behavior.
develop and overt anxiety MAC-R = 23 She has an average score
is usually apparent both (raw score) for her codetype. She re¬
to her and to others. Dis¬ ported that she used alco¬
abling guilt feelings may hol excessively and other
be present. drugs.
8(Sc) = 66 She feels alienated and re¬
mote from her environ¬
ment. Therapeutic
Content Scales
interventions should be di¬ (See Chapter 5)
rective and supportive. A NX = 69 She reported general symp¬
9 (Ma) = 49 The client has a normal toms of anxiety, nervous¬
activity level. ness, worries, and sleep and
0(Si) =81 She is introverted, shy, concentration difficulties.
and socially insecure. In She has difficulty making
addition, she withdraws decisions. She finds life a
from and avoids signifi¬ strain and works under a
cant others, which serves great deal of tension and
to exacerbate her prob¬ stress.
lems since others might be FRS = 46 She has a typical number of
able to help her. The like¬ specific fears.
lihood of acting out is de¬ OBS = 67 She has great difficulty mak¬
creased and ruminative ing decisions, ruminates ex¬
behavior is increased. cessively, worries
excessively, and has intru¬
sive thoughts. She dislikes
Supplementary Scales
change. She counts and
(See Chapter 5) She has a mild to moder- saves unimportant things.
A = 71 ate level of anxiety and DEP = 77 She has depressive mood
distress. She is malad¬ and thoughts. She feels blue
justed and emotionally and unhappy, and is likely
upset which may reflect a to brood. She is uncertain
situational crisis or a about her future and finds
more chronic problem. her life empty and meaning¬
Note that other first fac¬ less. She cries easily. She is
tor scales {Mt = 72, PK self-critical, guilty, and
= 75, PS = 69) are in the lonely. She reported suicidal
same range as A. ideation.
320 Chapter 7

HEA = 61 She has an usual number of she once was and she works
concerns about her health. under a great deal of ten¬
BIZ = 47 She does not report strange sion. She is tired, lacks en¬
thoughts or experiences. ergy, and is sick of what she
/WG = 47 She is not moody, irritable, has to do. She dislikes mak¬
or angry. She did not en¬ ing decisions and lacks self-
dorse items indicating that confidence. She gives up
she is angry or has problems easily and shrinks from fac¬
in controlling her anger. ing a crisis or problem.
CY7V - 58 She is not cynical. TRT —12 She dislikes going to doctors
,4&P = 75 She reported stealing things, and she believes that she
other problem behaviors, should not discuss her per¬
and antisocial practices dur¬ sonal problems with others.
ing her school years. She has She prefers to take drugs or
attitudes similar individuals medicine, since talking
who break the law, even if about problems does not
she is not engaging in antiso¬ help. She does not believe
cial behavior. that anyone understands or
77M = 41 She is not a hard-driving, cares about her. She gives
fast-moving, work-oriented up quickly and does not
person. care about what is happen¬
LSE = 68 She has a very low opinion ing to her, since nothing can
of herself, and is uncomfort¬ be done about her problems.
able if people say nice things
about her. She believes that Profile Interpretation
she is unattractive, awkward
and clumsy, useless, and a Test-Taking Behaviors
burden to others, who do not
The client endorsed the items consistently
like her. She sees herself as
(VRIN, E, Fb ) and accurately (Total T score
not as good or capable as
difference; Total critical items). She is experi¬
others, and she cannot do
encing a mild to moderate degree of emo¬
anything well.
tional distress (F scale). She is admitting to
SO£> = 77 She is very uneasy around
personal and emotional problems, is request¬
others and is happier by her¬
ing assistance with these problems, and is un¬
self. She sees herself as shy.
sure of her own capabilities for dealing with
She dislikes parties and other
these problems (validity scale configuration).
group events.
FAM = 68 She reports considerable fa¬
Cognitive Processes
milial discord. Her family
lacks love, support, and com¬ She has great difficulty making decisions,
panionship, and she wanted ruminates excessively, and has intrusive
to leave home. Family mem¬ thoughts (OBS). She vacillates between pity¬
bers are nervous and have ing herself and blaming others for her diffi¬
quick tempers. They are to culties (2-4/4-2 codetype). She did not report
be avoided and may be any strange thoughts or unusual experiences
hated. (BIZ). She has concentration and memory
07?/: = 78 She is not as able to work as difficulties because of her extensive depres-
 Interpreting the MMPI-2 Profile 321

sive symptomatology, and does not trust her life must be directly addressed (Scale 2,
judgment (Scale 2). DEP). Her suicidal ideation needs to be
evaluated carefully and monitored on an
Mood ongoing basis, particularly as her mood be¬
gins to improve (DEP). Depending on the
She is depressed, guilty, withdrawn, and self- severity of her depression and the sources of
deprecating (Scale 2). Pessimism, depression, distress in her current situation, antidepres¬
and hopelessness are prevading her life (Scale sant medication may be appropriate (Scale
2). She is uncertain about her future and
2).
finds her life empty and meaningless (DEP). Her family is a significant part of her cur¬
She reported suicidal ideation that should be rent distress and should be an initial focus in
evaluated carefully (DEP). She has a very low therapy (Pd,, FAM). Involvement of her
opinion of herself, and believes that she is un¬ spouse and/or other members of the family is
attractive, awkward and clumsy, useless, and important if meaningful behavior change is
a burden to others who do not like her (LSE). to occur (2-4/4-2 codetype).
She is tired, lacks energy, and is sick of what Therapeutic interventions should be di¬
she has to do (WRK). She does not believe rective and supportive (Scale 8). Group ther¬
that anyone understands or cares about her apy would be very beneficial since it would
(TRT). directly counteract her tendency to isolate
She also is worried, tense, and indecisive herself and avoid other people (Scale 0), and
(Scales 7, A, ANX). Her agitation and anxi¬ it would help her realize that other people
ety are readily apparent to her and to others have similar problems. Neither individual nor
(Scale 7).
group therapy should be confrontive until a
She does not describe herself as being good therapeutic relationship has been estab¬
angry or having problems in controlling her lished (LSE).
anger (ANG). She reported a history of antisocial be¬
haviors that needs to be reviewed to deter¬
Interpersonal Relationships mine whether such behaviors are still being
She is significantly more introverted than her expressed (HAP). These antisocial behaviors
peers (Scale 0). She is shy and socially inse¬ may reflect issues around her family or more
general antisocial behavior.
cure (SOD). She withdraws from and avoids
significant others, which serves to exacerbate
her problems since others might be able to Diagnostic Impression
help her (Scale 0). She is very uncomfortable
Axis I 300.40 Rule out Dysthymia
around others and is happier by herself
Axis II 301.82 Rule out Avoidant Person¬
(SOD).
ality Disorder
She has traditional feminine interests and
301.60 Rule out Dependent Per¬
activities (Scale 5).
sonality Disorder
She has considerable familial discord
(Pd,, FAM). Her family lacks love, support,
and companionship. Her family members are Computer Interpretation —
nervous and have quick tempers. The Minnesota Report™7

Treatment Profile Validity

Her significant depressive symptomatol¬ This is a valid MMPI-2 profile. The client has
ogy and general hopelessness about her cooperated in the evaluation, admitting to a
322 Chapter 7

number of psychological problems in a frank merous somatic concerns. She feels that life is
and open manner. Individuals with this pro¬ no longer worthwhile and that she is losing
file tend to be blunt and may openly com¬ control of her thought processes. According
plain to others about their psychological to her self-report, there is a strong possibility
problems. The client tends to be quite self- that she has seriously contemplated suicide.
critical and may appear to have low self-es¬ She has a self-acknowledged history of suici¬
teem and inadequate psychological defense dal ideation. It is important to perform a sui¬
mechanisms. She may be seeking psychologi¬ cide assessment and, if need be, take appro¬
cal help at this time since she feels that things priate precautions. She has acknowledged
are out of control and unmanageable. having suicidal thoughts recently. Although
she denies suicidal attempts in the past, given
her current mood an assessment of suicidal
Symptomatic Pattern potential appears indicated.
Individuals with this MMPI-2 profile tend to The client’s recent thinking is likely to be
show a pattern of chronic psychological mal¬ characterized by obsessiveness and indeci¬
adjustment. The client appears to be quite sion. She feels somewhat self-alienated and
anxious and depressed at this time. She may expresses some personal misgivings or a
be feeling some tension and somatic distress vague sense of remorse about past acts. She
along with her psychological problems and feels that life is unrewarding and dull, and
may want relief from situational pressures. finds it hard to settle down. She reports hold¬
Apparently quite immature and hedonis¬ ing some antisocial beliefs and attitudes, ad¬
tic, she may show a recent history of impul¬ mits to rule violations, and acknowledges a
sive acting-out behavior and substance abuse history of antisocial behavior in the past.
which resulted in considerable situational
stress. She shows a pattern of superficial guilt
Interpersonal Relations
or remorse over her behavior, but does not
accept much responsibility for her actions. She is probably experiencing disturbed inter¬
She may avoid confrontation and deny prob¬ personal relationships, possibly owing to her
lems. She reports no significant sex-role con¬ acting-out behavior. Her acting-out behavior
flicts. is likely to put great strain on her marriage.
In addition, the following description is She may be experiencing marital discord at
suggested by the content of this client’s re¬ this time.
sponses. She is preoccupied with feeling The content of this client’s MMPI-2 re¬
guilty and unworthy. She feels that she de¬ sponses suggests the following additional in¬
serves to be punished for wrongs that she has formation concerning her interpersonal rela¬
committed. She feels regretful and unhappy tions. She feels a moderate degree of family
about her life, and seems plagued by anxiety conflict at this time, and reported some trou¬
and worry about the future. She feels hope¬ blesome family issues. She feels that her fam¬
less at times and feels that she is a condemned ily life is not as pleasant as that of other peo¬
person. She has difficulty managing routine ple she knows. She feels like leaving home to
affairs, and the item content she endorsed escape a quarrelsome, critical situation, and
suggests a poor memory, concentration prob¬ to be free of family domination. Her social
lems, and an inability to make decisions. relationships are likely to be viewed by others
She appears to be immobilized and with¬ as problematic. She may visibly by uneasy
drawn and has no energy for life. She views around others, sits alone in group situations,
her physical health as failing and reports nu¬ and dislikes engaging in group activities.
 Interpreting the MMPI-2 Profile 323

Behavioral Stability manipulate others through suicidal gestures.

Thus, the possibility she might use prescrip¬
Individuals with this profile tend to have
tion medication for that purpose should also
long-standing personality problems and are
be taken into consideration.
presently experiencing situational distress.
The client endorsed item content, which
Although they might express a desire to
seems to indicate low potential for change.
change and feel remorse over past behavior,
She may feel that her problems are not ad¬
they tend to change only temporarily, eventu¬
dressable through therapy and that she is not
ally drifting back into the old pattern. Social
likely to benefit much from psychological
introversion tends to be very stable personal¬
treatment at this time. Her apparently nega¬
ity characteristic. Her generally reclusive in¬
tive treatment attitudes may need to be ex¬
terpersonal behavior, introverted life style,
plored early in therapy if treatment is to be
and tendency toward interpersonal avoidance
initiated successfully. Her item content sug¬
would likely be evident in any future test re¬
gests some family conflicts which are giving
sults.
her considerable concern at this time. She
feels unhappy about her life and resents hav¬
Diagnostic Considerations
ing an unpleasant home life. Psychological
Individuals with this profile are often diag¬ intervention with her could profitably focus,
nosed as having a Personality Disorder (De¬ in part, upon clarifying her feelings about her
pendent or Passive-Aggressive type) with a family.
Substance Use Disorder. The content of her In any intervention or psychological
responses underscores the antisocial features evaluation program involving occupational
in her history. These factors should be taken adjustment, her negative work attitudes could
into consideration in arriving at a clinical di¬ become an important problem to overcome.
agnosis. She holds a number of attitudes and feelings
that could interfere with work adjustment.
Treatment Considerations

Individuals with this profile may seek psy¬

chological therapy as an effort to reduce the
Computer Interpretation —MMPI-2
current situational distress they are experi¬
Adult Interpretive System8
encing. The sincerity of their motivation to Profile matches and scores for the client are
change their behavior should be carefully provided in Table 7-4.
evaluated. They may verbalize a great need
for help and show early gains, but as frustra¬
Configural Validity Scale Interpretation
tion mounts, they may terminate early. Act¬
ing-out behavior is a possibility once her This validity scale configuration is usually
anxiety and depression over her current prob¬ obtained by individuals who are admitting
lems diminish. She is probably experiencing personal and emotional problems, requesting
multiple problems that make it difficult to help with these problems, and are unsure of
focus treatment. Long-term behavioral change their own resources for dealing with them. As
may be difficult to obtain in her case. the elevation of the Fscale increases, these in¬
Individuals with this profile are often dividuals are acknowledging that they are ex¬
predisposed to Substance Use or Abuse Dis¬ periencing more problems and feeling worse
orders. Any treatment program involving or are overreporting their problems, perhaps
medication should be carefully monitored. to get help sooner.
Some individuals with this profile attempt to This configural interpretation should be
324 Chapter 7

TABLE 7-4 MMPI-2 Adult Interpretive System: Profile Matches and Scores for Example 2

Highest
Client Scale Best Fit
Scale Profile Codetype Codetype

Codetype match: 2-414-2 2-4Z4-2 (7)

Coefficient of Fit: .95 .96
Scores: ? (raw) 0
L 38 53 51
F 72 65 60
K 41 45 47
Hs U) 57 58 53
D (2) 92 76 71
Hy (3) 56 60 55
Pd (4) 84 78 73
Mf (5) 43 48 46
Pa (6) 52 59 54
Pt (7) 72 64 65
Sc (8) 66 61 57
Ma (9) 49 51 50
Si (0) 81 59 56

Mean Clinical
Elevation: 66 56 60
Ave age-males: 38 36
Ave age-females: 34 33
% of male codetypes: 4.0% 1.8%
% of female codetypes: 4.3% 1.2%
% of males within codetype: 66.5% 75.6%
% of females within codetype •
33.5% 24.4%

Configural clinical scale interpretation is provided in the report for the following codetype(s):
2-4Z4-2
2-4Z4-2 (7)

Note: Reproduced by special permission of Psychological Assessment Resources, Inc., from

MMPI-2 Adult Interpretive System by Greene, Brown, & PAR. Copyright 1 990. Further reproduc¬
tion is prohibited without permission from PAR, Inc.

the primary source of interpretive hypotheses L T = 38

for the L, F, and K validity scales. Individual
L scores in this range are usually ob¬
validity scale hypotheses, however, are also
tained by individuals who generally respond
presented in the following section.
frankly and openly to the test items and are
willing to admit to minor faults.
Validity Scales
F T = 72
? (raw) = 0 F scores in this range, if they are valid,
Scores in this range reflect a relatively suggest the increasing probability of serious
small number of unanswered items, which in psychological and emotional problems which
and of itself should not have an impact on the are often characteristic of severe neurosis,
validity of the profile. psychosis, or behavioral problems. Scores in
 Interpreting the MMPI-2 Profile 325

this range also may occur because individuals individuals with this codetype regardless of
have had some difficulty reading or under¬ the reasons for its occurrence: substance
standing the test items (evaluate measures of abuse, conflictual interpersonal and familial
consistency of item endorsement), or because relationships, and depression.
they have some motivation to overreport psy¬ These individuals often exhibit depres¬
chopathology (evaluate measures of accuracy sion and agitation in response to vocational
of item endorsement). or family problems, financial problems, legal
difficulties and/or substance abuse prob¬
K T = 41
lems. They perceive themselves as playing a
Scores in this range indicate limited per¬ significant role in these problems and are dis¬
sonal resources and open acknowledgment of tressed by them.
significant psychological distress. These indi¬ These individuals are usually very dissat¬
viduals are likely to have a relatively poor isfied with themselves and very dependent
self-concept, to be strongly dissatisfied with upon others. They tend to be somewhat intro¬
themselves but lacking the skills necessary to verted and shy, although they have adequate
change their situation, to be self-critical, social skills. They often are manipulative and
and/or to be extremely open and revealing. passive-dependent in their relationships with
Scores in this range may also reflect low ego others.
strength, a lack of insight into one’s self-mo¬
tivation and behavior, and ineffectiveness in Treatment
dealing with the problems of daily life. Prog¬
nosis for psychological intervention is usually The prognosis is generally poor with respect
guarded. to traditional methods of individual psycho¬
therapy. Marital or family therapy may be
somewhat effective in instances other than
Configural Clinical Scale Interpretation
characterologic or extremely severe pathology.

2-414-2 Codetype (High Match)

Possible Diagnoses
Clinical Presentation Axis I Rule out Adjustment Disorder
Rule out Mood Disorders
This codetype is very frequent and much Dysthymia
more common in men than women. It also is Major Depression
one of the more difficult codetypes to inter¬ Rule out Psychoactive Substance-
pret because of the multitude of factors that Abuse Disorders
can produce it. It is imperative that the clini¬ Axis II Rule out Antisocial Personality
cian note the other clinical scales that are ele¬ Disorder
vated and the relative elevations of the Rule out Passive Aggressive Per¬
supplementary and content scales. For exam¬ sonality Disorder
ple, a patient with this codetype who also ele¬ Rule out Borderline Personality
vates Pd2 (Authority Problems), ANG Disorder
(Anger), CYN (Cynicism), and ASP (Anti¬
2-414-2 (7) Codetype (Best Fit)
social Practices), is very different from a
patient who does not elevate these same
Clinical Presentation
scales and elevates ANX (Anxiety), DEP
(Depression), LSE (Low Self-esteem) and This codetype is very frequent and much
Scale 0. more common in men than women. It also is
A number of features are common to all one of the more difficult codetypes to inter-
326 Chapter 7

pret because of the multitude of factors that Computer Interpretation — Caldwell

can produce it. It is imperative that the clini¬ Report9
cian note the other clinical scales that are ele¬
vated and the relative elevations of the Test-Taking Attitude
supplementary and content scales. For exam¬
The client was open and mildly self-critical in
ple, a patient with this codetype who also ele¬
taking the MMPI-2. The basic validity scales
vates Pd2 (Authority Problems), ANG
were well within the acceptable limits.
(Anger), CYN (Cynicism), and ASP (Antiso¬
The supplemental validity scales show a
cial Practices), is very different from a pa¬
mild to moderate elevation on the “fake-
tient who does not elevate these same scales
bad” scale (Ds). This suggests an open will¬
and elevates ANX (Anxiety), DEP (Depres¬
ingness to report distressing symptoms if not
sion), LSE (Low Self-esteem) and Scale 0.
some overemphasis on or exaggeration of
These individuals often exhibit depres¬
them. It should be noted, however, that a se¬
sion and agitation in response to vocational
rious disorder, especially if it involved any
or family problems, financial problems, legal
mental confusion or even marginally psy¬
difficulties and/or substance abuse prob¬
chotic elements, could also contribute to such
lems. They perceive themselves as playing a
a willingness to self-disclose. Thus, most of her
significant role in these problems and are dis¬
mild elevation on scale F may be attributable to
tressed by them.
her self-criticalness, although some of it may be
These individuals are usually very dissat¬
secondarily due to psychopathology.
isfied with themselves and very dependent
The score on the scale measuring cur¬
upon others. They tend to be somewhat intro¬
rently attained, recently experienced, or
verted and shy, although they have adequate
wished for socioeconomic status (Ss) is below
social skills. They often are manipulative and
average, which is consistent with her K score.
passive-dependent in their relationships with
This overall pattern of scores suggests a per¬
others.
son of less than average sophistication and
socioeconomic status identification who is
Treatment openly reporting her distresses if not mildly
exaggerating them. Unless the following in¬
The prognosis is fair with traditional meth¬
terpretation is unequivocally a poor clinical
ods of individual psychotherapy. fit, however, it would not be justified to re¬
ject the test results as malingered.
Possible Diagnoses
Symptoms and Personality
Axis I Rule out Adjustment Disorder
Characteristics
Rule out Mood Disorders
Dysthymia The profile shows a severe level of anxiety
Major Depression and depression. The patterns suggests ex¬
Rule out Psychoactive Substance- treme low moods and complaints of ner¬
Abuse Disorders vousness, worry, fears, self-doubts, feelings
Axis II Rule out Avoidant Personality Dis¬ of inferiority, and loss of initiative. She ap¬
order pears prone to overreact with excessive anxi¬
Rule out Passive Aggressive ety and poorly regulated emotions to minor
Personality Disorder matters or even fancied threats. She is apt to
Rule out Borderline Personality become quite tense and ruminative and to
Disorder have chronic difficulties in getting to sleep.
 Interpreting the MMPI-2 Profile 327

The current level of her day-to-day coping Diagnostic Impression

and immediate practical self-sufficiency tests
The diagnoses most commonly associated with
as partially disorganized in a variety of areas.
this profile are of depressive and anxiety neuro¬
The profile indicates a severe passive-ag¬
ses. A secondary personality disorder diag¬
gressive or related personality disorder. She is
nosis such as dependent or passive-aggressive
likely to get many secondary gains from her
personality would also be typical. A few of
symptoms, even though her undercontrol of
these patients showed secondary schizoid
her impulses and lapses of judgment are self-
trends. A secondary diagnosis reflecting
defeating in the long run. Her ability to con¬
chronic dependence on alcohol may also be
form socially tests as poor. Repeated difficul¬
indicated.
ties and conflicts over limits on her behavior
are suggested. She appears quite immature
and insecure with indications of repeated Treatment Considerations
misunderstandings and longstanding resent¬ The profile suggests a mild to moderate sui¬
ments in her close personal relationships. She cide risk. In some similar cases the use of al¬
tests as vulnerable to increasing difficulties cohol effectively became a slow form of
with alcohol. suicide. Antidepressants and energizers
Persisting problems in regulating her ex¬ have been of limited benefit with patients
pressions of anger are indicated along with who obtained similar profiles. On the one
chronic, underlying resentments over depen¬ hand, her pattern is often associated with
dency frustrations. Fears of confirming her episodes of serious if not uncontrolled
self-dislike would lead to a self-protective in¬ chemical abuse; on the other hand, her
terpersonal distancing. These fears would, score on the MacAndrew Alcoholism Scale
however, repeatedly block self-assertive ex¬ was just below the chronic alcoholic and
pressions of anger. She would be acutely sen¬ chemical dependence range. Her responses
sitive to criticism. suggest asking if she has been in trouble with
She tests as severely introverted and so¬ the law. If currently involved, the stress of this
cially shy. She appears mildly to moderately could have precipitated or aggravated her
withdrawn. Her balance of masculine and symptoms or otherwise have led her to make
feminine interests is within the normal range professional contact.
for women. The profile emphasizes the importance of
Her profile is related to the dependency contacts with her relatives or other informants.
manipulative “Daddy’s girl’’ life-style pat¬ The clarification of all the precipitating circum¬
tern. She would play a daughter-to-father stances would be particularly indicated, in¬
role toward her husband, relating in depen¬ cluding possible adjustments to them that she
dent and immature ways to him as well as to may fail to mention. The family can be of
other males in her life. She is likely to be specific benefit in minimizing secondary
seen as clutching onto men in masochistic gains.
or even self-destructive relationships and as Treatment motivation is apt to decline as
becoming symptomatic when she was soon as her situation begins to improve and
threatened with losing such quasi-paternal external stresses are reduced. Many patients
supports. The husbands of patients with this with this pattern have terminated treatment
pattern have a notably high frequency of psy¬ before the therapist felt it to be complete,
chological breakdowns. This pattern is typi¬ some against therapeutic advice. She tests as
cally associated with histories of almost no pre¬ very prone to manipulate treatment, and the
marital dating. character problems strongly warn against in-
328 Chapter 7

volvement in her manipulations. She is apt to tions were having a hard time deciding
outwardly obliging because of her needs for whether the 2-4/4-2 codetype reflected a per¬
attention and affection, which would obscure sonality disorder, some form of depressive
her underlying resentments and passive-ag¬ disorder, and/or a substance use disorder.
gressive tendencies. The treatment relation¬ This confusion is real since all three diagnoses
ship may develop slowly because of her fears were appropriate for the client given her his¬
of being hurt and her emotional distortions. tory. Her discharge diagnoses from the hos¬
In general her emotional constrictions and pital were Heroin Dependence; Alcohol De¬
her tendency to declare certain topics “off pendence, in Remission; Dysthymia; and
limits” could necessitate careful handling Dependent Personality Disorder.
and patience in therapy. It is interesting to note that all four inter¬
The expected response to short-term pretations only discussed suicidal ideation
treatment is fair. She tests as prone to focus since the client did not report any suicide at¬
in interviews on her fears, worries, and short¬ tempts despite her history and reasons for
comings. She could benefit from a greater being hospitalized. In this specific instance,
awareness of the ways in which she sacrifices all four interpretations were incorrect be¬
long-term goals for immediate gratifications cause the client did not provide accurate in¬
and relief of anxiety. She could also benefit formation. All four interpretations did stress
from an increased awareness of the manipu¬ the importance of evaluating her suicidal ide¬
lations and countermanipulations of guilt ation carefully.
around her victim role. New activities in The client made satisfactory progress
which she needs to exert initiative with a risk while hospitalized and was discharged with a
of failure are apt to require repeated encour¬ significant improvement in her mood. Al¬
agement and reassurances. These could in¬ though she verbalized insight into her famil¬
clude projects to develop new activities and ial problems and the changes that she needed
interests as well as initiative in dealing with to make in her life, she chose to return to live
current personal dilemmas. with her disabled spouse. Subsequent fol¬
lowup interviews as an outpatient did not re¬
veal any significant changes in her life cir¬
General Comments
cumstances.
All four interpretations are in general agree¬
ment about the primary features of this ENDNOTES
client’s profile: mild to moderate depression 1. Reproduced from the MMPI-2 by per¬
and anxiety with associated guilt; difficulties mission. Copyright © 1943, (renewed 1970), 1989
with interpersonal relationships, particularly by the University of Minnesota. Published by the
with members of her family; significant so¬ University of Minnesota Press. All rights reserved.
cial introversion and avoidance of social in¬ 2. The parenthetical references to specific
teraction; history of antisocial behavior; scales or indexes are provided so clinicians can see
presence of suicidal ideation; and obsessive the source(s) for the statements in the interpreta¬
and ruminative thoughts. tion. These parenthetical references would be de¬
leted in an actual report.
The four interpretations did not concur
3. The permission of the University of Min¬
whether the client was angry and hostile; an¬
nesota Press to reproduce this report is gratefully
tidepressant medications would be appropri¬
acknowledged.
ate; acting-out will occur; substance use or 4. Reproduced by special permission of
abuse is a problem; and the client should have Psychological Assessment Resources, Inc., from
a personality disorder diagnosis. It is appar¬ MMPI-2 Adult Interpretive System by Greene,
ent that the computer-based test interpreta¬ Brown, & PAR, Copyright 1990. Further repro-
 Interpreting the MMPI-2 Profile 329

duction is prohibited without permission from nesota Press to reproduce this report is gratefully
PAR, Inc. acknowledged.
5. The permission of Dr. Alex B. Caldwell, 8. Reproduced by special permission of
Caldwell Report, Los Angeles, California, to re¬ Psychological Assessment Resources, Inc., from
produce this report is gratefully acknowledged. MMPI-2 Adult Interpretive System by Greene,
6. The parenthetical references to specific Brown, & PAR, Copyright 1990. Further repro¬
scales or indexes are provided so clinicians can see duction is prohibited without permission from
the source(s) for the statements in the interpreta¬ PAR, Inc.
tion. These parenthetical references would be de¬ 9. The permission of Dr. Alex B. Caldwell,
leted in an actual report. Caldwell Report, Los Angeles, California, to re¬
7. The permission of the University of Min¬ produce this report is gratefully acknowledged.
 CHAPTER 8

Specific Groups:
Adolescents, the Aged, Blacks,
and Other Ethnic Groups

The original standardization sample of the adolescents, administered the MMPI to 3,971
MMPI consisted of white Minnesota adults, ninth-graders (mean age about 15) in the
primarily between the ages of 16 and 55 Minneapolis public school system during the
(Hathaway & McKinley, 1940). Investigators 1947-1948 school year. They also micro¬
suggest that the validity of the MMPI with filmed each student’s school record at the
persons who differ in age or ethnicity from time of testing. Two and four years later,
the original standardization sample may be they determined how many of these students
improved by modifying either the norms or had records with the local juvenile division of
the scales themselves. The MMPI-2, which the police department or probation office.
has a more nationally representative norma¬ During the spring of 1954 these research¬
tive sample, will begin to provide some current ers tested 11,329 additional ninth-graders
data on the issues that are raised in this chapter. who represented a sample of the entire state
This chapter will review the use of the of Minnesota. They again microfilmed each
MMPI with adolescents, the aged, and several student’s school record, had each student
ethnic groups. The MMPI-2 as currently pub¬ complete a personal data sheet, and obtained
lished is not intended for use with adolescents the teacher’s prediction of which students
(i.e., anyone under 18 years of age) (Butcher, were likely to have legal or emotional diffi¬
Dahlstrom, Graham, Tellegen, & Kaemmer, culties. Three years later they determined
1989, p. 15). An adolescent revision of the how many of these students had records by
MMPI is planned for publication in August examining the student’s local community po¬
1991. Until that revision is published, the original lice and court files.
MMPI is intended to be used with adolescents. In 1957, when most of this second set of
ninth-graders were now twelfth-graders (mean
age about 18), Hathaway and Monachesi read¬
ADOLESCENTS
ministered the MMPI to 3,976 students.
Hathaway and Monachesi (1963), in an ex¬ Hathaway and Monachesi obviously have a
tensive study of the MMPI performance of wealth of data, only part of which is relevant

331
332 Chapter 8

to the issue of how adolescents differ from Hathaway and Monachesi (1963) did not an¬
adults in terms of their MMPI performance. swer this question.
(The reader who is interested in the use of the The students who were retested with the
MMPI to predict delinquency in adolescents, MMPI in the twelfth grade provided some in¬
which was the main thrust of Hathaway and teresting information on profile stability.
Monachesi’s research project, should consult Test-retest reliability coefficients were high¬
their book.) est for the K scale—.52 for males and .56 for
Relevant to the current topic of MMPI females—and Scale 0 (Social Introversion)—
norms, Hathaway and Monachesi found that .54 for males and .61 for females. By con¬
ninth-graders had mean scores with K-correc¬ trast, Scales 4 (Psychopathic Deviate)—.36
tions on Scales 4 (Psychopathic Deviate), 8 for males and .38 for females—and 6 (Para¬
(Schizophrenia), and 9 (Hypomania), which noia)—.32 for males and .36 for females—
were approximately 10 T score points higher had the lowest reliability coefficients.
than the original Minnesota normative sam¬ As might be expected, profile stability as
ple of adults. On the rest of the validity and defined by the single high-point scale was
clinical scales, the ninth-graders scored very highest when that scale was greater than a T
similarly to the adult sample. Thus, these ad¬ score of 69. For example, more than half of
olescents were more likely to have socio- the adolescents, both male and female, with
pathic (Scale 4 elevations) or psychotic (ele¬ Scale 4 greater than a T score of 69 on initial
vations on Scales 8 and 9) profile patterns if testing, had Scale 4 as one of the two highest
adult MMPI norms were used. They also scales when retested. This relationship, how¬
were less likely to have profiles in which all ever, did not hold for all scales. Scale 5 (Mas¬
clinical scales were below a T score of 70 than culinity-Femininity), for example, was likely
were the normal adults. to shift from a high-point scale to one of the
The issue that Hathaway and Monachesi three lowest clinical scales across this three-
did not directly address is whether these year interval.
MMPI scale elevations in normal adolescents Hathaway and Monachesi wondered
reflect some form of psychological distress whether there may be an active differentia¬
and maladjustment that is characteristic of tion of gender roles during this time span that
the turmoil of adolescence or whether these results in Scale 5 scores changing so dramati¬
elevations reflect mere differences in the fre¬ cally. Whatever the reason, Scale 5 scores did
quency of item endorsement that may not change drastically in some adolescents during
have psychopathologic implications. this time interval.
Hathaway and Monachesi seem to favor Following the lead of Hathaway and
the former interpretation since they did not Monachesi, Ball (1962) also examined the re¬
advocate the use of special adolescent lationship of personality to social deviancy
norms with the MMPI. Instead, they sug¬ among Kentucky adolescents. Only the data
gested that both the standard adult profile on the nondelinquent adolescents will be re¬
and an adolescent-normed profile should viewed here since it is germane to the issue of
be plotted so that the clinician can under¬ how normal adolescents differ from the orig¬
stand the contrast between adolescents and inal Minnesota normative sample on the
adults. The clinician, however, when pro¬ MMPI. Both male and female adolescents
vided with the potentially divergent and con¬ achieved mean T scores near 60 on Scales F, 4
tradictory information from two profiles on (Psychopathic Deviate), 7 (Psychasthenia), 8
the same adolescent, needs to know which (Schizophrenia), and 9 (Hypomania), with
source of information is more accurate, and the other clinical scales averaging near a
 Specific Groups: Adolescents, the Aged, Blacks, and Other Ethnic Groups 333

mean T score of 50. This pattern of scores is and 1965 from both rural and urban, public
very similar to what Hathaway and and private school students residing in Ala¬
Monachesi (1963) reported for their Minne¬ bama, California, Kansas, Missouri, North
sota adolescents who were tested nearly 10 Carolina, and Ohio. All of these students
years earlier. were presumed to be white and at the time of
In a comprehensive study of black-white testing were neither institutionalized nor
differences on a variety of intellectual, aca¬ being treated for emotional disturbance
demic, and personality factors, Baughman (Marks et al., 1974).
and Dahlstrom (1968) reported the MMPI By combining these two groups of stu¬
performance of eighth-graders. Only discus¬ dents, the adolescent norms for the standard
sion of the data on white adolescents will be validity and clinical scales for boys and girls
reviewed in this section; the data examining in four age groups (14 and below, 15, 16, and
black-white differences in MMPI perfor¬ 17 and 18) were constructed. Only if adoles¬
mance will be reviewed in a later section of cents were still living with their parents were
this chapter. Again, a similar pattern of 18-year-olds included in the oldest group. If
MMPI scale scores was found in the white an 18-year-old was not living with his or her
eighth-graders. Scales F, 4 (Psychopathic De¬ parents, it was deemed more appropriate to
viate), 7 (Psychasthenia), 8 (Schizophrenia), use adult norms, although there has not been
and 9 (Hypomania) were elevated to a mean any research to demonstrate which set of
T score of nearly 60 in both girls and boys, norms is better.
with Scales 8 and 9 elevated almost to a mean Although Marks and colleagues (1974)
T score of 70 in boys. advocate that emotional disturbance in ado¬
It appears that normal adolescents dis¬ lescents needs to be established against ado¬
play significant elevation on a number of the lescent norms, they do concur with
standard MMPI validity and clinical scales Hathaway and Monachesi (1963) that adoles¬
when the adult norms from the original Min¬ cent scores on the MMPI also should be com¬
nesota normative sample are used. If the cli¬ pared to adult norms. Thus, Marks and col¬
nician is interested in knowing how an ado¬ leagues recommend that both an adult- and
lescent client compares with normal an adolescent-normed profile should be con¬
adolescents rather than with the original, structed for adolescents.
adult normative group on the MMPI, an ad- The clinician who desires to construct a
olescent-normed profile, described in the profile from adolescent norms must locate
next section, should be constructed. the correct table for the age and gender of the
Marks, Seeman, and Haller (1974) have adolescent and determine the T score equiva¬
done the most extensive MMPI research on lent of the raw score on each scale; these ta¬
emotional disturbance in adolescents. They bles are not K-corrected. These T scores are
also have described how the adolescent then directly plotted on the standard profile
norms for the MMPI were derived, primarily sheet. The clinician should indicate on the
crediting Peter F. Briggs. Briggs selected 100 profile sheet that the profile was plotted using
boys and 100 girls aged 14, 15, and 16 years, adolescent norms; this procedure will prevent
and 80 boys and 40 girls aged 17 years from another clinician who uses the profile from
the students studied by Hathaway and assuming that the profile has been
Monachesi (1963). misplotted, since the adolescent and adult T
In order to obtain a larger and more na¬ scores for each raw score will not be the
tionally representative sample, 1,046 addi¬ same. Adolescent profile sheets are available
tional MMPI profiles were collected in 1964 from Psychological Assessment Resources
334 Chapter 8

(P.O. Box 998, Odessa, FL 33566, 800-331- pages long, and the clinician will need to con¬
8378) so that clinicians can plot the T score sult Marks and associates’ (1974) book to use
equivalents of the raw scores directly without their interpretive system.
consulting other data. For profiles that do not fit into one of
In addition to describing the develop¬ these 29 codetypes, Marks and colleagues rec¬
ment of adolescent norms, Marks and associ¬ ommend that the second highest scale should
ates (1974) assessed the empirical correlates be disregarded and that the profile should be
of MMPI codetypes in a sample of 834 white reclassified using the first and third highest
boys and girls between the ages of 12 and 18 scales. If the profile is still not classifiable
who were not mentally retarded but who had using the first and third highest scales, the
adjustment problems causing them to seek or profile is considered to be unclassifiable, and
be referred for professional services. Each of their interpretive system cannot be used.
these teenagers completed a personal data Only one study (Williams & Butcher,
sheet, and for each teenager a therapist com¬ 1989b) has attempted to validate the corre¬
pleted a case data schedule, an adjective lates of any of the codetypes developed by
checklist, and a Q-sort after a minimum of 10 Marks and colleagues in another sample of
hours of therapy. All of the therapist’s rat¬ adolescents; this study will be reviewed
ings were made independent of the MMPI. below.
Each of the descriptors from the infor¬ Marks and associates (1974) dealt with
mation provided by the teenager and the ther¬ the problem of testing adolescents with the
apist was evaluated for gender differences MMPI by using adolescent norms, which
and deleted if significant. Marks and col¬ sometimes lead to codetypes different from
leagues also eliminated descriptors with ex¬ those that would be derived if the adult
ceedingly high (> 90 percent) or low (< 10 norms were used. In addition, they used an
percent) frequencies of occurrence. This pro¬ adolescent population to develop a set of be¬
cedure resulted in a pool of 1,265 descriptors, havioral correlates, or narrative descriptions,
which were then related to the codetypes on for these codetypes. Marks and associates’
the MMPI based on adolescent norms. narrative descriptions have not been cross-
When ties occurred among the high- validated in another adolescent population,
point scales, the profile was classified into a and the research in this area has yielded
codetype on the basis of the scale with the mixed results.
lowest number. For example, if Scales 7, 2, Ehrenworth and Archer (1985) found
and 3 were the three highest scales and had that interpretations based on the descriptions
equivalent T scores, the profile was classified by Marks and colleagues were rated as being
as a 1-2/2-1 codetype. Similarly, if Scale 2 less accurate than other interpretations in a
was the highest and Scales 5, 4, and 7 were sample of adolescent inpatients, whereas
tied for the second highest scale, the profile Archer, Gordon, Giannetti, and Singles
was classified as a 2-3/3-2 codetype. (1988) reported that the clinical correlates of
Marks and colleagues were able to iden¬ single clinical scales in adolescent inpatients
tify 29 codetypes, irrespective of scale order, were similar to those of Marks and col¬
with a minimum of at least 10 cases per high- leagues. Consequently, clinicians should use
point pair. They then developed an actuarial the Marks and associates’ descriptions cau¬
description based on the descriptors that were tiously.
significantly associated with each of these 29 There has been an almost geometric in¬
codetypes. The narrative description for each crease in research on the use of the MMPI
of these codetypes is approximately two with adolescents in the last few years. This in-
 Specific Groups: Adolescents, the Aged, Blacks, and Other Ethnic Groups 335

crease in research has been spearheaded by interpretation once adolescent norms serve as
the work of Archer (cf. Archer, 1984, 1987, the base to derive a codetype. Using adoles¬
1988) and Williams (cf. Williams, 1986; Wil¬ cent and adult norms, Lachar, Klinge, and
liams & Butcher, 1989a, 1989b). Their contri¬ Grisell (1976) obtained valid profiles on 100
butions should be reviewed by any clinician adolescents, most of whom were hospital¬
who is using the MMPI with adolescents. ized. Interpretations of the two profiles for
Archer (1987) provides a single, comprehen¬ each adolescent were generated using
sive overview of this entire area. Lachar’s (1974) automated interpretive sys¬
Several issues arise when the MMPI is tem for adults. Clinicians were then asked to
used with adolescents. The first issue is rate these interpretations for accuracy.
whether adult norms, adolescent norms, or The interpretations of profiles generated
both sets of norms should be used with ado¬ from adolescent norms, in comparison to in¬
lescents. As noted above, Hathaway and terpretations of profiles generated from adult
Monachesi (1963) and Marks and associates norms, were rated more accurate 61 times, as
(1974) recommend that adult- and adoles- accurate 13 times, and less accurate 26 times.
cent-normed profiles should be constructed Only 10 percent of the adolescent norm inter¬
for adolescents. However, Archer (1984, pretations were judged to be inaccurate, and
1987) advocates that adolescent norms 20 percent of the adult norm interpretations
should be used exclusively with adolescents, were judged to be inaccurate. Wimbish
and he presents a cogent rationale for their (1984) reported similar findings in a sample
use that should be read by clinicians. Wil¬ of inpatient adolescent substance abusers.
liams (1986) also indicated that adolescent Hence, it appears that even when an in¬
norms are the most appropriate for adoles¬ terpretive system (Lachar’s) based on adults
cents, but she suggests that both profiles is used, profiles based on adolescent norms
should be plotted for adolescents. provide a more accurate description of ado¬
Once it has been decided that adolescent lescents than do profiles based on adult
norms will be used, the next issue is whether norms. It remains to be seen, however,
T scores of 70 and above should continue to whether other adult interpretive systems and
be the criterion for defining a significant ele¬ individual clinicians will demonstrate the
vation on the scales. Since adolescent norms same improvement in accuracy with the use
produce a general lowering of the entire pro¬ of adolescent norms.
file, Archer (1984, 1987; Ehrenworth & Once the decision has been made to use
Archer, 1985) has recommended that T adolescent norms, the clinician now has three
scores of 65 and above be used to define sig¬ different sets of norms from which to choose:
nificant clinical elevations similar to the pro¬ (1) Marks and colleagues (1974), which have
cedure now being used on the MMPI-2. been described above and which have served
Additional empirical research that as¬ as the standard for adolescent MMPIs since
sesses whether a T score of 65 or 70 facilitates they were originally published; (2) Gottes-
the interpretation of adolescent MMPIs is man, Hanson, Kroeker, and Briggs (1987);
needed. In the interim, clinicians would seem and (3) Colligan and Offord (1989).
to be well advised to follow Archer’s (1984, Gottesman and associates developed
1987) suggestion and use adolescent norms their norms by using the MMPI responses of
for adolescents and interpret their profiles at 12,953 15-year-olds and 3,492 18-year-olds,
T scores of 65 and above. who composed the entire sample of adoles¬
A final issue is whether the adolescent or cents studied by Hathaway and Monachesi
the adult correlates provide a more accurate (1963). Instead of using linear T scores as tra-
336 Chapter 8

ditionally had been done with the MMPI, tions as suggested previously by Archer
Gottesman and colleagues used a normalizing (1987) and Williams (1986).
procedure based on power transformations Although, Williams and Butcher (1989b)
of the raw scores to produce T scores that found that the relative frequency of the vari¬
had equivalent percentile ranks across the ous codetypes in their sample was very com¬
scales. parable to those reported by Marks and col¬
Colligan and Offord (1989) developed leagues (1974), they were unable to replicate
their norms by collecting a sample of adoles¬ the correlates of specific codetypes found in
cents in the Mayo Clinic catchment area in a Marks and colleagues or the adult codetype
similar manner as they had collected their literature. This latter finding suggests that the
adult data (see Chapter 1). Colligan and validity of traditional MMPI codetypes in ad¬
Offord also used normalized T scores in their olescents may be limited.
adolescent sample as they had in their adult Since the development of an adolescent
sample (see Chapter 2). Since the same issues interpretive system for the MMPI is still in its
of codetype concordance can be raised infancy, one solution may be to use adoles¬
among these various sets of adolescent norms cent norms and empirical correlates of indi¬
as were discussed in Chapter 6 about code¬ vidual validity and clinical scales based on
type concordance between the MMPI and the adolescents when testing this population. An
MMPI-2, they will not be reiterated here. alternative procedure, which can be followed
Archer, Pancoast, and Klinefelter (1989) while an adolescent interpretive system such
reported the codetype concordance among as Marks and associates (1974) is being cross-
these three sets of adolescent norms to range validated, is to use adolescent norms to gen¬
from 22 to 31 percent for an adolescent sam¬ erate the profile and then to use an adult in¬
ple of 100 inpatients, 100 outpatients, and terpretive system.
100 normals. They did not find the two newer This procedure, as Lachar and associates
sets of norms to have higher hit rates in clas¬ (1976) and Wimbish (1984) demonstrated,
sifying adolescents as normal or psychologi¬ provides some improvement over utilizing
cally distressed, but they did substantiate both adult norms and adult correlates of a
higher sensitivity rates using a T score of 65 codetype to interpret an adolescent’s MMPI.
rather than a T score of 70 to classify an ado¬ However, it must be noted that Williams and
lescent as psychologically distressed. They Butcher (1989b) were unable to replicate the
cautioned clinicians about using these newer correlates of specific adolescent codetypes,
sets of adolescent norms to produce a code¬ which suggests that clinicians should be cau¬
type that would be interpreted based on the tious in using this procedure.
existing MMPI research because of the lim¬
ited concordance rates.
THE AGED
Williams and Butcher reported the em¬
pirical correlates of the MMPI standard va¬ There have been only limited investigations
lidity and clinical scales (1989a) and of the performance of the normal aged on the
codetypes (1989b) in a large, inpatient, ado¬ MMPI. Brozek (1955) reported MMPI data
lescent sample. These adolescents had similar on 233 business and professional men be¬
descriptors associated to the individual tween the ages of 45 and 55 who were partici¬
MMPI scales that had been found in adults. pating in a program examining the effects of
Williams and Butcher (1989a) concluded that aging on the cardiovascular system. All of
their results supported the use of adult de¬ these men had normal blood pressure and
scriptors for adolescent MMPI interpreta¬ were in good physical health.
 Specific Groups: Adolescents, the Aged, Blacks, and Other Ethnic Groups 337

Brozek compared the performance of striking stability of scores over this 30-year
these older men with a group of 157 male col¬ interval for Scales 5, 9, and 0, and 23.9 per¬
lege students. He reported T scores for only cent of these men had the same codetype.
three validity scales (?, L, and K ) and Scales Leon and associates (1979) suggest that
1 (Hypochondriasis) and 0 (Social Introver¬ the elevations on Scales 2 and 3 in an aging
sion). The mean T score on all five of these sample may reflect realistic somatic concerns
scales was nearly 50 in both groups of men; rather than the usual depressive and histrion¬
that is, no scale was noticeably different from ics correlates found in younger adults.
the mean T score of the original Minnesota The most comprehensive examination of
normative sample. MMPI performance in the aged was con¬
Brozek found that the older men en¬ ducted by Swenson (1961). He asked 210 per¬
dorsed four to nine more items on Scales 1 sons aged 60 years or more who had com¬
(Hypochondriasis), 2 (Depression), 3 (Hyste¬ pleted a brief attitude questionnaire to take
ria), 5 (Masculinity-Femininity), 6 (Para¬ the MMPI. Only 95 (45.1 percent) of these
noia), and 0 (Social Introversion) and en¬ persons completed the MMPI: 31 men and 64
dorsed six to seven fewer items on Scales 7 women with a median age of 71.4 years.
(Psychasthenia), 8 (Schizophrenia), and 9 None had a known mental disorder sufficient
(Hypomania) than the younger men. He con¬ to warrant a psychiatric diagnosis. (It is pos¬
cluded that within the age ranges (20 to 55) of sible that the large percentage of older indi¬
his study, the T scores derived from the origi¬ viduals who did not complete the MMPI may
nal Minnesota normative sample were valid. somehow bias the representativeness of this
Brozek also provided a detailed analysis of sample, but the nature of these biases is un¬
items that separated the older and younger known.)
age groups. The reader who is interested in The scores of these 95 persons served as
this information should consult Brozek’s the basis for constructing the T score norms
(1955) article. for the aged. They scored from 6 to 9 T score
Leon, Gillum, Gillum, and Gouze (1979) points higher on Scales 1,2,3, and 5 and 3 T
have reported 30-year followup data on the score points lower on Scale 9 then the original
MMPI performance of the healthy profes¬ Minnesota normative sample. Scales 1, 2, 3,
sional men described by Brozek. These re¬ and 0 were the most frequently occurring
searchers were able to obtain MMPI data on high-point scales in both genders. The me¬
71 of these men who had completed the dian profiles for this aged sample were com¬
MMPI at four different times (1947, 1953, monly of a neurotic pattern (elevations on
1960, and 1977). The mean age of the 71 men Scales /, 2, and 3) and only rarely gave evi¬
in 1977 was 77 years. dence of a psychotic or a behavior disorder
When the scale scores from 1947 were pattern.
contrasted with those in 1977, these men Swenson, Pearson, and Osborne (1973)
achieved significantly higher scores on all va¬ provided MMPI data on 1,733 male and
lidity and clinical scales except the L scale, al¬ 1,471 female medical outpatients 70 years of
though all scale scores were still below a mean age and older who were seen at the Mayo
T score of 62. When scale scores from 1960 Clinic between 1962 and 1965. They excluded
were compared with those in 1977, these men any patient referred primarily for psychiatric
achieved significantly higher scores on Scales evaluation and treatment. Both male and fe¬
L, 2, 7, 8, and 0. The largest difference in male patients scored about a T score of 60 on
scale score between 1960 and 1977, however, Scales 1, 2, and 3, a T score of 55 on Scale 0,
was only 5 T points on Scale 2. There was and a T score of 48 on Scale 9. This pattern of
338 Chapter 8

scores is very similar to what Swenson (1961) 7, 2, and 3 about 10 T score points, elevate
reported. Swenson and colleagues (1973), Scale 0 about 5 T score points, and lower
however, report similar elevations on Scales Scale 9 about 3 T score points compared to
7, 2, and 3 in medical patients aged 16 to 69. the original Minnesota normative sample.
Thus, it is not possible to conclude whether Using aged norms in constructing a profile
the elevations on Scales 7, 2, and 3 in patients for a person 70 years or older will allow the
70 years of age and older reflect age per se or clinician to compare the person’s scores to
some other factor. others of similar age. There is no published
Fillenbaum and Pfeiffer (1976) collected research, however, on how to interpret an
the Mini-Mult (Kincannon, 1968) as a por¬ MMPI profile constructed on aged norms.
tion of a broad questionnaire administered to Again, research is needed to determine
a 10 percent random sample of noninstitu- whether correlates of profiles constructed on
tionalized persons 65 years and older living in the original Minnesota normative sample will
a southern community. The Mini-Mult was apply in a similar fashion to an aged sample.
read to the person by a trained interviewer, Since the representativeness of the sample
and responses were obtained to all 71 items used in constructing the aged-norm tables is
on the Mini-Mult from 249 men (171 white, unclear because of the limited sample size
78 black) and 391 women (250 white, 141 and the large number of persons who did not
black). An additional 298 persons answered complete the MMPI, the clinician should use
some of the items, and 59 answered none of the aged norms very cautiously.
the items; their responses were not ana¬ It might be best for the clinician to plot
lyzed. two profiles for an aged client: the standard
Fillenbaum and Pfeiffer found only profile and an aged-norm profile. By examin¬
one item (25) that was endorsed differently as ing the latter profile, the clinician could de¬
a function of gender or race. The mean T termine how the client compares on the vari¬
scores in both genders and races for the eight ous scales with persons in his or her age
clinical scales measured by the Mini-Mult range. The clinician, however, probably
(which excludes Scales 5 and 0) were approxi¬ should interpret the standard profile and use
mately 60, except for Scale 9, which was ap¬ the aged norm profile only to supplement the
proximately 50. Despite there being only one interpretation.
item that was significantly different for
blacks and whites, blacks had significantly
ETHNIC GROUPS1
higher T scores (2 to 5 points) on Scales F, 4,
8, and 9. Most research on the MMPI in minority
If this sample can be considered to be rel¬ groups has focused on blacks, although re¬
atively normal, it appears that the Mini- cently there has been some research with His-
Mult overestimates the amount of psycho¬ panics,2 Asian-Americans, and American
pathology in an aged sample and hence will Indians. The first issue, of course, is whether
have limited usefulness. Clinicians should re¬ minority group status does influence MMPI
view the discussion of short forms of the results when other potential confounding fac¬
MMPI in Chapter 5 before using them with tors such as socioeconomic status and educa¬
any clients. tional level are controlled. If minority group
The limited amount of research on status does affect MMPI scale scores when
MMPI performance in normal-aged samples these factors are controlled, the critical issue
allows for only a few general conclusions. It becomes whether MMPI interpretations
appears that aged persons may elevate Scales based on white or minority group norms are
 Specific Groups: Adolescents, the Aged, Blacks, and Other Ethnic Groups 339

more valid. Each of these issues will be exam¬ be considered that can potentially affect the
ined in turn. interpretation of any ethnic differences that
The MMPI also has been translated into are found.
a number of different languages for use Researchers tend to be very casual about
around the world (cf. Butcher, 1985; Butcher reporting the criteria used to specify member¬
& Clark, 1979; Butcher & Pancheri, 1976), ship within the ethnic group whose MMPI
which allows for the investigation of the effects performance is being evaluated. Most often,
of cultural factors on MMPI performance the persons are described as black, Hispanic,
from an international perspective. The reader Asian-American, and so on, with little con¬
who is interested in this perspective should con¬ sideration of whether they have any actual
tact these resources for additional information. identification with their ethnic group.
Dahlstrom, Lachar, and Dahlstrom Sometimes Hispanic subjects, and even
(1986) have provided a thorough, in-depth less frequently, Asian-American subjects,
analysis of the influence of ethnic group will be classified on the basis of their sur¬
membership on MMPI performance, and name, which at least assures that at some
Dana (1990) has provided a general overview point in their family heritage they had mem¬
of multiethnic assessment. Also, there have bership in the ethnic group. However, a sur¬
been a number of other recent reviews of the name does not determine whether persons ac¬
research in this area that should be consulted tually are members of the ethnic group, or
for a variety of perspectives on this topic more important, whether they have any iden¬
(Costello, Tiffany, & Gier, 1972; Gynther, tification with that group.
1972, 1979, 1989; Gynther & Green, 1980; Recently, in an attempt to address this
Pritchard & Rosenblatt, 1980a, 1980b). problem, some researchers have used the Ac¬
The prototypic investigation of the ef¬ culturation scale for Mexican-Americans
fect of ethnic group membership on MMPI (Cuellar, Harris, & Jasso, 1980) as a modera¬
performance consists of obtaining a sample tor variable in examining MMPI perfor¬
of individuals from some specific ethnic mance in Hispanic individuals (e.g., Mont¬
group and then plotting the obtained profile gomery & Orozco, 1985). In a similar vein,
against the original normative group or an¬ Costello (1977) proposed a scale that can be
other ethnic group. If differences occur be¬ used to assess “functional” identification
tween the two groups on any scale, the typical with black culture, although it has not yet
conclusion is that the MMPI as a whole or been used in any reported study.
some subset of scales is affected by member¬ Clearly, researchers need to become
ship in that ethnic group. more sophisticated in assessing membership
Rarely do such prototypic studies deter¬ in and identification with a specific ethnic
mine whether these obtained differences are group when they are interested in examining
of sufficient magnitude to affect clinical in¬ MMPI performance, because the cultural
terpretation and, more importantly, they vir¬ factors that are so important in determining
tually never determine whether these differ¬ how psychopathology will be manifested in a
ent scores actually affect the empirical person may be only loosely related to ethnic
correlates of the scale. A number of issues group membership.
must be considered before the conclusion is
justified that the observed results reflect dif¬
Black-White Comparisons
ferences as a function of ethnic group mem¬
bership. Greene (1987) has summarized a The early research of black-white differences
number of methodologic issues that need to on the MMPI compared scores on the stan-
340 Chapter 8

dard validity and clinical scales. The general Table 8-1 summarizes the extant litera¬
procedure for these studies was to administer ture that has examined black-white differ¬
the MMPI to a group of blacks and whites ences on the standard validity and clinical
and then to compare their mean scores on the scales of the MMPI. Since MMPI perfor¬
validity and clinical scales. mance varies as a function of the setting, this
Ball (1960) contrasted the MMPI perfor¬ table has been separated into normal, prison,
mance of 31 black ninth-graders with 167 substance abuse, medical/welfare, and psy¬
white students in the same classes. The black chiatric samples. Even a cursory review of
male students earned higher scores on Scale 1 Table 8-1 should suggest that there is no con¬
(Hypochondriasis) than the white male stu¬ sistent pattern to black-white differences on
dents. The black female students earned the MMPI.
higher scores on Scales F, 8 (Schizophrenia), Clearly, a statement that blacks or
and 0 (Social Introversion) and lower scores whites routinely score higher on any specific
on Scales K and 3 (Hysteria) than the white validity or clinical scale is not warranted
female students. Ball noted that, in addition across this entire range of studies. Even
to ethnicity, his two groups differed in school within specific populations, there are few
grades, level of intelligence, and social class. generalizations that can be made in a reliable
McDonald and Gynther (1962) con¬ manner.
ducted a similar study using black students For example, in normal populations it
from segregated high schools. The black male has been stated that blacks frequently score
students achieved higher scores than white higher on the F scale and Scales 8 (Schizo¬
male students on Scales L, F, K, 1 (Hypo¬ phrenia) and 9 (Hypomania) than whites.
chondriasis), 2 (Depression), and 9 (Hypo- However, a majority of the 27 comparisons
mania), whereas the black female students within normal samples found no reliable dif¬
achieved significantly different scores from ferences on these three scales, although when
white female students on all scales except K differences do occur, blacks score higher
and 8. McDonald and Gynther (1962) noted than whites. In fact, there is no scale on
that their black students were primarily lower which a majority of the studies was consis¬
class, and the white students were middle and tently higher either in blacks or whites in any
upper class. Since socioeconomic status could sample. Consequently, it appears inappropri¬
have been a factor in producing the black- ate to conclude that blacks routinely score
white differences, McDonald and Gynther higher than whites on any of the standard va¬
(1963) examined another group of black and lidity or clinical scales in any sample.
white students from segregated high schools It is possible that the black samples in
who were matched on the basis of their these studies may have differed in their iden¬
parents’ occupation. tification with a black culture, which would
They reported that social class as mea¬ tend to obscure any real black-white differ¬
sured by parents’ occupation was not related ences. Harrison and Kass (1968), for exam¬
to scores on any MMPI scale. Both male and ple, found that northern blacks scored be¬
female black students achieved higher scores tween southern blacks and whites on Scales ?,
on Scales L, F, 2, 8, and 9 and lower scores F, 9, and 0, which led them to conjecture that
on Scale 3 than their white counterparts. Mc¬ northern blacks may be acquiring white cul¬
Donald and Gynther concluded that these ture. Moreover, Erdberg (1975) found that
black-white differences on the MMPI were black-white differences were less in urban
culturally determined and did not reflect a than in rural settings.
bias in social class. The three studies (Erdberg, 1975; Me-
 Specific Groups: Adolescents, the Aged, Blacks, and Other Ethnic Groups 341

TABLE 8-1 Black-White Differences on MMPI Standard Validity and Clinical Scales

Validity
Scales Clinical Scales

Study L F K 1 2 3 4 5 6 7 8 9 0

Normal Samples

Ball (1960)
Female B W W B B
Male B
Butcher, Ball, & Ray (1964)
Female B B W
Male B B W w W B W
Erdberg (1975)
Rural female W B W w B W B B B
Rural male W B W B B B B B B
Urban female W W W W w B W
Urban male
Gynther, Fowler, & Erdberg
(1971)
Female B W B B B B B B B
Male B W B B B B B B B B
Harrison & Kass (1967),
female
King, Carroll, & Fuller (1977),
male B
Kirk & Zucker (1 979),a male
McDonald & Gynther (1962)
Female B
Male B B
McDonald & Gynther (1963)b
Female, social class 1-2 B B
Female, social class 3
Female, social class 4
Male, social class 1-2 B B
Male, social class 3 B B
Male, social class 4 B B B B
Moore & Handal (1980)
Female B
Male B B B B B
Muller & Bruno (1988), male
Sutker & Kilpatrick (1973)
Female B
Male
Walters, Greene, & Jeffrey
(1 984), male

Prison Samples

Caldwell (1953), male B B

Cooke, Pogany, & Johnston
(1974), male W B
continued
342 Chapter 8

TABLE 8-1 continued

Validity
Scales Clinical Scales

Study L F K 1 234567890

Prison Samples

Costello, Fine, & Blau (1973),

male
Elion & Megargee (1975),c
male B
Flanagan & Lewis (1969),
male B
Fry (1 949), male B
Holcomb & Adams (1982),
male B
Holcomb, Adams, & Ponder
(1984), male B W
Ingram, Marchioni, Hill,
Caraveo-Ramos, & McNeil
(1 985), male
Recidivist B
Nonrecedivist B
McCreary & Padilla (1977),
male B W B
Murphree, Karbelas, & Bryan
(1 962),d male
Panton (1959a), male B W B
Rosenblatt & Pritchard
(1978), male B W B B
Stanton (1956), male
Walters (1986), male
Control W
Index B W B

Substance Abuse Samples

Hill, Haertzen, & Glaser

(1 960), male W B
Patalano (1 978)
Female W B W W
Male B W W
Patterson, Charles, Wood¬
ward, Roberts, & Penk
(1981), male
Penk et al. (1982), male
Penk, Robinowitz, Roberts,
Dolan, & Atkins (1981),
male
Penk, Woodward, Robinowitz,
& Hess (1978), male W W
Sutker, Archer, & Attain
(1 978), female & male W W W W
 Specific Groups: Adolescents, the Aged, Blacks, and Other Ethnic Groups 343

TABLE 8-1 continued

Validity
Scales Clinical Scales

Study L F K 1 234567890

Substance Abuse Samples

Sutker, Archer, & Allain

(1 980)e
Female (NARA) W W W W W B W W W
Male (NARA)
Female (Fetter) W W W W W B W W W W
Male (Fetter) W W B W
Walters, Greene, & Jeffrey
(1 984), male
Weiss & Russakoff (1977),
female & male W W

Medical/Welfare Samples

Hokanson & Calden (1960),

male W B B B
McGill (1980), female

Psychiatric Samples

Bertelson, Marks, & May

(1982), female & male B B B
Butcher, Braswell, & Raney
(1983)
Female B B
Male
Costello et al. (1973)
Female B W B B B B
Male B
Davis (1 975), male
Schizophrenic
Nonschizophrenic
Davis, Beck, & Ryan (1973),
male
High education B B W W W W
Low education W W W W W W
Davis & Jones (1974), male
High education schizo¬
phrenic W
High education nonschizo¬
phrenic W
Low education schizo-
phrenic B B B B
Low education nonschizo-
phrenic B W W B B
Genthner & Graham (1976),
female & male B B
continued
344 Chapter 8

TABLE 8-1 continued

Validity
Scales Clinical Scales

Study L F K 1234567890

Psychiatric Samples

Kirk & Zucker (1979), male B

Klinge & Strauss (1976),
female & male
Liske & McCormick (1976),
male
Miller, Knapp, & Daniels
(1968), male B W B
Miller, Wertz, & Counts
(1961), male B W
Peteroy & Pirrello (1982)/
female & male
Smith & Graham (1981),9
female & male

Adapted from "Ethnicity and MMPI Performance: A Review" by Roger L. Greene, 1 987, Journal
of Consulting and Clinical Psychology, 55, pp. 500-502. Copyright 1 987 by the American Psy¬
chological Association. Reprinted by permission.
Note: B = Blacks scored higher than whites; W = Whites scored higher than blacks. Differences
less than 5 T points between two groups were not charted, regardless of statistical significance.
When a study reported multiple comparisons between two groups, only the results of the most
rigorous comparisons were charted.
aCompared only Scale 2.
bSocial classes 1 and 2 = independent or dependent occupations involving skill and the supervi¬
sion of others; Social class 3 = dependent occupations involving skill but little supervision of
others; Social class 4 = dependent occupations involving little skill and little supervision of others.
cCompared only Scale 4.
dCompared only Scales K and 4.
eNARA = Narcotic Addict Rehabilitation Act Program, Tulane University School of Medicine; Fet¬
ter = Franklin C. Fetter Drug Abuse Program, Charleston, SC.
'Compared only Scale F, 8, and 9.
9Compared only Scale F.

Donald & Gynther, 1962, 1963) that reported ture independent of geographic location—
the preponderance of significant black-white may be contributing to whether black-white
differences on the MMPI in normal individu¬ differences are found. Costello (1977) has de¬
als were conducted in the South, which lends veloped an index to help control for this fac¬
some credence to Harrison and Kass’s (1968) tor in studying black-white differences on the
hypothesis. Ball (1960) and Baughman and MMPI.
Dahlstrom (1968), however, also conducted In response to the finding that blacks
their research in the South, and these investi¬ usually score higher on the standard MMPI F
gators found few black-white differences. scale, Gynther, Lachar, and Dahlstrom
Thus, some factor(s) other than ethnic¬ (1978) developed a new F scale for blacks.
ity—perhaps identification with a black cul¬ They identified 33 items that were endorsed
 Specific Groups: Adolescents, the Aged, Blacks, and Other Ethnic Groups 345

by less than 10 percent of a combined normal tion of the pattern of black-white differences
male and female black sample. Only 22 of within and between populations is not readily
these 33 items (66.7 percent) are on the stan¬ apparent without giving adequate consider¬
dard F scale. ation to the multitude of moderator variables
Blacks and whites, however, responded which may affect performance.
quite similarly to the 33 items on the new F One final comment about black-white
scale for blacks; 28 of these 33 items also differences on the standard validity and clini¬
were endorsed by less than 10 percent of a cal scales seems warranted. The more rigor¬
white sample. Gynther and colleagues (1978) ously that moderator variables and profile
suggest that raw scores of 3, 7, and 11 on the validity issues are controlled by an investiga¬
new F scale for blacks should be considered tor, the less likely it becomes that black-white
equivalent to T scores of 50, 60, and 70, re¬ differences will be found. For example, Cos¬
spectively. This new/7scale for blacks can be tello and colleagues (1972) reported that no
used as a measure of the black client’s en¬ black-white differences were found if invalid
dorsement of items that are infrequently en¬ profiles were excluded. Similarly, Penk and
dorsed by black normals, in a fashion similar colleagues (Penk, Robinowitz, Dolan, Atkins,
to the standard F scale. It remains to be seen, & Woodward, 1982; Penk et al., 1981) have
however, whether black clients who elevate found no black-white differences in substance
the new F scale for blacks will demonstrate abuse patients when age, education, socio¬
the same behavioral correlates as are found in economic status, and intelligence were con¬
white clients who elevate the standard F trolled statistically.
scale. Thus, it appears that moderator vari¬
The specific population that is being ex¬ ables, such as socioeconomic status, educa¬
amined also appears to play an important in¬ tion, and intelligence, as well as profile valid¬
fluence on the nature of the black-white dif¬ ity issues, are important factors to be
ferences that will be found. Thus, in controlled in any MMPI research and partic¬
substance abuse samples, whites rather fre¬ ularly when the potential effect of ethnic sta¬
quently score higher than blacks on those tus is being examined. Rather than discussing
scales for which reliable differences have the fact that normals from an ethnic minority
been found, whereas in psychiatric samples tend to have more elevated scores on the stan¬
there does not appear to be any reliable pat¬ dard validity and clinical scales than whites,
tern to the black-white comparisons. it seems much more important to consider the
It has been conjectured (cf. Penk, role of the moderator variables listed above
Robinowitz, Roberts, Dolan, & Atkins, 1981) such as socioeconomic status, education, and
that whites who become substance abusers intelligence.
are more disturbed than blacks, which would As stated by Pritchard and Rosenblatt
account for the higher elevations in the for¬ (1980a), it seems premature to conclude that
mer group. Since groups of substance abusers new norms for blacks are needed. However,
have not been matched for type or severity of it does appear necessary for clinicians and re¬
psychopathology and then examined for searchers to become more aware of the multi¬
black-white differences, there are no data to tude of potential factors which can affect
address this issue directly. However, when MMPI performance.
psychiatric patients were matched on educa¬ The paucity of studies that have exam¬
tion and for type of psychopathology (Davis, ined directly whether there are distinct empir¬
Beck, & Ryan, 1973), no black-white differ¬ ical correlates of the obtained black-white
ences were found. differences is amazing, particularly when it is
Again, it seems that any simple explana¬ realized that this issue has been investigated
346 Chapter 8

for nearly 40 years. Investigators seem con¬ correlates of the black-white differences that
tent to report small black-white differences in were found, and two of the three studies
performance on a specific scale without con¬ (Elion & Megargee, 1975; Genthner & Gra¬
sidering whether these differences have any ham, 1976) concluded that there were no dif¬
clinical correlates or relevance. ferences in the external correlates of the
Gynther, Fowler, and Erdberg (1971) ad¬ MMPI.
ministered the MMPI to 88 rural, isolated Four studies have examined the corre¬
blacks and found that almost half (41/88) lates of specific codetypes and scale eleva¬
of these persons had raw scores greater tions (Clark & Miller, 1971; Gynther, Alt¬
than 16 (T score of 80) on the Fscale. Scale man, & Warbin, 1973a, 1973b; Strauss,
8 (Schizophrenia) was the most frequent Gynther, & Wallhermfechtel, 1974). In none
high-point scale, followed by Scale 9 (Hypo- of these four studies were the black correlates
mania) in men and Scale 6 (Paranoia) in identical to the white correlates, although
women. The mean profile for this group of Clark & Miller (1971) reported that the cardi¬
blacks would have been classified as psy¬ nal features of their 8-6 profiles were remark¬
chotic by most MMPI interpretive systems. ably similar. Shore (1976) reviewed the statis¬
Clearly, this black sample was being over- tical analysis of the data reported by Strauss
pathologized using standard MMPI interpre¬ and colleagues (1974) and concluded that
tive criteria. there was no clear evidence for greater misdi¬
In a related study Gynther, Altman, and agnosis among blacks than whites.
Warbin (1973b) were unable to replicate any Thus, at least three of these seven studies
of the correlates of raw scores greater than 25 have found different correlates for blacks
(T score of 98) on the standard F scale in and whites on the MMPI. The interested
black psychiatric patients; they were, how¬ reader should consult Pritchard and
ever, successful in replicating these correlates Rosenblatt’s (1980a, 1980b) criticisms of
in white patients. Gynther and associates’ these studies, however, before assuming that
findings suggest that the meaning of high F these correlates are reliable. Clearly, addi¬
scale scores in blacks remains unclear. tional research is needed to investigate the
In one of the few additional validity stud¬ clinical correlates of black-white differences
ies, Elion and Megargee (1975) tested the va¬ in MMPI performance when the potential in¬
lidity of Scale 4 (Psychopathic Deviate) in fluence of moderator variables have been
blacks and concluded that there was no need controlled adequately.
to derive a new Scale 4 to assess antisocial be¬ Only a few studies have investigated
havior patterns among blacks. They did sug¬ whether there are differences in the dis-
gest that 5 T points should be subtracted criminability of blacks and whites within and
from Scale 4 after AT-correction in black cli¬ between diagnostic groups using the MMPI.
ents to offset the tendency that they found Both Cowan, Watkins, and Davis (1975) and
for blacks to score slightly higher than whites Davis (1975) examined black-white differ¬
on this scale. Finally, Marks and associates ences in matched groups of schizophrenics
(1974) found few black-white differences in and nonschizophrenics and found no signifi¬
the correlates of their adolescent codetypes, cant differences for ethnicity. Cowan and as¬
although they only had a small sample of 61 sociates did find that blacks with less than 12
black adolescents. years of education were more likely to be mis-
Only three studies (Elion & Megargee, classified than blacks with 12 years or more
1975; Genthner & Graham, 1976; Smith & of education or whites. Their results appear
Graham, 1981) have examined the empirical to be compatible with Rosenblatt and Pritch-
 Specific Groups: Adolescents, the Aged, Blacks, and Other Ethnic Groups 347

ard (1978), who found race differences only Witt, 1976; Harrison & Kass, 1967, 1968;
in patients with IQs below 94. Jones, 1978; Miller, Knapp, & Daniels, 1968;
Strauss and colleagues (1974) examined Witt & Gynther, 1975). Although from 58 to
black-white differences between groups of 213 items have been found to differentiate
patients diagnosed as exhibiting behavior dis¬ blacks from whites in a given study, there has
orders and psychoses. Using a discriminant been limited overlap among these items
function analysis, they reported that black across the various studies. In reviewing six
patients with behavior disorders were mis- studies of item-level differences among
classified more than any of the other groups blacks and whites, Costello (1977) reported
of patients. But Shore (1976) indicated that that “three items differentiated [blacks and
this result reflected a computational error whites] six of six times, seven items discrimi¬
and concluded that there were no significant nated five of six times, and 22 items discrimi¬
black-white differences in the Strauss and nated four of six times’’ (p. 515).
colleagues’ (1974) study. Again, the need to control for the poten¬
Because of the limited amount of re¬ tial effect of moderator variables must be un¬
search, it is difficult to conclude whether derscored in these studies. Consequently,
there are black-white differences between there does not appear to be any real conclu¬
various diagnostic groups as assessed by the sion that can be drawn from these findings
MMPI. There does appear to be a tendency other than that item-level endorsement fre¬
for blacks with limited education and IQs quencies may vary in different samples with a
below average to score differently than small number of items consistently differenti¬
whites on the MMPI, and the clinician should ating blacks and whites.
consider this when interpreting a black Harrison and Kass (1967) examined the
client’s MMPI. individual validity and clinical scales to deter¬
There has been very little research on mine which ethnic group was more likely to
black-white differences for the numerous endorse the items, and concluded that in
MMPI supplementary scales (see Table 8-2). many cases ethnic differences may cancel out
The limited research appears to show that because there are approximately the same
few reliable black-white differences are number of items favoring each ethnic group
found with fewer significant results than on on many of the scales. They discovered more
the standard validity and clinical scales. Even items that whites were likely to endorse only
given the fact that moderator variables gener¬ on Scale 3 and found more items that blacks
ally were not controlled in most of these stud¬ were likely to endorse on Scales F, /, 6, 8, and
ies, few reliable differences are found. It is in¬ 9. These findings are somewhat consistent
teresting to note that black and white with the results presented in Table 8-1, which
substance abusers seem to score in the same reveals a tendency for blacks to score higher
range on the MacAndrew Alcoholism scale on Scales F, 1, 8, and 9; whites, however,
(MacAndrew, 1965), yet normal blacks score tend to score higher on Scale 6 as well as on
higher than whites. Since the MacAndrew Al¬ Scale 3.
coholism scale is used so widely, more re¬ Harrison and Kass (1967) conducted a
search on potential ethnic differences, partic¬ factor analysis of the 150 items that most reli¬
ularly in interaction with the type of setting, ably distinguished blacks from whites and
is needed. identified 20 factors, 16 of which could be
A number of investigators have reported conceptually labeled. The first 8 of these fac¬
black-white differences at the item level on tors are estrangement, intellectual and cul¬
the MMPI (Costello, 1973, 1977; Gynther & tural interests, denial of major symptoms,
TABLE 8-2 Black-White Differences on MMPI Supplementary Scales

Study Gender Population Result

Dependency scale (Navran, 1954)

Ingram, Marchioni, Hill, Caraveo-Ramos, & McNeil (1985) M Prison
King, Carroll, & Fuller (1977) M Normal
Patalano (1978) M & F Substance abuse
Dominance scale (Gough, McClosky, & Meehl, 1951)
Ingram et al. (1985)
Nonrecidivist M Prison W
Redicivist M Prison
King et al. (1977) M Normal W
Patalano (1978) M & F Substance abuse
MacAndrew Alcoholism scale (MacAndrew, 1965)
McCreary & Padilla (1977) M Prison
Snyder, Kline, & Podany (1985) M Substance abuse
Snyder et al. (1 985) F Substance abuse
Walters, Greene, Jeffrey, Kruzich, & Haskin (1983) M Normal B
Walters et al. (1 983) M Substance abuse
Zager & Megargee (1981) M Prison/substance abuse
Overcontrolled Hostility scale (Megargee, Cook, & Mendelsohn, 1967)
McCreary & Padilla (1 977) M Prison
Welsh Anxiety Scale (Welsh, 1965)
Butcher, Ball, & Ray (1964) F Normal
Butcher et al. (1 964) M Normal W
Hokanson & Calden (1960) M Medical
King et al. (1977) M Normal
Miller, Knapp, & Daniels (1968) M Psychiatric
Miller, Wertz, & Counts (1961) M Psychiatric
Sutker, Archer, & Allain (1978) M Substance abuse
Sutker, Archer, & Allain (1980) M & F Substance abuse
Welsh Repression scale (Welsh, 1965)
Butcher et al. (1 964) F Normal
Butcher et al. (1 964) M Normal
King et al. (1977) M Normal
Miller et al. (1961) M Psychiatric
Miller et al. (1968) M Psychiatric
Sutker et al. (1978) M Substance abuse
Sutker et al. (1 980)a
Fetter F Substance abuse B
Fetter M Substance abuse
NARA F Substance abuse B
NARA M Substance abuse
Wiggins Content scales (Wiggins, 1 966)b
Dolan, Roberts, Penk, Robinowitz, & Atkins (1983) M Substance abuse \N(FAM)
B (FEM)
Penk, Woodward, Robinowitz, & Hess (1978) M Substance abuse \N(FAM)
B (FEM)
B (PHO)
\N(SOC)

Adapted from "Ethnicity and MMPI Performance: A Review" by Roger L. Greene, 1987, Journal of
Consulting and Clinical Psychology, 55, pp. 504-505. Copyright 1 987 by the American Psychological
Association. Reprinted by permission.
Note: B = Blacks scored higher than whites; W — Whites scored higher than blacks. Differences less
than 5 T points between two groups were not charted, regardless of statistical significance. When a
study reported multiple comparisons between two groups, only the results of the most rigorous compari¬
son were charted.
aFetter = Franklin C. Fetter Drug Abuse Program, Charleston, SC; NARA = Narcotic Addict Rehabilita¬
tion Act Program, Tulane University School of Medicine.
bFAM - Family Problems; FEM = Femininity; PHO = Phobias; SOC = Social Maladjustment.

348
 Specific Groups: Adolescents, the Aged, Blacks, and Other Ethnic Groups 349

cynicism, admission of minor faults, roman¬ to determine which interpretation of black-

tic interests, somatic tension, and impulse- white differences at the item level is more ap¬
ridden fantasy. Blacks achieved significantly propriate. In the interim, clinicians should be
higher scores on all 8 of these factors except cautious of interpreting blacks’ perfor¬
denial of major symptoms, on which whites mances on Harrison and Kass’ (1967) factors
achieved significantly higher scores. as reflecting either psychopathology or dis¬
Baughman and Dahlstrom (1968) exam¬ trust of society. Jones’ (1978) investigation
ined the performance of their sample of black suggests that other factors, such as the type
and white eighth-graders on the first 5 of of items found in the MMPI, may be contrib¬
these factors. The black eighth-graders uting to these results.
achieved significantly higher scores than
white eighth-graders on 3 of the 5 factors: es¬
Hispanic-White Comparisons
trangement, intellectual and cultural inter¬
ests, and cynicism. The white male eighth- Table 8-3 summarizes the literature that has
graders achieved higher scores on denial of examined Hispanic-white differences on the
major symptoms than did white female or standard validity and clinical scales of the
black eighth-graders. The similarity of the re¬ MMPI. It should be apparent that there has
sults between Baughman and Dahlstrom been very little research in this area, and con¬
(1968) and Harrison and Kass (1967) on these sequently, conclusions are even more diffi¬
factors is remarkable considering the differ¬ cult to make than in black samples. There
ences in sample composition and geographic does not appear to be any pattern to these
location. comparisons and there are few data to sup¬
Gynther (1972) suggested that the com¬ port the contention that Hispanics frequently
mon theme in both of these studies is the score higher on the L scale and lower on Scale
marked distrust of society exhibited by blacks 5 (Masculinity-Femininity; cf. Greene, 1980),
rather than any indication of psychopathol¬ although when differences are found they
ogy. Jones (1978) has contended that this tend to be in that direction. There does appear
conceptualization of blacks’ performance on to be a tendency for fewer Hispanic-white dif¬
the MMPI may reflect the types of items ferences to be found than when black-white
found in the MMPI rather than blacks’ comparisons are made (see Table 8-1).
marked distrust of society. Velasquez and colleagues (Velasquez,
In support of this contention, Jones ex¬ 1984; Velasquez & Callahan, 1990a, 1990b;
amined black-white differences in college stu¬ Velasquez, Callahan, & Carrillo, 1989;
dents on 361 items, which included items Velasquez & Gimenez, 1987) have provided
from the California Psychological Inventory the most systematic research on the use of the
(Gough, 1957) that are less pathologic in con¬ MMPI in Hispanics. There is not a simple
tent than many MMPI items. He found that generalization that summarizes their re¬
288 (80 percent) of these items significantly search, although their results tend to be sim¬
differentiated blacks and whites. Instead of ilar to what has been reported by others.
the alienated and distrustful conceptualiza¬ Velasquez and Gimenez (1987) did not find
tion of blacks’ performance on the MMPI as any relationship between DSM-III diagnosis
proposed by Gynther (1972), Jones found his and the standard validity and clinical scales
item differences to suggest that young blacks on the MMPI in Hispanic inpatients; this fail¬
were assertive, poised, outspoken, tough- ure to find a specific relationship between the
minded, and so on. MMPI and DSM-III diagnoses has been re¬
Further research is obviously necessary ported frequently (cf. Greene, 1988).
TABLE 8-3 Hispanic-White Differences on MMPI Standard Validity and Clinical Scales

Validity
Scales Clinical Scales

Study L F K 1 234567890

Normal Samples

Montgomery & Orozco

(1985), female & male H H
Padilla, Olmedo, & Loya
(1982), female & male3
N-L factor H H W
SES factor W W H
Reilley & Knight (1970), fe-
male & male

Psychiatric Samples

Hibbs, Kobos, & Gonzalez

(1979)
Female H H
Male W
Lawson, Kahn, & Heiman
(1982), female & male H
Plemons (1977), female &
male H H W W W W
Velasquez & Callahan (1990b) H H H H H H H

Prison Samples

Holland (1 979), male W

McCreary & Padilla (1977),
male H

Substance Abuse Samples

Page & Bozlee (1982), male H

Penk, Robinowitz, Roberts,
Dolan, & Atkins (1981),
male W
Velasquez & Callahan (1990a) W W W W W W W

Welfare Sample

McGill (1980), female H W

Adapted from "Ethnicity and MMPI Performance: A Review" by Roger L. Greene, 1987, Journal
of Consulting and Clinical Psychology, 55, p. 506. Copyright 1 987 by the American Psychologi¬
cal Association. Reprinted by permission.
Note: H = Hispanics scored higher than whites; W = Whites scored higher than Hispanics. Differ¬
ences less than 5 T points between two groups were not charted, regardless of statistical signifi¬
cance. When a study reported multiple comparisons between two groups, only the results of the
most rigorous comparison were charted.
aN-L factor - Normality-Language factor; SES factor = socioeconomic status.

350
 Specific Groups: Adolescents, the Aged, Blacks, and Other Ethnic Groups 351

Velasquez and Callahan (1990a) found blacks, whereas McCreary and Padilla’s
that Hispanic alcoholics had lower scores on (1977) Hispanic prisoners scored higher on
the MMPI than white alcoholics similar to Scales K, 1 (Hypochondriasis), 2 (Depres¬
the relationship found in black-white com¬ sion), and 3 (Hysteria). McGill (1980) found
parisons in alcoholics (see Table 8-1). This that Hispanic women welfare recipients
latter finding that Hispanic alcoholics, like scored higher on Scales L and K and lower on
black alcoholics, have lower scores on the Scale 9 (Hypomania) than blacks. Penk and
MMPI when differences are found requires colleagues (1981) reported that Hispanic male
that any study of ethnic differences on the substance abusers scored higher on Scale 2
MMPI include the setting in which the test is and lower on Scale 5 than blacks. Finally,
administered as a potential confounding fac¬ Velasquez and Callahan (1990b) did not find
tor. Finally, it must be noted that Velasquez, any differences between black and Hispanic
like many other investigators in this area of schizophrenics. It should be evident that
research, does not report how ethnicity is de¬ there is no clear pattern to be found in these
fined in any of his studies. black-Hispanic comparisons.
There have been only three studies McCreary and Padilla (1977) are the
(Dolan, Roberts, Penk, Robinowitz, & only investigators to report black-Hispanic
Atkins, 1983; McCreary & Padilla, 1977; comparisons on any MMPI supplementary
Page & Bozlee, 1982) that have reported scales. They found that Hispanic prisoners
Hispanic-white differences on any supple¬ scored lower on the MacAndrew Alcoholism
mentary scales, with almost no overlap in the scale (MacAndrew, 1965) than blacks.
scales reported. McCreary and Padilla (1977)
found no differences between Hispanic and
Indian-White Comparisons
white prisoners, and Page and Bozlee (1982)
found no differences between Hispanic and Table 8-4 summarizes the literature that has
white substance abusers on the Mac Andrew examined Indian-white differences on the
Alcoholism scale (MacAndrew, 1965). Mc¬ standard validity and clinical scales of the
Creary and Padilla reported that Hispanics MMPI. For the first time in the comparison
scored higher on the Overcontrolled-Hostil- of the MMPI performance of two ethnic
ity scale (Megargee, Cook, & Mendelsohn, groups, there appears to be a clear pattern for
1967), and Dolan and associates (1983) found normal Indians to score higher on most of the
no differences larger than 5 T points on any clinical scales than their white counterparts
of the Wiggins (1966) Content scales. even though there are no differences on the
There have been no studies of Hispanic- validity scales. This conclusion must be tem¬
white differences at the item level. pered by the fact that only two studies have
been conducted, especially since similar
trends are not evident in psychiatric and sub¬
Black-Hispanic Comparisons
stance abuse samples. In these latter two sam¬
Only five studies have investigated black- ples, much as in black-white and Hispanic-
Hispanic differences on the standard validity white comparisons, there is no consistent pat¬
and clinical scales (Holland, 1979; McCreary tern at all.
& Padilla, 1977; McGill, 1980; Penk et al., Only one study (Uecker, Boutilier, &
1981; Velasquez & Callahan, 1990b). Richardson, 1980) has investigated the per¬
Holland’s (1979) sample of Hispanic formance of Indian samples on any of the
prisoners scored lower on Scales 5 (Masculin¬ MMPI supplementary scales. Uecker and
ity-Femininity) and 8 (Schizophrenia) than colleagues found no differences between In-
352 Chapter 8

TABLE 8-4 Indian-White Differences on MMPI Standard Validity and Clinical Scales

Validity
Scales Clinical Scales

Study L F K 1 2 3 4 5 6 7 8 9 0

Normal Samples

Arthur (1944)a
Female 1 1 1
Male 1
Herreid & Herreid
(1966)
Female native 1 1 1 1 1 1
Male native 1 1 1
Female nonnative 1
Male nonnative 1

Psychiatric Samples

Butcher, Braswell, &

Raney (1 983)
Female W W
Male W W
Pollack & Shore
(1 980), female &
maleb 1 1 1 II II 1 1 1

Substance Abuse Samples

Kline, Rozynko, Flint,

& Roberts (1 973),
male
Page & Bozlee
(1982), male
Uecker, Boutilier, &
Richardson
(1980), male W W

Adapted from "Ethnicity and MMPI Performance: A Review" by Roger L. Greene, 1 987, Journal
of Consulting and Clinical Psychology, 55, p. 508. Copyright 1 987 by the American Psychologi¬
cal Association. Reprinted by permission.
Note: I = Indian scored higher than whites; W = Whites scored higher than Indians. Differences
less than 5 T points between two groups were not charted, regardless of statistical significance.
When a study reported multiple comparisons between two groups, only the results of the most
rigorous comparison were charted.
aReported data only for Scales 1-4.
bCompared their Indian psychiatric patients with the original Minnesota normative sample, which
produced the larger number of significant results.
 Specific Groups: Adolescents, the Aged, Blacks, and Other Ethnic Groups 353

dian and white substance abusers on the Mac- me,” because he worried about them even
Andrew Alcoholism scale (MacAndrew, though he was not married and had no chil¬
1965). Uecker and colleagues did not report dren.
data from an Indian sample who were not
substance abusers, so it is not known whether
Asian-American-White Comparisons
they also would elevate the MacAndrew Al¬
coholism scale as seen in normal blacks as Table 8-5 summarizes the literature that has
mentioned earlier. examined Asian-American-white differences
There have been no studies of Indian- on the standard validity and clinical scales.
white differences at the item level, although As can be readily seen in Table 8-5, there
Arthur (1944) discussed the various interpre¬ have been almost no studies of the perfor¬
tations that her Indian students gave to some mance of Asian-Americans on the MMPI.
of the items. For example, one Indian student Since Sue and Sue (1974) found a number of
answered “true” to the items, “I worry differences between their samples, it appears
about my wife,” and “My children worry clear that more research is need in this ethnic

TABLE 8-5 Asian-American-White Differences on MMPI Standard Validity and Clinical Scales

Validity
Scales Clinical Scales

Study L F K 7 2 3 4 5 6 7 8 9 0

Normal Samples

Marsella, Sanborn,
Kameoka, Shizuru,
& Brennan (1 975)a
Chinese female A
Chinese male
Japanese female
Japanese male A

Psychiatric Samples

Sue & Sue (1974)

Female A A A
Male AA AA A AAA A
Tsushima & Onorato
(1982)
Female
Male W

Adapted from "Ethnicity and MMPI Performance: A Review" by Roger L. Greene, 1987, Journal
of Consulting and Clinical Psychology, 55, p. 509. Copyright 1 987 by the American Psychologi¬
cal Association.
Note: A = Asian-Americans scored higher than whites; W = Whites scored higher than Asian-
Americans. Differences less than 5 T points between two groups were not charted, regardless of
statistical significance. When a study reported multiple comparisons between two groups, only
the results of the most rigorous comparison were charted.
aCompared only Scale 2.
354 Chapter 8

group. It is particularly important for the of the effect of ethnic status on MMPI per¬
study by Sue and Sue to be replicated, since in formance need to give more serious consider¬
the other ethnic groups that were reviewed ation to the role of cultural factors and stop
above, there was a tendency toward fewer assuming that identity with the minority cul¬
differences between groups in psychiatric ture is defined by the person’s race or sur¬
samples than in normal samples. name. Some reliable and valid means of as¬
There have been no studies which have sessing persons’ degree of identification with
examined Asian-American-white perfor¬ their culture would greatly enhance research
mance on any of the MMPI supplementary on this topic.
scales. There also have been no studies of the Third, researchers need to stop reporting
differences in the pattern of item endorse¬ small mean differences between two groups,
ments between Asian-Americans and whites. which are frequently poorly defined, and
begin to assess the empirical correlates of any
differences that are found. The fact that
Summary
fewer than 10 studies actually have examined
After reviewing the MMPI performance of the empirical correlates of ethnic status and
various ethnic groups, there appear to be a MMPI performance should indicate the di¬
number of comments that can be made about rection for future research.
the research being conducted in this area. Finally, there is a real dearth of studies
First, the failure to find any consistent pat¬ on the multitude of MMPI supplementary
tern of scale differences between any two eth¬ scales in any ethnic group. Since a number of
nic groups in any population would suggest these supplementary scales are scored rou¬
that it is very premature to begin to develop tinely, research is needed to examine the ef¬
new norms for various ethnic groups. It ap¬ fect of ethnic status on performance on these
pears that moderator variables, such as socio¬ scales. It should go without saying that the
economic status, education, and intelligence, above stated guidelines for research on the
as well as profile validity, are more important standard validity and clinical scales also
determinants of MMPI performance than apply to research on these supplementary
ethnic status. Definitely, research is needed scales.
that examines the role of identified cultural
factors on MMPI performance when appro¬ ENDNOTES
priate controls are instituted for the multi¬ 1. This section has drawn substantially on
tude of factors which can affect the results. Greene’s (1987) review of ethnicity and MMPI
Second, the frequent failure to assess in performance.
any manner the persons’ identification with 2. A multitude of terms have been used to
their ethnic group is quite notable. It seems describe persons within a specific ethnic group
very questionable to be making statements such as Chicano, Hispanic, Mexican, Mexican-
about the effects of ethnic status on MMPI American, and so on. In order to provide consis¬
performance without some means of insuring tency throughout this chapter, the terms
Asian-American, Hispanic, and Indian will be
that the persons actually identify with and be¬
used.
long to the ethnic group. Thus, investigations
Appendixes

A: MMPI-2: Item Composition of Validity, Clinical, and

Supplementary Scales

B: MMPI-2: T Score Conversion Tables

C: MMPI-2: Critical Items

D: Prototypic Scores for Specific Codetypes in Psychiatric

Settings

E: Item Overlap among MMPI-2 Scales

F: MMPI-2 to MMPI Conversion Tables

355
 APPENDIX A

MMPI-2: Item Composition

of Validity, Clinical, and
Supplementary Scales

TABLE A-1 Basic Scales

L — Lie (15 items)

True
none

False
16 29 41 51 77 93 102 107 123 139 153 183 203 232 260
Males: Mean 3.53; S.D. 2.28. Females: Mean 3.56; S.D. 2 08.

F - Infrequency (60 items)

True
18 24 30 36 42 48 54 60 66 72 84 96 114 138
144 150 156 162 168 180 198 216 228 234 240 246 252 258
264 270 282 288 294 300 306 312 324 336 349 355 361

False
6 12 78 90 102 108 120 126 132 174 186 192 204 210
222 276 318 330 343
Males: Mean 4.53; S.D. 3.24. Females: Mean 3.66; S.D. 2.91

continued

Source: From the Minnesota Multiphasic Personality Inventory-2 Manual by J. N. Butcher, W. G.

Dahlstrom, J. R. Graham, A. Tellegen, & B. Kaemmer, Appendix B, pp. 65-84. Copyright © the
University of Minnesota 1 942, 1 943, 1 951, 1 967 (renewal 1 970), 1 989. This Manual, 1 989.
Reproduced by permission.
357
358 Appendix A

TABLE A-1 continued

K - Correction (30 items)

True
83

False
29 37 58 76 110 116 122 127 130 136 148 157 158 167
171 196 213 243 267 284 290 330 338 339 341 346 348 356
365
Males: Mean 15.30; S.D. 4.76. Females: Mean 15.03; S.D. 4.58.

1 Hs-- Hypochondriasis (32 items)

True
18 28 39 53 59 97 101 111 149 175 247

False
2 3 8 10 20 45 47 57 91 117 141 143 152 164
173 176 179 208 224 249 255
Raw scores without K:
Males: Mean 4.92; S.D. 3.87. Females: Mean 5.93; S.D. 4.51.
Raw scores with K:
Males: Mean 12.78; S.D. 3.86. Females: Mean 13.69; S.D . 4.05.

2D- Depression (57 items)

True
5 15 18 31 38 39 46 56 73 92 117 127 130 146
147 170 175 181 215 233

False
2 9 10 20 29 33 37 43 45 49 55 68 75 76
95 109 118 134 140 141 142 143 148 165 178 188 189 212
221 223 226 238 245 248 260 267 330
Males: Mean 18.32; S.D 4.59. Females: Mean 20.14 ; S.D. 4.97.
 MMPI-2: Item Composition of Validity, Clinical, and Supplementary Scales 359

TABLE A-1 continued

3 Hy - Conversion Hysteria (60 items)

True
11 18 31 39 40 44 65 101 166 172 175 218 230

False
2 3 7 8 9 10 14 26 29 45 47 58 76 81
91 95 98 110 115 116 124 125 129 135 141 148 151 152
157 159 161 164 167 173 176 179 185 193 208 213 224 241
243 249 253 263 265
Males: Mean 20.87; S.D . 4.73. Females: Mean 22. 08; S.D. 4.72.

4 Pd - Psychopathic Deviate (50 items)

True
17 21 22 31 32 35 42 52 54 56 71 82 89 94
99 105 113 195 202 219 225 259 264 288

False
9 12 34 70 79 83 95 122 125 129 143 157 158
160 167 171 185 209 214 217 226 243 261 263 266 267
Raw scores without K:
Males: Mean 16.57; S.D. 4.60. Females: Mean 16.21; S.D. 4.65.
Raw scores with K:
Males. Mean 22.65; S.D. 4.67. Females: Mean 22.21; S.D. 4.53.

5 Mf-m - Masculinity-Femininity (Masculine) (56 items)

True
4 25 62 64 67 74 80 112 119 122 128 137 166 177
187 191 196 205 209 219 236 251 256 268 271

False
1 19 26 27 63 68 69 76 86 103 104 107 120 121
132 133 163 184 193 194 197 199 201 207 231 235 237 239
254 257 272
Males: Mean 26.01; S.D. 5.08.

continued
360 Appendix A

TABLE A-1 continued

5 Mf-f - Masculinity-Femininity (Feminine) (56 items)

True
4 25 62 64 67 74 80 112 119 121 122 128 137 177
187 191 196 205 219 236 251 256 271

False
1 19 26 27 63 68 69 76 86 103 104 107 120 132
133 163 166 184 193 194 197 199 201 207 209 231 235 237
239 254 257 268 272
Females: Mean 35.94; S.D. 4.08.

6 Pa - Paranoia (40 items)

True
16 17 22 23 24 42 99 113 138 144 145 146 162 234
259 271 277 285 305 307 333 334 336 355 361

False
81 95 98 100 104 110 244 255 266 283 284 286 297 314
315
Males: Mean 10.10; S.D. 2.87. Females: Mean 10.23; S.D. 2.97.

7 Pt - Psychasthenia (48 items)

True
11 16 23 31 38 56 65 73 82 89 94 130 147 170
175 196 218 242 273 275 277 285 289 301 302 304 308 309
310 313 316 317 320 325 326 327 328 329 331

False
3 9 33 109 140 165 174 293 321
Raw scores without K:
Males: Mean 11.24; S.D. 6.61. Females: Mean 12.69; S.D. 7.19.
Raw scores with K:
Males: Mean 26.43; S.D. 5.00 Females: Mean 27.70; S.D. 5.10.
 MMPI-2: Item Composition of Validity, Clinical, and Supplementary Scales 361

TABLE A-1 continued

8 Sc - Schizophrenia (78 items)

True
16 17 21 22 23 31 32 35 38 42 44 46 48 65
85 92 138 145 147 166 168 170 180 182 190 218 221 229
233 234 242 247 252 256 268 273 274 277 279 281 287 291
292 296 298 299 303 307 311 316 319 320 322 323 325 329
332 333 355

False
6 9 12 34 90 91 106 165 177 179 192 210 255 276
278 280 290 295 343
Raw scores without K:
Males: Mean 11.20; S.D. 7.12. Females: Mean 11.24; S.D. 7.57.
Raw scores with K:
Males: Mean 26.40; S.D 5.92. Females: Mean 26.25; S.D. 5.97.

9 Ma - Hypomania (46 items)

True
13 15 21 23 50 55 61 85 87 98 113 122 131 145
155 168 169 182 190 200 205 206 211 212 218 220 227 229
238 242 244 248 250 253 269

False
88 93 100 106 107 136 154 158 167 243 263
Raw scores without K:
Males: Mean 16.88; S.D. 4.51. Females: Mean 16.07; S.D. 4.50.
Raw scores with K:
Males: Mean 19.93; S.D. 4.29. Females: Mean 19.09; S.D. 4.26.

0 Si - Social Introversion (69 items)

True
31 56 70 100 104 110 127 135 158 161 167 185 215 243
251 265 275 284 289 296 302 308 326 337 338 347 348 351
352 357 364 367 368 369

False
25 32 49 79 86 106 112 131 181 189 207 209 231 237
255 262 267 280 321 328 335 340 342 344 345 350 353 354
358 359 360 362 363 366 370
Males; Mean 25.86; S.D. 8.57. Females: Mean 27.98; S.D. 9.18.
362 Appendix A

TABLE A-2 Harris-Lingoes Subscales

D, - Subjective Depression (32 items)

True
31 38 39 46 56 73 92 127 130 146 147 170 175 215
233

False
2 9 43 49 75 95 109 118 140 148 178 188 189 223
260 267 330
Males: Mean 6.86; S.D. 3.79. Females: Mean 7.65; S.D. 4.21.

D2 - Psychomotor Retardation (14 items)

True
38 46 170 233

False
9 29 37 49 55 76 134 188 189 212
Males: Mean 5.34; S.D 1.82. Females: Mean 5.74; S.D. 1.83.

D;, - Physical Malfunctioning (11 items)

True
18 117 175 181

False
2 20 45 141 142 143 148
Males: Mean 2.89; S.D. 1.23. Females: Mean 3.22; S.D 1.36.

D4 - Mental Dullness (15 items)

True
15 31 38 73 92 147 170 233

False
9 10 43 75 109 165 188
Males: Mean 2.42; S.D. 2.09. Females: Mean 2.55; S.D. 2.21.
 MMPI-2: Item Composition of Validity, Clinical, and Supplementary Scales 363

TABLE A-2 continued

D5 - Brooding (10 items)

True
38 56 92 127 130 146 170 215

False
75 95
Males: Mean 1.83; S.D 1.77. Females: Mean 2.50; S.D. 1.95.

Hy., - Denial of Social Anxiety (6 items)

True
none

False
129 161 167 185 243 265
Males: Mean 3.89; S.D. 1.86. Females: Mean 3.85; S.D. 1.87.

Hy2- Need for Affection (12 items)

True
230

False
26 58 76 81 98 110 124 151 213 241 263
Males : Mean 6.69; S.D 2.58. Females: Mean 6.88; S.D. 2.44

Hy3 - Lassitude-Malaise (15 items)

True
31 39 65 175 218

False
2 3 9 10 45 95 125 141 148 152
Males : Mean 2.55; S.D 2.20. Females: Mean 2.73; S.D. 2.48

continued
364 Appendix A

TABLE A-2 continued

Hy4 - Somatic Complaints (17 items)

True
11 18 40 44 101 172

False
8 47 91 159 164 173 176 179 208 224 249
Males: Mean 2.50; S.D. 2.06. Females: Mean 3.22; S.D. 2.53.

Hy5 - Inhibition of Aggression (7 items)

True
none

False
7 14 29 115 116 135 157
Males: Mean 3.29; S.D. 1.32. Females: Mean 3.46; S.D. 1.29.

Pd! - Familial Discord (9 items)

True
21 54 195 202 288

False
83 125 214 217
Males: Mean 1.78; S.D. 1.52. Females: Mean 2.05; S.D. 1.65.

Pd2 - Authority Problems (8 items)

True
35 105

False
34 70 129 160 263 266
Males: Mean 3.29; S.D. 1.52. Females: Mean 2.35; S.D. 1.34.
 MMPI-2: Item Composition of Validity, Clinical, and Supplementary Scales 365

TABLE A-2 continued

Pd3 - Social Imperturbability (6 items)

True
none

False
70 129 158 167 185 243
Males: Mean 3.64; S.D. 1.71. Females: Mean 3.37; S.D. 1.73.

Pd4 - Social Alienation (13 items)

True
17 22 42 56 82 99 113 219 225 259

False
12 129 157
Males: Mean 3.74; S.D. 1.89. Females: Mean 3.98; S.D. 1.93.

Pd5 - Self-Alienation (12 items)

True
31 32 52 56 71 82 89 94 113 264

False
9 95
Males: Mean 3.39; S.D. 2.08 Females: Mean 3.35; S.D. 2.07.

Pa1 - Persecutory Ideas (17 items)

True
17 22 42 99 113 138 144 145 162 234 259 305 333 336
355 361

False
314
Males: Mean 1.74; S.D. 1.66. Females: Mean 1.79; S.D. 1.67.

continued
366 Appendix A

TABLE A-2 continued

Pa2 - Poignancy (9 items)

True
22 146 271 277 285 307 334

False
100 244
Males: Mean 2.36; S.D. 1.43. Females. Mean 2.57; S.D. 1.58.

Pa3 - Naivete (9 items)

True
16

False
81 98 104 110 283 284 286 315
Males: Mean 4.84; S.D. 2.09. Females: Mean 4.95; S.D. 2.09.

Sc, - Social Alienation (21 items)

True
17 21 22 42 46 138 145 190 221 256 277 281 291 292
320 333

False
90 276 278 280 343
Males: Mean 2.72; S.D. 2.42. Females: Mean 3.11; S.D. 2.59.

Sc2 - Emotional Alienation (11 items)

True
65 92 234 273 303 323 329 332

False
9 210 290
Males: Mean 1.05; S.D. 1.04. Females: Mean 1.12; S.D. 1.09.
 MMPI-2: Item Composition of Validity, Clinical, and Supplementary Scales 367

TABLE A-2 continued

Sc3 - Lack of Ego Mastery, Cognitive (10 items)

True
31 32 147 170 180 299 311 316 325

False
165
Males: Mean 1.31; S.D. 1.66. Females: Mean 1.18; S.D. 1.62.
CO

Lack of Ego Mastery, Conative (14 items)

o
1

True
31 38 48 65 92 233 234 273 299 303 325

False
9 210 290
Males : Mean 2.13; S.D. 1.85 Females: Mean 2.17; S.D. 1.95.

Sc5 - Lack of Ego Mastery, Defective Inhibition (11 items)

True
23 85 168 182 218 242 274 320 322 329 355

False
none
Males: Mean 1.42; S.D. 1.43. Females: Mean 1.57; S.D. 1.57.

Sc6 - Bizarre Sensory Experiences (20 items)

True
23 32 44 168 182 229 247 252 296 298 307 311 319 355

False
91 106 177 179 255 295
Males: Mean 1 90; S.D. 2.04. Females: Mean 2.07; S.D. 2.20.

continued
368 Appendix A

TABLE A-2 continued

Ma, - Amorality (6 items)

True
131 227 248 250 269

False
263
Males: Mean 1.97; S.D. 1.29. Females: Mean 1.56; S.D. 1.20.

Ma2 - Psychomotor Acceleration (11 items)

True
15 85 87 122 169 206 218 242 244

False
100 106
Males: Mean 5.29; S.D. 2.07. Females: Mean 5.07; S.D. 1.99.

Ma3 - Imperturbability (8 items)

True
155 200 220

False
93 136 158 167 243
Males: Mean 3.51; S.D. 1.66. Females: Mean 3.06; S.D. 1.56.

Ma4 - Ego Inflation (9 items)

True
13 50 55 61 98 145 190 211 212

False
none
Males: Mean 3.04; S.D. 1.53. Females: Mean 3.09; S.D. 1.64.
 MMPI-2: Item Composition of Validity, Clinical, and Supplementary Scales 369

TABLE A-3 Wiener-Harmon Subtle-Obvious Subscales

D-0 - Depression, Obvious (39 items)

True
15 18 31 38 39 46 56 73 92 127 130 146 147 170
175 215 233

False
2 9 10 20 33 43 45 49 75 95 109 118 140 141
142 165 188 223 245 248 260 330
Males: Mean 7.72; S.D. 4.29. Females: Mean 8.75; S.D. 4.88.

D-S - Depression, Subtle (18 items)

True
5 117 181

False
29 37 55 68 76 134 143 148 178 189 212 221 226 238
267
Males: Mean 10.60; S.D. 2.61. Females: Mean 11.39; S.D. 2.44.

Hy-0 - Hysteria, Obvious (32 items)

True
11 18 31 39 40 44 65 101 166 172 175 218

False
2 3 8 9 10 45 47 91 95 115 125 141 152 159
164 173 179 208 224 249
Males: Mean 4.78; S.D 3.61. Females: Mean 5.67; S.D. 4.21.

Hy-S - Hysteria, Subtle (28 items)

True
230

False
7 14 26 29 58 76 81 98 110 116 124 129 135 148
151 157 161 167 176 185 193 213 241 243 253 263 265
Males: Mean 16.09; S.D 4.31 Females: Mean 16.41; S.D. 4.09.

continued
370 Appendix A

TABLE A-3 continued

Pd-0 - Psychopathic Deviate, Obvious (28 items)

True
17 22 31 32 35 42 52 54 56 71 82 94 99 105
195 202 225 259 264 288

False
9 12 34 79 95 125 261 266
Males: Mean 6.16; S.D. 3.74. Females: Mean 5.40; S.D. 3.72.

Pd-S - Psychopathic Deviate, Subtle (22 items)

True
21 89 113 219

False
70 83 122 129 143 157 158 160 167 171 185 209 214 217
226 243 263 267
Males: Mean 10.41; S.D. 2.58. Females: Mean 10.80; S.D. 2.48.

Pa-0 - Paranoia, Obvious (23 items)

True
17 22 23 24 42 99 138 144 146 162 234 259 277 285
305 307 333 336 355 361

False
255 266 314
Males: Mean 2.55; S.D. 2.19. Females: Mean 2.76; S.D. 2.33.
 MMPI-2: Item Composition of Validity, Clinical, and Supplementary Scales 371

TABLE A-3 continued

Pa-S - Paranoia, Subtle (17 items)

True
16 113 145 271 334

False
81 95 98 100 104 110 244 283 284 286 297 315
Males: Mean 7.55; S.D. 2.36 Females: Mean 7.47; S.D. 2.34.

Ma-0 - Hypomama, Obvious (23 items)

True
15 23 50 61 85 87 145 155 168 182 190 205 218 227
229 238 242 250 253 269

False
100 106 107
Males: Mean 6.84; S.D. 2.95. Females: Mean 6.38; S.D 2.97.

Ma-S - Flypomania, Subtle (23 items)

True
13 21 55 98 113 122 131 169 200 206 211 212 220 244
248

False
88 93 136 154 158 167 243 263
Males: Mean 10.04; S.D. 2.68. Females: Mean 9.69; S.D 2.60.
372 Appendix A

TABLE A-4 Si Subscales

Si, - Shyness/Seif-Consciousness (14 items)

True
158 161 167 185 243 265 275 289

False
49 262 280 321 342 360
Males: Mean 4.80; S.D. 3.44. Females: Mean 5.31; S.D. 3.69.

Si2 - Social Avoidance (8 items)

True
337 367

False
86 340 353 359 363 370
Males: Mean 3.12; S.D. 2.33. Females: Mean 2.75; S.D. 2.19.

Si3 - Alienation —Self and Others (17 items)

True
31 56 104 110 135 284 302 308 326 328 338 347 348 358
364 368 369

False
none
Males: Mean 4.99; S.D. 3.35. Females: Mean 5.23; S.D. 3.59.
 MMPI-2: Item Composition of Validity, Clinical, and Supplementary Scales 373

TABLE A-5 Supplementary Scales

A - Anxiety (39 items)

True
31 38 56 65 82 127 135 215 233 243 251 273 277 289
301 309 310 311 325 328 338 339 341 347 390 391 394 400
408 411 415 421 428 442 448 451 464 469

False
388
Males: Mean 10 02: S.D. 7.10 Females: Mean 11.64; S.D. 7 90

R - Repression (37 items)

True
none

False
1 7 10 14 37 45 69 112 118 120 128 134 142 168
178 189 197 199 248 255 256 297 330 346 350 353 354 359
363 365 422 423 430 432 449 456 465
Males: Mean 15.18; S.D. 4.53. Females: Mean 16.34; S.D. 3.81.

Es - Ego Strength (52 items)

True
2 33 45 98 141 159 169 177 179 189 199 209 213 230
245 323 385 406 413 425

False
23 31 32 36 39 53 60 70 82 87 119 128 175 196
215 221 225 229 236 246 307 310 316 328 391 394 441 447
458 464 469 471
Males: Mean 37.34; S.D 4.46. Females: Mean 34.37; S.D 4 90

continued
374 Appendix A

TABLE A-5 continued

MAC-R - MacAndrew Alcoholism-Revised (49 items)

True
7 24 36 49 52 69 72 82 84 103 105 113 1 15 128
168 172 202 214 224 229 238 257 280 342 344 387 407 412
414 422 434 439 445 456 473 502 506 549

False
73 107 117 137 160 166 251 266 287 299 325
Males: Mean 21.72; S.D. 4.32. Females: Mean 19.78; S.D. 3.65.

Fb - Back F (40 items)

True
281 291 303 311 317 319 322 323 329 332 333 334 387 395
407 431 450 454 463 468 476 478 484 489 506 516 517 520
524 525 526 528 530 539 540 544 555

False
383 404 501
Males: Mean 1.86; S.D. 2.44. Females: Mean 1.94; S.D. 2.58.
 MMPI-2: Item Composition of Validity, Clinical, and Supplementary Scales 375

TABLE A-5 continued

VRIN - Variable Response Inconsistency (67 item-response pairs)

For each of the following response pairs add one point.

3 T - 39 T 125 T - 195 T 349 T-515 F

6 T - 90 F 125 F- 195 F 349 F - 515 T
6 F - 90 T 135 F-482 T 350 F- 521 T
9 F - 56 F 136 T- 507 F 353 T - 370 F
28 T - 59 F 136 F- 507 T 353 F - 370 T
31 T - 299 F 152 F-464 F 364 F - 554 T
32 F-316T 161 T- 185 F 369 F- 421 T
40 T - 176 T 161 F - 185 T 372 T- 405 F
46 T - 265 F 165 F - 565 F 372 F - 405 T
48 T- 184 T 166 T-268 F 380 T - 562 F
49 T - 280 F 166 F - 268 T 395 T - 435 F
73 T-377 F 167 T- 243 F 395 F - 435 T
81 T - 284 F 167 F-243 T 396 T - 403 F
81 F - 284 T 196 F - 415 T 396 F - 403 T
83 T - 288 T 199 T -467 F 411 T-485 F
84 T- 105 F 199 F-467 T 411 F - 485 T
86 T - 359 F 226 T - 267 F 472 T - 533 F
95 F - 388 T 259 F - 333 T 472 F - 533 T
99 F- 138 T 262 F - 275 F 491 T - 509 F
103 T - 344 F 290 T - 556 F 506 T - 520 F
HOT - 374 F 290 F - 556 T 506 F - 520 T
110 F 374 T 339 F - 394 T 513 T- 542 F
116 T - 430 F
Males: Mean 4.52; S.D. 2.39. Females: Mean 4.47; S.D. 2.28.

TRIN - True Response Inconsistency (23 item-response pairs)

1) For each of the following response pairs add one point:

3 T - 39 T 65 T - 95 T 209 T-351 T
12 T - 166 T 73 T - 239 T 359 T - 367 T
40 T- 176 T 83 T - 288 T 377 T - 534 T
48 T - 184 T 99 T - 314 T 556 T - 560 T
63 T- 127 T 125 T- 195 T
2) For each of the following response pairs subtract one point:
9 F - 56 F 140 F - 196 F 262 F - 275 F
65 F - 95 F 152 F-464 F 265 F - 360 F
125 F - 195 F 165 F- 565 F 359 F - 367 F
3) Then add 9 points to the total raw score.
Males: Mean 8.95; S.D. 1.41. Females: Mean 9 00; S.D. 1.32

continued
376 Appendix A

TABLE A-5 continued

O-H - Overcontrolled Hostility (28 items)

True
67 79 207 286 305 398 471

False
1 15 29 69 77 89 98 116 117 129 153 169 171 293 344
390 400 420 433 440 460
Males: Mean 12.51; S.D. 2.94 Females: Mean 13.53; S.D. 2.74.

Do - Dominance (25 items)

True
55 207 232 245 386 416

False
31 52 70 73 82 172 201 202 220 227 243 244 275 309 325
399 412 470 473
Males: Mean 16.62; S.D. 2 95. Females: Mean 16.27; S.D. 2.89.

Re - Social Responsibility (30 items)

True
100 160 199 266 440 467

False
7 27 29 32 84 103 105 145 164 169 201 202 235 275 358
412 417 418 430 431 432 456 468 470
Males: Mean 20.09; S.D. 3.89. Females: Mean 21.02; S.D. 3.36.

Mt - College Maladjustment (41 items)

True
15 16 28 31 38 71 73 81 82 110 130 215 218 233 269
273 299 302 325 331 339 357 408 411 449 464 469 472

False
2 3 9 10 20 43 95 131 140 148 152 223 405
Males: Mean 11.30; S.D. 6.44 Females: Mean 12.31; D 6.99.
 MMPI-2: Item Composition of Validity, Clinical, and Supplementary Scales 377

TABLE A-5 continued

GM - Gender Role-Masculine (47 items)

True
8 20 143 152 159 163 176 199 214 237 321 331 350 385 388
401 440 462 467 474

False
4 23 44 64 70 73 74 80 100 137 146 187 289 351 364
392 395 435 438 441 469 471 498 509 519 532 536
Males: Mean 37.49; S.D 4.56. Females: Mean 28.81; S.D. 6.16.

GF - Gender Role-Feminine (46 items)

True
62 67 1 19 121 128 203 263 266 353 384 426 449 456 473 552

False
1 27 63 68 79 84 105 123 133 1 55 197 201 220 231 238
239 250 257 264 272 287 406 417 465 477 487 510 51 1 537 548
550
Males. Mean 27.32; S.D. 4.70. Females: Mean 36.86; S.D 3.85.

PK - Post Traumatic Stress Disorder-Keane (46 items)

True
16 17 22 23 30 31 32 37 39 48 52 56 59 65 82
85 92 94 101 135 150 168 170 196 221 274 277 302 303 305
316 319 327 328 339 347 349 367

False
2 3 9 49 75 95 125 140
Males: Mean 8.01; S.D. 5.99. Females: Mean 8.52; S.D. 6.56.

continued
378 Appendix A

TABLE A-5 continued

PS - Post Traumatic Stress Disorder-Schlenger (60 items)

True
17 21 22 31 32 37 38 44 48 56 59 65 85 94 116
135 145 150 168 170 180 218 221 273 274 277 299 301 304 305
311 316 319 325 328 377 386 400 463 464 469 471 475 479 515
516 565

False
3 9 45 75 95 141 165 208 223 280 372 405 564
Males: Mean 10.49; S.D. 7.98 Females: Mean 11.82; S;.D. 8.96.

TABLE A-6 Content Scales

ANX - Anxiety (23 items)

True
15 30 31 39 170 196 273 290 299 301 305 339 408 415 463
469 509 556

False
140 208 223 405 496
Males: Mean 5.53; S.D. 4.17 Females: Mean 6.53; S.D. 4.51.

FRS - Fears (23 items)

True
154 317 322 329 334 392 395 397 435 438 441 447 458 468 471
555

False
115 163 186 385 401 453 462
Males: Mean 3.80; S.D. 2.96. Females: Mean 6.59; S.D. 3.60.
 MMPI-2: Item Composition of Validity, Clinical, and Supplementary Scales 379

TABLE A-6 continued

OBS- Obsessiveness (16 items)

True
55 87 135 196 309 313 327 328 394 442 482 491 497 509 547
553

False
none
Males: Mean 4.93; S.D. 3.06. Females: Mean 5.50; S.D. 3.32.

DEP - Depression (33 items)

True
38 52 56 65 71 82 92 130 146 215 234 246 277 303 306
331 377 399 400 411 454 506 512 516 520 539 546 554

False
3 9 75 95 388
Males: Mean 4.79; S.D. 4.62. Females: Mean 5.86; S.D. 5.02.

HEA - Health Concerns (36 items)

True
11 18 28 36 40 44 53 59 97 101 111 149 175 247

False
20 33 45 47 57 91 117 118 141 142 159 164 176 179 181
194 204 224 249 255 295 404
Males: Mean 5.29; S.D . 3.91. Females: Mean 6.16; S.D. 4.47.

BIZ - Bizarre Mentation (23 items)

True
24 32 60 96 138 162 198 228 259 298 311 316 319 333 336
355 361 466 490 508 543 551

False
427
Males: Mean 2.30; S.D. 2.50. Females: Mean 2.21; S.D. 2.49.

continued
380 Appendix A

TABLE A-6 continued

ANG - Anger (16 items)

True
29 37 116 134 302 389 410 414 430 461 486 513 540 542 548

False
564
Males: Mean 5.63; S.D. 3.31. Females: Mean 5.68; S.D. 3.08.

CYN - Cynicism (23 items)

True
50 58 76 81 104 110 124 225 241 254 283 284 286 315 346
352 358 374 399 403 445 470 538

False
none
Males: Mean 9.50; S.D. 5.35. Females: Mean 8.73; S.D. 5.16.

ASP- Antisocial Practices (22 items)

True
26 35 66 81 84 104 105 110 123 227 240 248 250 254 269
283 284 374 412 418 419

False
266
Males: Mean 7.91; S.D. 4.19. Females : Mean 6.17; S.D . 3.70.

TPA - Type A (19 items)

True
27 136 151 212 302 358 414 419 420 423 430 437 507 510 523
531 535 541 545

False
none
Males: Mean 8.08; S.D. 3.68. Females: Mean 7.41; S.D. 3.34.
 MMPI-2: Item Composition of Validity, Clinical, and Supplementary Scales 381

TABLE A-6 continued

LSE - Low Self-Esteem (24 items)

True
70 73 130 235 326 369 376 380 411 421 450 457 475 476 483
485 503 504 519 526 562

False
61 78 109
Males: Mean 4.25; S.D. 3.69. Females: Mean 5.16; S.D 4.24.

SOD - Social Discomfort (24 items)

True
46 158 167 185 265 275 281 337 349 367 479 480 515

False
49 86 262 280 321 340 353 359 360 363 370
Males: Mean 7.65; S.D. 4.77. Females: Mean 7.53; S.D. 4.80.

FAM - Family Problems (25 items)

True
21 54 145 190 195 205 256 292 300 323 378 379 382 413 449
478 543 550 563 567

False
83 125 217 383 455
Males: Mean 5.32; S.D. 3.52. Females: Mean 6.14; S.D 3.77.

WRK - Work Interference (33 items)

True
15 17 31 54 73 98 135 233 243 299 302 339 364 368 394
409 428 445 464 491 505 509 517 525 545 554 559 566

False
10 108 318 521 561
Males: Mean 7.30; S.D>. 4.98. Females: Mean 8.51; S.D. 5.45.
382 Appendix A

TABLE A-6 continued

TRT - Negative Treatment Indicators (26 items)

True
22 92 274 306 364 368 373 375 376 377 391 399 482 488 491
495 497 499 500 504 528 539 554

False
493 494 501
Males: Mean 4.70; S.Ci. 3.71. Females: Mean 5.02; S.D. 3.98.
 APPENDIX B

MMPI-2: T Score
Conversion Tables

Source: From the Minnesota Multiphasic Personality Inventory-2 Manual by J. N. Butcher, W. G.

Dahlstrom, J. R. Graham, A. Tellegen, & B. Kaemmer, Appendix A, pp. 53-64. Copyright © the
University of Minnesota 1 942, 1 943, 1 951, 1 967 (renewal 1 970), 1 989. This Manual, 1 989.
Reproduced by permission.

383
 UJ cooM^-oa>oof^<£>ifi'tcoc\j^-QCT)oot^cpiO’>4coc\j»-oaoooi^c£)m^rnc\j’-ocooor^c£)io
I cc r— •— r^-'
-cocococococococococoioiotr>i7)i-‘~~~‘.* * ' • J-
a: O
o
co
CD '<4 CO »- ( i <J) i I f—- I iuo'<rcocM’-a)ooi^c£)Ln'tcocM>-ocnr^i icorroocMT-oaoco
CO 0) 0) <J) 0) i > 00 I I oo I loooocooocof^-i^'r^t^i^i^-r^r^-r^h'coco' icocococococoloio
_ * OCOlOCNOlCDCOON^
<0 CM N------00000)0)
4

u OCT)aocomcocM’-ooooc£)mooc\ioa)N-co,^<r)»-Qooi^inTtc\i’-ajoocomcoc\ioa)iv'C£)un
CO CM --O 0) CD 0) 0) OO 0) 0) 00 00 CO 00 00 00if-—r—
OOOOOOI I cd co
4

* oa>f'-irwc\joao< iiocO’-a>r'~iO'3-c\jooo(£>'4coT-CT>r-incoc\jocxicO'4
>ooo0>0)0)0)0)0)aoaoaooocor-i''-f-i'-r-.r-c0coco
4

CO
cl

0®10®0®®NOOJ
cocooooo^-o-rrrrmio
C/)
0)
CO ■d •4 0®®®ON®OjON4CM®Ntf^
c\iT-»-r-»-oooocr)a)cr>ooooooco
E Q.
CD
LL
>. O 00 lO CO COCO’’4-r-<T>C£>'4C\JCT>r-rfCMO
CM ■»— *— 1 00000)0)0)00 00 00 00 00 00
X

OCOCO,'TCM0)f--i0CO»-0)COr}-CMOCOCOCO'--
(\Jr-r-r-^000000)0)0)0)0)®ffl®®
Uniform and Linear T Score Conversions for Basic Scales with K-Corrections

oo m 0®(D®’-0)N®®T-0)N
CMr---r-^000000)0)

1 co c- cp 0 co cm O O) r- co to -4 co cnj o O) oo r- co m co cm o O) oo co m ^ co cm t-
CO 8 : i O) O) O) 0)0)0)0)0)ooooaooocoooooaot'^r'-f'-f-r'-r'-r-r-<--cocococococococo

«
CO *
CM OC-M'OC-'cf’-OO'^-
cmt-t-^ooocdo)
4

*
o
CO
v—
4
CM S!
>-'-'->-oooooo)0)aiO)0)o>ffl®®®®®NNNNNN(DO®
* O0)f--ir)C0t-0)C0^cMQ00CD'^r’-0)r-cnc0T-a)r-.TrcMO
CMr-^r-^r-000000)0)0)05®®®®®NNNNN
4

a?

8 r-.ir)cO’-0)r-cncoT-o)r-mcO’-0)cocjo^-CMOco
00000)0)0)®®®®®®®NNNNSN®

<f>
CO
n o- ON®CMON®OJi
(\|t-r-r-i-OOOl
N®CJON4CJO)
iO)O)O)O)0QCO00C-
4
00)CO'4’-0)CO'4’-O)CO'4''-0>CO^-
I (\j^r-^r-ooooo)0)0)0)®m®

CC|i-^rT-t-^OOOOl >ooc-inco’-0)r-mco
O0)f-ir)'4CMOOOCO'0CMi
i0)0)crj0)0)ooooooco

oo m O0)O'^-CMO00Lnc0’-
CM^t-t-^^-0000

Z
TABLE B-1

o
h-
o
<
cr
LL
cr
* O
<
cr
o
co
CO CM 00)OOC-‘OlD’^-COCM’-00)OOf--C£)lO'tCOCMT-00)aOf--COLnM’COCM’-00>COC^COLn
~ ~--
f— f— I—- CO CO CO

384
O’ CO CM o CT) CO Is- CD in O’ CO CM T- O CT) CO Is- CD in O’ CO CM

6
5
4
3
2
o CT) CO Is- i- o
CO CO CO CO CO CM CM CM CM CM CM CM CM CM CM 1—

*
Is- in O’ CO CM O CT) CO r- CD in CO CM O CT) CO Is- CD in O’ CO CM
in in in in m in in O O’ o- o o O’ O’ O’ O’ CO CO CO CO CO CO CO CO CO
o
c/5

T- CO in CM CT) CD O’ i— 00 in CM CT) CD CO -i— CT) Is- in CO 1— CT) Is- in CO T— O 03 CM

CT) CO CO CO Is- Is- Is- Is- CD CD CD in in in in O’ O’ O’ O’ O’ CO co CO CO CO CO
+

CO CM o CT) Is- m co CM O CO CD O’ CM CT) Is- CD O’ CO CM o O

CO CD CD m m in in m m O’ O’ O’ O’ o CO co CO CO CO CO CO CO CO
+

C\J t- 0) N- in CO i— CT) Is- O’ CM o

CT) Is- in O’ CM o
Q_
CD CD m m in in in O’ o- O’ O’ O’ CO co CO CO CO CO CO
+

o 00 O’ 1_ r- CO O CD CM CT) LO CO O’ o 1^ CO CT) CD CM CT) in CM CT) Is O’ CM r- O

CM i— i- 1— o o O CT) CT) CO CO 00 Is- Is- Is- CD CD in in m O’ O’ O’ CO CO CO CO CO CO Q_
1 1 1 1

in Is- o CM m Is- CT) CM O’ Is- CT) CM O’ 1^ CT) CM O" CD CT) O’ CD CT) O’ CD CO O

m in CD CD CD CD CD Is- Is- Is- Is- 00 CO CO C0 CT) CT) CT) CT) O o O o 1— 1— CM
T_ T~ T- 1— T_
CD
CD
CT) CD CO T— CO CD CO o 00 m CO T— CT) Is in CO T” CT) Is- CD O’ CM o ra ~o o
Is- Is- r- Is- CD CD CD CD in in in m O’ O’ o O O’ CO co co CO CO CO
E Cl
+
CD
LL
r- m CO o 00 in CO t— 00 CD o T— CT) r- m CO CT) CO CD in O’ CM CM o >)
Is- Is- Is- Is- CD CD CD CD in m in in O’ O’ O’ O O’ CO CO CO CO CO CO CO co CO I

cn Is- in CM O CO CD O’ CM CT) Is- in CO CT) 1^ CD O’ CM O CO CD O’ CM o L_)

r- Is- Is- Is- Is- CD CD CD CD in m m m in O’ O O’ O’ O’ O’ CO CO CO CO CO

uo cm o CO CD o- CM O CO CD O’ 1— CT) r-. in CO -1— CT) Is- O’ T— CT) CD CO o CO in CO o (f) in

<j> CT) CD 00 CO CO CO CO Is- Is- Is- Is- CD CD CD CD CD in in in in O’ O’ O’ O’ CO CO CO CO X
+

_
CO CO CD O" CM o Is- in CO CT) CD O’ CM oCO CD CO CT) Is- in CM o *
CO CO Is Is- r- Is Is CD CD CD CD uo in in in in O’ O’ O’ O’ CO CO CO CO CO

o CD CO CT) CD CO CT) CD CM CT) in CM CT) in CM 00 in com-i-coo- T- N-

CM 1— 1— o O o CT) CT) CT) CO 00 00 Is- Is- Is- CD CD CD m in m o- o- oo LL
1- T~ T~ 1

in o in O CD ^_ CD 1— CD CM Is- CM Is- CO CO CO
o o CT) CT) CO CO 1^ h- CD cd in in o- o- co co 1j
i— ■v—

*
CT) CO Is- CD m O’ CM O CT) CO Is- in O’ CO CM o CO Is- CD in O’ CO t— o
in in in m m in m in in O’ O’ O- O’ O’ O’ O’ O’ O’ CO CO CO CO CO CO CO CO CO

T~ 00 m i— CO m CM CT) in CM CT) CD CO T— CT) Is- in CO T— CT) CO CO m CO T— o 03 CM

CT) CO CO CO Is- Is- CD CD CD in m lO in O’ O" O’ O’ O’ CO CO CO CO 00 CO CO
+

CO CM o CO CD m co CT) Is- m o CM o CT) Is- CD in O’ CO CM o o

CD CD CD in in in in in O’ O" O’ o o o
CO co CO CO CO CO CO CO CO CO
+

■
CD o CM CT) Is- in CO CT) Is- O’ CO CT) N- CD o CO CM O
CD CD CD m m in in in O’ O- O’ O’ O’ CO CO CO CO CO CO CO CO 0-
+

o CT) CD CM CO in Is- O’ o CD CO CT) n CM CO O’ 1— Is CO CT) CD CM CT) Is- O’ CM r- o

v.U
CM T*“ T— o o o O) CT) CT) 00 CO Is Is- Is- CD CD CD in in o O’ O CO CO CO CO CO CO Q_
1 T“ ' i—

*
cd o- CM o 00 CD O’ CM o CO CD O’ CM O CO CD O’ CM o
CD CD CD CD in in m in in O’ O’ O’ O’ O’ CO CO CO CO CO

0)
CD ■a
r— O’ CM CT) Is- O’ CM CT) Is- O’ CM o CO CD O’ CM O CT) Is- in O’ CO T— o o
Is- Is- Is- CD CD CD CD in in in in m O’ O’ O’ o o CO co CO CO CO CO CO 03 0-
+

i- CT) CD O’ CT) CD O’ i— CT) Is- O’ CM o Is- in CO CM o CO n O’ CO CM T— o >.

CO Is- r- Is- Is- CD CD CD CD in m in in m o- O’ O’ O’ O’ CO CO CO CO CO CO CO CO I

T- O 00 CD •o CM o CO CD O’ CM CT) Is- O’ CM o Is- in CM o CO CD O’ CM o

1_1
00 CO Is- Is- Is- r- Is- CD CD CD CD 5 in m m in m O" O’ O’ O’ CO CO CO CO CO

CT) Is- ■o CM o CO CD O’ T— CT) r- in CO o CO CD O’ CM O) h- O’ T— 00 in CM CT) 1^ m CO CM -i- T- o CO in

CT) CT) CT) CT) CT) CO co CO 00 Is- Is- Is Is- Is- CD CD CD CD in in in in O’ O’ O’ CO CO CO CO CO CO CO CO X
+

_ *
CT) Is- in CM O 00 CD O" CM o CO CD O’ CT) Is- in CO CT) Is- in CO o
5 Is- Is- Is- Is- Is- CD CD CD CD CD in m in in O’ ■o O’ o o CO co CO CO CO

o CT) CD co O Is- O’ CO m CM CT) n CM CT) CD CO o Is- O’ i— CO in t- oo m cm ct) cd

CM o o O CT) CT) CT) CO 00 CO Is- Is- Is Is- CD CD CD in in in O’ O’ O’ co co LL
-r_ 1—
’— ’r~ ’r_ T” ’r~
O
o CD i— Is- co CO O’ O in T- CD CM CO CO CT) in
1f—\
o CT) CT) CO CO r- r- Is- CD cd in in o- o- co co 1J CJ
<
cc
LL.
6
5
4
3
2

O CO CM r— O CT) 00 Is- CD in O’ CO CM O CT) CO Is- CD in O’ CO CM T- o CT) C0 I-. T- O

T— 1— 1-
CO CO CO CO CO CM CM CM CM CM CM CM CM CM CM T~ T— t_

385
 > LU (MT-OmcON(DO'TCO(NjT-Oa)CON(OD,tCOWT-0(3)CONCDlOO-COC\J - O O) CO N CD
KNNCDCDCDCflCDCDCDCDCDCOcflcncfilOlOininiOLOlOO-'TO-O-O-'TO-O- O' O’ CO CO CO CO
cr O
O
co

LO'tCOr-OOCDNCDCfiO'COCMT-aiCONCDlO^fCOCM'-OCnNCDinO'COCO'-Oa)
co OlO)aiCDa)COCOCOCOCOCOCX)COOONNNNNNNNNNCD(r)(DCDCDCiDCD(DCDlfi

CO O (Tj CD O' CO CD OJ
C\J -r- o o o

o O CO N CD CD ■cf CO CM OUDKCDlDO'CO(MOa)CONCDlDO‘COCJOO)CDNCDlD^CO'-Oa)
OOOOOOOOOffiO)CnO)0)ffiO)roffiCOCOCDCOCOCDCOCOOON
co

CONIDM-COt-OCDNCD'cCCOW
cncncDcricnairococDcDcococo

CO
CL

OCOlOCOOCODNO
COCOCOCOO-TO-O-lD
CD
0)
O CO CD CO CT) CD Tt CM <T> h-. LO
CO "O
CM O O O O CT) CT) CT)
E CL
CD

> o co co co coco^r^-cnicD'O-CMCor^.^rcM
CM T- T- T- ooooffiffiairococooDco
X
Uniform and Linear T Score Conversions for Basic Scales without K-Corrections

OCOCDM-CMONIDCO’-CJICDM'IMOCOCDCO
Q N^->-^T-000000)a)0)0)0)COCOCO

co
X

OCT)CON(DM'COCM'-OaiNCDlDM'COCMOO)OONCDincOC\jT-oaiCOCDin\fCOCO
Oa)0)O)O)O)CnCna)aHOCOCOCOCOCOCOCONNNNSSSNN(DCDCDCDCDCDCD
CO

CO O CX) UO CM CT) <X) CO

CM ■’ ■’ ■' O O O

o O CT) N- CD LO CO OCT)CONLOCTCOC\JOO)CONCDM'COC\J’— OCONCDLDM'CO^

t-OOOOOOOOCT)CT)CT)CT)CT)CT)CT)CT)CT)COCOCOCOCOCOOO
co

M'CO^OCOMOM'C\J'-O)CO(0
CL OOOOCT)CT)CT)CT)CT)CT)COfflCO

co
CL

a)NlOCO’-C3)NCOCOr-CT)NlOCOT-CT)COCOM'C\IO
OOOOOCT)®CT)CT)CT)COCOCOCOa3NNNNNN

CO
_CD TO O N LO CO CT)CDM-CMONLOCO
CO CM T- T- OOOOOCT)CT)CT)
CL

O CT) CO •M- CT)CD^t-CT)CDM-t-CT)CD

X CMT-r-r- OOOOCT)CT)CT)CT)OOCO

O CT) C'- LOM'C\JOOOCDM'C\JOCONlOCOf-0)NlO

Q CM 1- r-T--r-i-OOOOOCT)CT)CT)CT)CT)COCOCX)

co
X
TABLE B-2

LU
CL
§ O
< o CM-’-OCT)00r"-<X)cO^rC0CM''-OCT)C0r'^(0l0^t-C0CM''— OCT)COr^tDLOMrcOCM'<-OCT)COI'^CO
CC CO NNNlDCDCDCDCDCDCDCDCDCDLOiOLOCOlDlDiOlOlOlOM'M'M'M'M'M'M'M'M'M'COCOCOCO

386
in d CO CM T- o CT) CO td CO m d CO CM o CT) CO e-- CO n d CO CM o CD CO i^. CO n d CO CM o $ LU
00 CO CO CO co CO CM CM CM CM CM CM CM CM CM CM T- T— T— T— 1— < (X
CO o
o
CO

*
CO f- in CO CM O CT) CO c- CO m CO CM 1_ o CT) 00 1^- CO n d CO CM o
LO m in LO m in in in d d d d d d ^r ■CT rT CO CO co CO CO CO CO CO CO CO

o e— d "T- CT) CO CO o r- in CM CT) CO d T“ CO n CO CT) n CO o 00 CO d CM o CD

o CT) CT) CT) 00 CO CO CO r- fd c- CO CO CO CO n n n n ■d •sr d d d d CO CO CO CO CO

00 e— CO m d CM o O) CO id CO in CO CM o CT) 00 CO n CO CM O 00 r^ n CO CM CT) e'¬ n CM o o

e— l~- f—■ t'- r- N- r— f- CO CO co CO CO CO CO 5 CO LO n n n n n n in n d d d d d CO en CO CO CO CO

o CT) CO CO in CO CM CT) CO id in d CM o CO f'- CO n CO CM o CO r- CO d CO O) n CO o

u_
CO r- L- r- e-. e-. [-. id CO CO co CO CO CO 5 CO n n n n n n n n d d d d d d CO CO CO CO CO CO

o co d ^_ r- CO o CO CM CT) n 00 rt o r"- CO CD CO CM CD n CM CT) e'¬ d CM o CD

CM 1— T— i— c o o CT) CT) CO CO CO i"- e'¬ co CD n n n d d d CO en CO CO CO CO CL
1 1 1 1 T~

CM in h- O CM m e— CT) CM d fd CT) CM d e^ CT) CM ■d- CO er) T- d CO CD _ d CO CO O *

uo in in CO CO CO CO CO Id |d id r- CO co 00 00 0) CT) CT) CT) O o o o ■<— T— T— T— CM d
i— 1 —
"■ ■ "!— "r~
C0
CD
CM o 00 m CO _ 00 CO d CM 0) r- in CM o CO n CO CT) n CO CD n CO o 03 "O
CT) CT") 00 CO CO CO e-- e— c- fd CO co CD CO CO n n n n ■d" sd d d d CO CO CO CO CO CO E 0-
CD
U_

o h- in CO o CO in CO CO CO d 1- CT) r-- n CO y— CT) 00 CO n d CM CM T_ o :>

00 r— e-. N- e— CO CO CO co in m m m d ■'t ■'d- sj- CO CO CO CO CO CO CO CO CO I

_
CT) !—■ LO CM o CO CO d CM CT) r— in CO CT) CO sd CM o 00 CO d CM o Q
CO r- I-. r- C— r— CO CO CO CO in m in m in ^r "Sf •d- -d" d" CO CO CO CO CO

_
o CO e— in CO CT) |d in CO CM o 00 CO "St CM O CD i-- n CO CT) r- CO d CM CT) r- d r- CO 0)
o CT) 0) CT) CT) CT) CO co CO CO CO 00 e'¬ e^ e^ i-- CO co CO CO CO n n n n n d d d d co CO I

*
CO CO CO d CM o |d in CO er) CO sd CM o CO CO CO CT) e'¬ n CM o
CO 00 r- |d id f- e- co CO CO S n n n n n d d d d CO en CO CO CO

_
o CO CO CT) CO CO CT) CO CM CT) n CM CD n CM 00 n CO n T- CO d I-- *
CM 1— 1— o o o CT) CT) CT) 00 00 00 f- r- I'- CO CO CO n n n d d d co LL

n o n o co CO CO CM r-- CM [■- CO CO CO
o o 0) CT) CO 5 r- e'¬ CO CD n n d d CO CO Li

CT) 00 I-. CO m d CM o CT) CO r- LO n- CO CM O 00 CO n d CO o

CO in in in in m m in in m d d d d 't 'cr rr ■sr sd CO CO CO CO CO CO CO CO CO

o I-. d CO in CM CT) CO CO O ed d 00 co CO CT) r'- n CO CM o 00 e'¬ n en o CD

o CT) CT) CT) CO CO C0 e— e- e'¬ r- co CO CO n n n n ^d ^d ■d- d d d CO en CO CO CO CO
T“

o CT) CO CO in d CO CM o er) co fd CO d CO CM o 00 e-. CO n d CM CT) CO CO n CO CT) e'¬ n CM o o

CO I-. t'- h- e— e— e— Id e— CO CO CO CO CO CO CO S CO n n n n n n n d d d d d d CO en CO CO CO CO

m CO CM O CT) CO CO m co CM o CT) e— CO ■"t CO T- o 00 e^ CO d CO CM o CT) i'- CO d CM CT) i-- d CM o

o!
00 00 CO CO h~ e— e- fd r- r- e— CO CO CO CO CO CO CO n n n n n n n d d d d d d CO co CO CO CO

o CT) CO CM 00 m id d o CO CO CT) n CM 00 d i— r-- CO CT) CO CM CT) e'¬ d CM i— o CD

CM T— T— T— o o o CT) CT) CT) CO CO r-- r^. i'- CO CO CO n n d d d CO en CO CO CO CO CL
•Y~~ "r” 1 11—

CO CO d CM o CO CO d CM o CO CO d CM O 00 CD ^d" CM o
CO CO CO CO CO m in in in in d d d d •3" CO CO CO CO CO

CO
0)
CT) CO d CM o e- in CO 00 CO d CM O N- n CO CT) CO d CM o CO CO d CO o "O
CD CO 00 00 00 CO e- id fd h- CO CO CO CO CO n n n n ■sd d d d d CO CO CO CO CO CO 03 Q_

CT) c- d CM o N- n CO CM o 00 e'¬ n d CO CM 5-
d 7- CT) CO d CT) co d o
CO 00 e— e- e- t- CO CO CO 5 in m in m n ■sf ■d" ■d" ^d CO en CO CO CO CO CO CO I

CO o CO CO d CM o CO CO d CM CT) e- CM o e^ n CM O CO CO d CM o o
CO 5 CO r- t'- e— i-. r- CO CO CO CO 5 in n n n n ■d" ■d" d d CO CO CO CO CO

r- m CO CT) ld in CO CT) e- n CO CT) 1-. n CO

_
CT) i'- n CO
_
CT) r- d CM CT) CO CM CO d (/)
o o o o CT) CT) CT) CT) CT) CO CO CO 00 00 e^ e^- e'¬ h- CO co CO CO CD n n n n d d d CO CO X
T“ 7— "r”

CT) Id in CM o 00 CO CM o CO CO ^d
_ er) n CO
_ CT) h- n CO o
CO Id [d id f— e'¬ CO CO CD CO CD n n n n d d d CO CO CO CO CO

O CT) CO er) o id CO n CM C!) n CM CD CO CO o e- O'

^
CO
_ n _ CO n CM CD CO *
Linear T Scores

CM o o o CT) CD CT) 00 00 CO h- e'¬ r~- h- co CD CO n n n d d d CO CO LL

t—

o co t'- CO 00 O’ o n CO CM 00 CO CT) n
o CD er) co 00 e'¬ h- r- CO CO n n d d CO CO Lj
1—
UJ
X
<; o
in d CO CM o CT) 00 [d CO in d CO CM •*— o CT) CO l'- CO LO CO CM -T— en O) co CD n d CO CM T— o <
CO CO
CO CO CO CO CO CO CM CM CM CM CM CM CM CM CM CM ’r~ "S” T~ i- 'r_

387
 > UJ cg<-oo3coN(DU3oro)iAiT-o LO O' CO CM T~ o 03 00 Is- CO LO O' CO CM 1— o
isu
COCOCOCMCMCMCMCMCMCMCMCMCM 1-

C/3

03 CM co 03 CO CO o CO Is- o
(XI
00 CO Is- CO CO LO LO ■O' co CO

co
Is- LO 03 CO Is- 1— LO o
o- ■^r
03
Is- Is- CO LO LO CO CO

03
CM
co CO 00 co 00 CO 03 O' 03 ■^r o
Is- Is- CO CO LO LO o- O' CO CO CO

o- CO CO o CM LO
03
5 Is- CO LO LO ■^r CO

,_
ID O 03 ^r 03 O' 03 Lfi o LO o LO o LO o LO CD
a CM T- O o 03 03 03 CO CO Is- Is- CO CD LO LO ■O' ■O'
co t_ T—

Is o COCO 03 CM LO CO _ ^r Is- o
o T— T— o 03 00 CO Is- CO CO LO ^cr O'
co T— 1—

or O' 03 CO CO CM Is- CM CO
__ LO o LO 03 ■^r CT3
a o o 03 03 00 CO Is- Is- CO CO LO ■O' ^r CO
co 1—

CO CO CO o o- CO CM CO o CO CM
O O 03 03 CO Is- Is- CO CO LO xr O'
CO 1

CM o Is- Is- 00 CO CO 03 03 o o
O CM 1— o 03 CO Is- CO LO LO o
CO 'T— ’r— T_

o N CO 03 LO
_ r- CM 00 O o CO CM 00 o- 03 LO y_ Is- CO a3
(J CM T- X- O O o CT3 03 CO 00 CO Is- Is- CO CO LO LO LO ■o ^r CO
CO 1

co _ o
03 o LO o CO CO CO CM
cl Is- CO CO LO LO o- o- CO CO CO

_
Linear T Score Conversions for Harris-Lingoes Subscales for Males

CM
ex CO 03 CM CO 03 CM LO CO O'
CL 03 CO CO Is- CO CO LO 'O CO

O CO CM COo O' CO CM CO o o- CO CM CO o
03 CM T— T— o o 03 CO CO Is- Is- CD LO LO O'
CL T~ T_ T— ’

uo _
Tj
r— CM Is- CM Is- CO co CO CO CO 00 o
Cl
03 CO CO Is- Is- CQ co LO LO O' ^r CO CO

o
TJ 03 o- CO CO CO CO Is- CM Is- CO CO o
Cl
03 03 CO CO Is— Is- co CD LO LO o ■o CO CO

CO
T3 o- CO CM CO o LO o
CL
CO LO LO o ^r CO CO

CM
TJ o- CO LO CO CM LO o
Cl 5 Is- CO 5 LO O CO CO

TJ co T— o- CO LO CO LO CO
Cl
03 03 CO Is- Is- CO LO LO ■^r CO

ui
> 00 1— 00 LO CO o co o
X Is- h- CO LO o ■o CO CO

O LO
_ lO
_ CO
__ CO CM r- CM CM Is- CM CO co CO
>- CM T- T— O O 03 03 CO 00 Is- Is- CD CO LO LO ■o ^r CO
X T_ 7—
CO CD CM Is- CO CO o- 03 LO o CD _ Is- CM 00 co CO
O O 03 03 CO co Is- Is- Is CD CO LO LO 'cr ^r CO
X ’’ 1

>
CM
1— Is- co 03 LO
_ h- CO o CO CM o
X t-- CO co LO LO LO o- o- CO CO CO

>- CO LO o o
X CD LO LO XT ■O' CO CO

uo co 1— LO 03 CO CM Is- LO o
Q 03 03 CO Is- CD CO LO LO ^r o

o LO
1_ co 1_ CD CM Is- CM N. CM CD CO CO CO CO
Q
T— o O 03 03 CO CO Is- Is- CD CO LO LO ^r ■'3' CO
X- ■’

CO COo T_ CO LO Is- 03 CO LO o
Q
T— o o 03 00 h- co LO LO ’Lt CO CO
T“

OJ CO CM Is- cc o LO 03 ^r co CO Is- CM o
TABLE B-3

Q CT3 03 00 00 Is- Is- CO LO LO O' o- CO CO CO

co o- CO CO CO COrnOONiONffiK ^_ 03 CO o- _ CO CD CO o CO LO CM o r- LO CM
Q o o o CTIOIOICJICOCOCDNN h- CO CD co CO LO LO LO LO o- o- o CO CO CO

LL)
cc

<o CM’-oCT3cor-cOLn^-cocM’-oa3coh--coLn
LO or CO CM o 03 00 Is- CD LO O' CO CM o
CC CO COCOCOCMCMCMCMCMCMCMCMCMCM'-r-’-

388
 > LU c\jr-oa3coN-cDLr>^rcoc\ji- O 03 CO N LO O' CO CM T— 03 Is- LO T—
o CO CO O' CO CM o
C\| 1- T-

i§
COCOCOCMCMCMCMCMCMCMCMCM T- T-

C/3

03 CO o O- CO CM CO 03 CO Is-
CO CO Is- CO CO LO O' o co CO

CO
03 CM LO CD CM CO o CO I-- o
CO Is- CO CO LO LO o co CO

CM
03 o LO O LO o LO o LO o LO o
00 Is- Is- CO CO LO LO O' O' CO CO

03 Is 03 o CM o LO r-
CO Is- Is- CO LO o CO

,_ _ _
CD o cn o- o LO CO Is- CM CO CO 03 O' o LO
O CM o o o 03 03 00 CO Is- Is- CO CO LO LO LO o O'
CO T- T— T—

co O o- Is- ^_ LO ao CM LO 03 CO CO o
a T— o 03 03 CO Is- Is- CO LO LO O' O'
co 1- T—
^_ CO LO LO LO LO 03 03 03
O o o o O' o
o 1— o O 03 03 CO CO r- r— CO LO LO o o CO
CO 'I—

co CO CO o Is- -_ LO 03
O' CM O CO
o o 03 03 CO CO r- CO CO LO O' O'
CO T—

CM O CO o- LO CO CO Is CO 03 o
O CM T— o 03 CO Is- CO LO O' O'
CO ,— T—

03 LO co o- CO CM CO O' Is- CO 03 LO T— Is- CO o CO CM CO

a o o 03 03 CO 00 CO r-~ Is- CO CO CO LO LO LO O' o CO
CO

CO
03 O) LO o LO o LO CO o
CL CO CO CO LO LO O' o CO CO CO
Linear T Score Conversions for Harris-Lingoes Subscales for Females

CM
03 o- CO CM LO 03 CO CO o O'
CL 03 CO Is- r- CO LO LO O' o CO

Is _ Is- 1— Is
o LO 03 CO LO 03 CO T— LO 03
03 CM T— T— o 03 03 00 00 Is- CO CO LO LO o CO
CL 1 ‘T“ 1
cO
CM Is- CM Is- CM 00 co CO CO CO CO 03 O'
"O CD CO CO Is- Is- CO CO LO LO O' o CO CO
CL

Is- CM CO CO CO o LO o LO o LO o
■a
CD CD CO 5 Is- Is- CO CO LO LO O' O' CO CO
CL

co
T3 LO 03 O' CO CM CO
CO m LO O' O' CO CO
CL

CM
CM LO Is- o CM LO Is- o CM
~o 03 Is- Is- LO
CL CO CO o- o CO

TO CM CO o o- co CM CO o O' CO
03 CO CO Is- CO CO LO LO o CO
CL

lO
> Is- o CM O' CO 03 o
Is- Is- CO LO O' CO CO CO
I

_ r- Is- CO _ r-
LO T- CO 03 LO CO 03 LO 03 LO
>- o o 03 03 00 CO 00 i—— Is- CO CO CO LO LO ■o O' CO
X

CO
> 03 LO h- CO 03 LO Is- CO 03 LO Is- CO 03
03 03 03 CO 00 Is Is- Is- CO CO LO LO LO O" O' CO
X

CM
>- Is- co 03 LO o CO CM 00 O' o
CO co LO LO LO O' o CO CO 03
X

>• CO LO o LO o
CO LO LO O' O' CO CO
X

in 03 CO 00 CO CO CO 00 CO Is- CM Is-
Q 00 CO Is- Is- CO CO LO LO O- O' CO

CO CM Is- co 00 O' 03 LO o CO Is- CM CO CO C30

o O 03 CD 00 CO Is- Is- Is- CO CO LO LO O' O' CO
Q

__ O'
Is- o C0 LO 00 o CO CO CO o
o o 03 CO Is- f- CO LO o O' CO CO
O
T~

C\J LO O 03 CO CO CM Is- 1_ CO
_
LO o
TABLE B-4

Q 03 03 CO Is- Is- CO CO LO LO O' CO CO

1_ Is
CO LO CO co co o- rocOTWCDNinoio Is- LO CO O 00 CO CO 00 CO o- 03 O' CM
o o o co CO CO CO LO cn LO LO O' O' O- M- CO CO CO CO
Q

LU
cc
50
< o CM^OOltDNOlOTONi- O 03 CO N LO O' CO CM o 03 CO Is- CO LO CO CM o
cr co COCOCOCMCMCMCMCMCMCMCMCM CM T" T- 1- i— i— -—

389
 > LU a3co^-coLn^rooc\ji-OCT3cor^coLn^rooc^r-oa3aDh--c£)Lr3'cr(?oc\jT-oa3coh~cDLn^-cocoT-o
5 oc Cr3rOCOOOCOrOCOrO(r3COC\l<MC\JCLjCMC\ICMC\ICcJCvjT-'^'^T-T-T--^7-T--^
a: O
O
cr

op T-NC/)0<OrclCO^ONC003lO'-N'tO(I)C\J
03 OCT3a303cooor^h-N-coiX3LnLnLn^r^r'croocr)

""

O cr3coa3CDc\ia3<X)<Aja3<x3CMcr3cocMa3Lnc\iCT3Lnc\iCT)Loc\i
03 ooa3cnCT3cooocor^r^-N-c£)cocoLnLnLr3^r^t^cocooo
2

op T-r^-c\jaD^tcT3LOT-r^c\joo^t03LnT-
03 03cooor^-h-co<x>cDinLn^^tcoco<n
CL
Linear T Score Conversions for Wiener-Harmon Subtle-Obvious Subscales for Males and Females

O OLnT-h'COco^fOLnT-r^cxicxD'cfOLn-r-i^coco
03 CM-^-^ooa3a3CT3coaDr^r--c£)coc£)LOLn'cr'cfco
1- 1- T— 1- 1-
Q.

op
LnT-N-coaiLn’r-i^-roaiLn'^N-cotDLni-
<s> "6 CT3C3cocoh-r^h~cocoimnLO^r^rcocoro
CD CL
03
E03
LL
o 1
■T-coLncoo^'LncMa3r^^rT-a3C£)cnocoLnc\joh~^rc\iC3cr3'cr'^coLO
T-ooooCT3CT303cocococor^r^i^h~cDCD<X3c£)LnLncn'cr^r'c}-'^coco
~rj
CL

op
co<x)ooT-a3CD'^--^-CT3CO-Nr-r-a3r^^i-cMa3r-^oj
t^.(^r^h~(X)CDcx3cDLOLnLnLn'sr^r^i-^j-cocoroco
X

O c\iocoLOcr)T-co(X)ro-^a3co^rc\jCT3r^Lnc^oi^Lnoooco<X)irOT-coco^T-C3i^
>, '-T-OOOOa)a)03C33a3COCOCONNNNNCDCDCD(flLruncmO^^\r^C003
I
"""

op r^r0CT3m-r-r^cr)C0'^-O(X)CM
6 i^-r-coco<x3Lnco^r^^nco

o coococD^c\jocoto^rooCT3N-ir)co-^-a3i^LoroT-03i^LncO'^-a3r--Lnco^-co<X)^rc\joco<x)'crc\i
r-r-OOOOO0)0303C33C0C0C0C0C0NNNNNCDCDCD<£iC0irum0L0ir3^\fM-^'cfC00303C0
Q t— i— i— t— i— i— y—

CO
' couoT-r^ooocDCMcnLO’-N-^rocDCMaciLOT-
Cu
a30303a3ooooi^t^<X)cDcDLnLnLn^r^}-rocnoo

o lot-colot—co^'—co^T-h-^T-’—i^-'^-T-t^'cror^'cj-o
03 ooCT3a3CT3cococor^i^h'<X)iX)CD'-mr3Ln^-'cf-^-coooro

0?
occcMi^coCT3Lnococ\jcocoCT3Ln'^-
03 CT)OOCOr^l^<X)COC£>LOLO^-cd-COC'OCO
0.

O cocMr^c\icocoa3^-OLnoc£)-r-r^coi^roco
CO ■^T-ooa3CT3cococor^r^cocc>LnLn^-^co
1 1 T T—
CL

CO
• un-^h~coa3CDc\)aDxfoci3c\jaDcn^r^co
~o CT3C3cr)coi^r^r-co<x>(£)LnLn'c3-^t'3-coco
C/3 D
i—L.
03
03
o CCCDCOOODcOCMON'tC\J03(D'tT-aDCD030COLOC\JON^(M03(0't
JJ ooooCT3a3a3a3coooQoi^h'^r^cDCDCDCDLDLnLr)Ln'Nr'c}-'cr(rococo
1 T“ 1 1
CL

CO
1
ooLOco-'-coco^rT-a3N.M-c\ior^iocD''-co<xwT-
i^-r^N-r^cDCDcocOLOLOLOLnuo^r^f-^r^rrocDcoco
X

o Oh'^t-r-C3C£)rooaDLOCcjcr)r^'^T-oocDcoor^^i-CcjCT3CDoO'^cOLOCMOi^
>> rcjT--r--^ooooa)a3a3oococoooi^r--r^i^-cDcDcDLnLnLnLO^r'cr^r^}-oo
X

0? OOin-r-t^-COCT>LnC\JCO^OCOC\J
6 ^r^r^cDcnininin^j-^-^-c'OC'O
TABLE B-5

O ooo<X3->crT-a3r-^}'CooN-LoroococDcO'r-a3<x)^tc\ja3i^Lnc\jocoLOco-^cocci^-^03r^-'crcM
cM-^'r-’-^-oooooa3a3 0')C3cooococ30[^r^r^h-ci3cococr)coLnLnLnLn^r^r^'cr<?ococooo
Q i •* •» i i •» i i i y

LU
cr
5 °
< o CT3ooN-cDLr3^-co(Aji-ocj)oot^coLri^(?o<MT-oa3cor^cDLO^rooc'j->-ocj>oor^cDLn^rcocMT-o
cr c/3 corocooo(r)rocorO(r3rococ\)c\jc\jc\jc\ic\jc\ic\iC\i-^-^-^T-T-T—i—t-t-t-

390
 MMPI-2 T Score Conversion Tables 391

Linear T Score Conversions for Si Subscales

Males Females

So S,2 Si3 Si, Si2 S.3

17 86 83
16 83 80
15 80 77
14 77 77 73 74
13 74 74 71 72
12 71 71 68 69
11 68 68 65 66
10 65 65 63 63
9 62 62 60 60
8 59 71 59 57 74 58
7 56 67 56 55 69 55
6 53 62 53 52 65 52
5 51 58 50 49 60 49
4 48 54 47 46 56 47
3 45 49 44 44 51 44
2 42 45 41 41 47 41
1 39 41 38 38 42 38
0 36 37 36 37
 > LU o O) CO h- CD LO CO CM T— o CT) 00 n- CD LO CO CM T- O CT) 00 CD LO CT CO CM 1— o
t cc CD LD LO LO LO LO LO LO LO LO LO ct CT CT CT CT CT CT CT CO CO CO CO CO CO CO CO CO CO
QC O
o
cn
ct CO CM o CT) CO h- CD LO CT CO CM o CT) CO N- CO LO CT CO T— O CT) 00 CD LO CT CO T— o
CO o O CT) CT) O) CT) O) CO 00 CO 00 CO CO CO CO 00 N- c- C- C- h~
CL O o CT) CT) CT) CT)
T~

,,_
\y
_v_
h- CD Td" CO o 00 CD LO CO CM O CT) 1^- CO CT CO
o O o o o o CT) CT) CT) CT) CT) CT) CO 00 CO CO 00
CL 1— 1—
y— 1 T~ T—

LL _ _ t'- LD CM
CT O) CD CO CO CD CO O 00 LO CO O
CD h- h- CD CD CD CD LO LO LO LO CT CT CT CT CO CO CO

O CO CD LO CO O 00 h- LO CO CM O CO r^- LD CT CM
CO h- n- r- h- CD CD CD CD CD CD LO LO LO LD LD
CD

_
O CO N- LO CT CM T~ o 00 LD
CT) CT) CO 00 CO 00 CO CO CO

<D l''-
CC N-

o
o
CD
Cl)
ro I
E 6
a)
LL z
cc
>
z
cc
1—
Linear T Score Conversions for Supplementary Scales for Males and Females

CO
LL

CC

6 o 0) CD CT r~ 00 LO CO O r^. CT CM CT) CD CT 00
< CM T— T- i— O O o O CT) CT) CT) CO CO CO 00 N-
T~ 1- 1- 1- T* * T~ T—

CO CD CM O CO CD CT CM o 00 CD ,_ CT) N- LO CO CT) h- LD CO
CT
LU 00 CO 00 CO r- L- N- L~ CD CD CD CD LO LO LO LO LO CT CT CT CT CT

^_
CT CM CT) CD CT CO CD
QC O O CT) CT) CT) CT CO 00
T— T—

^_ CO LD
LO CO CM O CD CO
< CO CO CO CO CO r^- N- 1^- N-

CM o CO CD CO CM CT) CO h- CD CT CO CM CT) 00 CD CT CO CM CT) 00 CD CT

00 T— T— -T- r^- l^~ h- h-
n o o O O O O o CT) CT) CT) CT) CT) CT) CT) CT) CO CO 00 00 00 00 00 <5

CO CM O CO f'- LO CO CM O CO h~. LO CO CM O CO
Z T—
n 1— 1— O o o o O O CT) CT) CT) CT) CT) CT) CO 00
T— T— T_ T— T_ T—

LL -_ N- _
O CO LO CO CT) LO CO 00 CD CT CM O CO CD
CD CT) CO 00 00 CO N- 1^ h- h- CD CD CD CD CD LD LD

CT) CD CT CM o 00 CD CO CT) 1^ LO CM o 00 CD CT
CD CD CD CD CD CD LO LO LO LO CT CT CT CT CT CO CO CO

CD LO
__ r— LO CM
^_
CT)
CO O CO CT
z CT) CT) CT) CT) CT) CO 00 00 CO CO CO h-

CD CD
CC h-

o
Q

CO
0) I
co
6

Z
CC
>
z
cc
1—
03
LL

cc
1
o CO CT) CD CT CM CT) K LO CM O 00 LO CO 00 CD CT CM CT)
< i— T— O O O O CT) CT) CT) CT) CT) CO 00 00 co h- C'. r^- h- CD
1 -T_ 1 *— 1- T—

to CO ,,_ 00 CD CM CT) h- LO CO O ao CD CT
_
CT) LO CM o 00 CD CT
LU CO CO h~- h- Is- N- CD CD CD CD CD LO LO LO LO CT CT CT CT CT CO CO CO
TABLE B-7

CC 00 CD CT CM CT) 1^. LD CO
CT) CT) CT) CT) CO CO 00 00

^_ _
< CT) CO LO CT CM O 00
CT) 00 CO CO 00 00 00 CO 00 N-
LLI
cc
§ o
< o o CT) CC' h- CD LO CO CM O CT) oo N- CD LO CT CO CM O CT) 00 CD LO CT CO CM o
CC CO CD LO LO LO LO LO LO LO LO LO LO O" CT CT CT CT CT CT CT CO CO CO CO CO CO CO CO CO CO

392
<y> CO Is- CO in O’ CO CM O 05 00 Is- CO in O’ CO CM O O) CO Is- CO in O’ CO CM ,— o > UO
CM CM CM CM CM CM CM CM CM CM T— T— T- 1— T— T— T— y— 9 cc
cr O
O
cr>
05 00 Is- CO O’ CO CM y— O O) 00 Is- co m O’ CM r- O (J) CO h- CO m CO CM o 05 CO Is- co
CO co CO CO CO co CO CO CO in in m in in in in m m O’ O’ O’ O’ O’ O’ O’ CO CO CO Cl

o 00 Is- in O’ CM 05 CO co O’ CO ^_ o 00 Is- m CM _ 05 CO to tn CO CM o CD Is-

CO 00 Is- Is- Is- Is- Is- Is- CO CO co CO co CO CO in m m m in m O’ O’ O’ O’ O’ O’ O’ CO CO 0-

o LL
CO O

o 05 Is- in •o CM 05 Is- CO O’ CM
m o O- o O’ O O’ CO CO CO CO CO CO CD

■o CM 1_ o CO Is- m O’ CM __ o 00 Is- m O’ CM o 00 r- in O’ CM T— o CO Is- in CM

—
r- Is- Is- r- CO CO CO CO CO CO CD m m m m m in m •o O’ O’ O’ O’ CO CO CO co CO

O’ T- 00 m CM 05 co CO O Is- O’ 1— CO in CM o 05
r-- Is- CO CO CO in m in in O" O’ CO co co co CL

O r- CO o CO CO 05 CO CO 05 co CM 05 m C\J o O
CO r- Is- r- CO CO m in in o O’ O’ CO co 00 co Q
cn
05
co 05 co CM 00 m ,_ Is- O’ o co CO 05 in CM CO O’ __ Is- co o 05 X
o 05 05 05 co 00 00 Is- Is- Is- CO CO m in in O’ O’ co co CO
1— E 6
05
LL z
o CO o- o co CM CO O’ o to CM 00 o to CM CO o CO CM 00 'sT o
CM 1— •t— T— o o 05 05 05 00 CO Is- Is- CO CO in in m o O’ co CO CO CL
s— ’r_ T~ s— >
1— 1— I— 1— 1— i— 1— 1— I— 1— o LL La¬ u_ LL lo¬ lo¬ LO¬ LL LO¬ z
o CO i— co m 00 o CO LO CO in CO in CO o co in CO y— CO
CM y— T— o 05 00 CO h- CD in in co h- CO CO 05 O y— QC
T_ y— T- T_ T— 'T— 1—
m
o CO CM CO m r-- CO 05 LO T— Is- O’ o CD CM CO O’ o CD CM u_
CM T— T” o o o 05 05 00 CO CO Is- i^ Is- CD to m in cn O’
T T— T- CL

,_ _ 6
uo CO O Is- O’ CM 05 CO CO CO m CM o O’ O
Is- Is- Is- co in <
CO CO in m m O’ O’ O o co CO CO CO

05 r- un CO o cn
CO CO CO CO CO CO LOJ

CO . co in CO o r- in CM o r- _
o CM 05 co O’ O) CD co o a:
CO CO Is- Is- Is- Is- co co CO CO in m m o O’ O’ O’ CO CO co CO CO

CM 05 CO Is- CO O’ CO CM 05 co r-. CO O’ CO CM o a) CO Is- in ■o co CM o 05 co h- Ln <

r— Is- CO co co CO CO CO CO CO in m m m m LT; in in o O- o O’ O’ o co co CO co

co CM 05 CO Is- CO O’ CO CM 05 co Is- co¬ O’ co CM T— 05 CO Is- CD CO 0-J 05 CO r- 00

Is- Is- Is- co CO co CO co CD CO to m in m in in in in m o O’ O- O’ O’ o O’ CO CO co CL

m CO OvJ o CO Is- in CO CM o co r- in co CM o co Is- in CO CM o 00 r- LO CO CM o CO r- z

CO CO CO CO Is- r- Is- Is- Is- co co co co co co in m m m m in O’ O’ O’ O’ O’ CO co CL

O’ 05 Is- m co T“ 05 Is- O" CM o u_

in in O’ o- o O’ O’ CO co CO CO co CD

o
CO co CD

T_
Is- co O’ CO T- o CO Is- m O’ CM o 05 r- 10 o co o 00 to in CO C\J o 05 to CM
Is- r- r- Is- Is- Is- CO CO CO CO CO CO m in m in m in 1/5 O’ O’ O’ O’ o CO co CO co CO

CO O co m CO o Is- m CM o ’—. m CM 05 Is- O’ CM o 05

Is- Is- co CO CO CO in m in in O’ O’ O’ co CO co co co CC

CO m CM CO m ^_ CO in T— co in i— co O’ o
__ o
h- Is- co CO CO m in in O’ o co CO CO co Q
cn
co 05 CO CO 05 co CM 05 CO CM 05 m CM 05 LO CM co in T— CO in — o 05 X
O 05 05 05 00 00 CO Is- r- Is- CO co CO 1/5 ID in O’ o CO co CO co
2
ro
6
o 00 m y— co O)
Is- co CM CO o CO CO 05 in T— Is- o CO CM CO y— z
CM t— T~ o o 05 05 05 00 CO co Is- Is- 10 co co m LO m O’ o CO co CO cc
T_ ’r_ T“” ’r~ >
1— 1— h- 1— I— 1— l- 1— 1— 1— o LL LL LL LO¬ lo¬ LO¬ LL LL LL z
o O’ o CO co 05 CM in I-. in r- O T— CO in OM 05 r-- O’
CM T— o o 05 co Is- Is- co m m CO h- Is- CO 05 05 o ’— cc
T~ ’r_ ’r_ s— 'T— 1—
CG
o CO CM 00 O’ o <o CM Is- CO 05 in h- co 05 in T- CD CM u_
CM ■1— 1- o O o 05 05 co CO Is- Is- Is- CD CO in in m O’
T_ 'r~ 1 T— CL

r- in CM o CO in CO 1_ CO CO O’ _ 05 Is- O’ C\J o
6
<
co co CO CO in m in in o O’ O’ o CO co CO CO CO

o cn
CO co UO

T“ co CO O’ CM 05 Is- m CO CO CO O’ CM o f^ in CO y— 05 to O" CM O CL
CO r- Is- Is- r— CO co CO CO CO in in m in m O’ O’ O’ O’ CO CO CO CO CO

Is- in O’ co o co r- in O’ CO i— o co CD o CO i—
Is- o 05 Is- CO CO CM o 05 h- CO <
Is- Is- Is- Is- Is- Is- co co co CO CO CO co m m LO in m LO in O’ O’ O o O o co CO CO
LL
cr

T— i 9
05 CO Is- co m o co CM r— o 05 00 Is- co in CO CM o 05 CO Is- CD m O’ CO CM y— o < o
CM CM CM CM CM CM CM CM CM CM T— QC CO

393
 > LLJ CD uo D" CD C\J y— o cn CO CD UO D" CD CD T— o <n 00 r- CD UO D- CD CD O CT) 00 e'¬ CD UO CO CD o
< cr CD CD CD CD CD CO CD CD CD CD CD CD CD CD CD CD cm T- 1— T— T— y “ T— T— y— 1— T—
cr o
O
CO
1— CD O X UO CD cn h- •^T CD CT) e- ■D- CD CT) e- D" 1- CT) r- en CD 1— cn CO CD CT) un
QI O O cn cn cn CO CO CO CO e- e- e— e- CD CD CD CD UO uo uo UO uo "M" O' D- CD CD
1— ’

CT) f— uo CD o CO CD D" CD o CO CD CO y— cn r- uo co CT) e-- uo "D" CD y— o 00 CD uo CO o X D" y—

cc CD oo cn cn cn CO CO CO CO CO X f~- e'¬ h- CD CD CD CD CD UO uo uo UO UO UO uo D" D- O' O' CD CD cn
$
_
cn CD D" en co CD CO U0 CD O CO uo CD o n- uo CD o X un CM CT) CD CM
< cn cn cn cn CO CO CO 00 e— r- r- e'¬ CD CD CD CD uo uo UO UO O' O' CD CD cn
u_

o f-. D" CD o uo CD o co uo en o CO CD D- CD 1— CT) CO CD O’ y— CT) un CM

o CO CO 00 CO h- Is- e- e- co CO CD CD U0 U0 UO uo UO D" •D- O’ O' CD CD CD
00

LU h- D" CD cn CO D" y- co co CD o CO UO CD o e- uo D- CD cn n- D- o en
00 cn cn cn co CO CO CO i-- e^ X e- CD CD CO CD uo uo UO uo UO O' D" D- CD
_l

< D- o CD T_ e- CD CT) o CD CD o CO en CD T— CO CD CD o
CL cn cn co CO h- e- CD CD CD UO UO uo D" D" ^r CO CD CD CD
t—

CL CO ^r — co uo CD CT) uo CD CT) CO CD CT) CO D" CD CT) n- LO CM CT) CO CD

CO cn cn cn co co co r- h- n- CD CO CD UO UO UO UO 'cr O' CD cn CD
cn <
0
CO z CD o n- LO CD cn r- D" T— 00 CD rcr CD o CO X CO d- CM o CO en CD
E co CO n- e- CD CD CD CD uo UO uo UO uo uo D" 'D' D" "D" O' O' CD CD CD
o o
LL
CD CO ■cr o CD CD CO D" o CD CD o LO CM CT) CO
z CO 00 00 r- e- CD CD CO uo uo uo D" O' o- CD cn CD
<
CD o co LO CD CT) CD CD o e- ■'t CT) CO CD o f- T- CO co CD h~ CT)
N T— T— o o O CT) CT) CT) CT) co co co n- e-. r- n- CD CD CD LO LO en CD
CD "r_ T_ ,~ 'T—

< 1"- uo CD T_ o CO cd CD O cn e- uo CD y- cn h- CD D- CD O co CD D" CD CT) n- uo CD y— CT) CO CO o CD CD

LU o o o O o cn cn cn cn cn 00 co co 00 CO Is- h- Is- Is- e- r-. CO CO CO CD CD uo uo uo UO uo D" O' O' 'cr CD CD
'r~ 'r~ 'T
X

CL e'¬ uo CD CD o CO X uo co CD o CO r- LO CD CD o co h- UO CD CD O CO e'¬ uo "D- CD O CO LO CD CT) D"

LU en cn cn cn cn 00 CO CO co co 00 e— Is- e- Is- i-- CD CO CD CD CD CD UO en uo uo uo UO O' O' D- CD CD
O

00 e^ CD CT) U0
_ r- CO O) CD CD o 00 CO O'
_
X CD
CD co CO r- e- e- co CD uo uo uo uo D" O' O' D- CD CD
O
1_
CO CO D" _ co uo ,_ co UO CD co uo CD CT) CD CD ^_ CO CO CO ^_ co en ^_
cc o cn cn CO co co CO r- e- r- CD CD CO UO UO uo uo D- D" O' O' CD CD CD
LL

X cn CD ■^r y— cn CD "nT T— CT) CD D" — CT) CD uo CD O) e— LO CO o X

z CO CO CO CO Is- Is- h- CD CD CD CO U0 UO uo uo uo D" •D- O' O' D" CD CD
<

1— D" cn CD D" CO CD CT) CD CT) CD D" CT) CD D- CD cn h- CD CT) un

cc o o cn cn cn CT) CO CO co 5 e- e- e- n- CO CD CD 5 uo uo UO en O' O' D- CD CD
1— 'r~
Uniform T Score Conversions for Content Scales

X ^_
cc CO CD d- CD o cn X uo CD _ cn CO CD O' CD o CO e- uo C0 CT) e- CD D" CD o CO CD O’ CT) CD CD
cn cn cn cn cn CO CO 00 CO co X X Is- n- Is- e'¬ CO CD co CO CD UO uo UO UO uo uo O' O' CD CD CD
<:

uo CD cn X ^r _ 00 en CD O e- ,_
CO CD CD O e- uo CD o O- __ X CD
O O cn cn cn cn 00 co CO CO e- r- X CD CD CD CD uo uo UO uo O' O- O' CD CD
l£ 1

Q
cn CD D" y- CO CD CO co uo CD o 00 uo D- CD o CT) e- uo CO T— CT) un CD
o 00 CO CO CO n- r- e-. Is- CO CO CD CD UO uo UO UO uo D' D- D- O’ O' CD CD CD
00
,_ ,_
LU CO CD CO co uo CD o e- uo CD O n- CD CT) e— UO CD ^_ CO en ^_ en
00 o cn cn cn cn co co CO co e- h- n- X co co CD uo UO UO UO LO O- D" CD
L
TPA

cn uo r- CD CO D" O CD CD O CO CD CD co CD CD O
co 00 CO h- X CO CD CO UO UO UO D- 'cr D- O' CO CD CD CD

CL
00 D" o CD cn CD CD CT) uo CD CO uo CD 1- CT) X co D- C\J o X "Cf O
cn cn CO co e- r~- e-- CD CD CD UO uo UO uo D- D" D" 'CT O' O' CD CD CD
<
C/)
0 z
CD o Is- O' 00 uo CD CT) CD D" CD CT) CO r- CD C0 o CO en CD
CO X
CO CO X n- e- CO CD CD UO U0 UO UO uo D" D" D" M- D" D- O' O' CD CD CD
o
2
CD
"7
x_
CD CD CO •D- O e- CD CT) CD CD O CO CO CO o CD CD
CO CO r- X r- CD CO UO uo uo UO ■D- O' O' CD CD
<

N o D uo CM 00 uo _ 00 D- ^_ CO D- y_ e-. D- o e- CD O O' ,_ CD CT)

CD T- T— T— o o o CT) CT) CT) CO CO 00 e— e-- e'¬ CD CD CD un LO en CD
CD T~ 1 T_ T_

< C\J o CO CD uo CD ,- cn x uo CD y- cn 1"- U0 CD ^_ o co CD D" CD O CO CD D" CD o er) CD CD ,_ co O' y_ e— CD

LU T— T- o o o O o cn cn cn cn cn 00 CO CO CO CO co Is- e- e- e- n- CD CO CD CD CD uo cn UO UO O' O' D" CD CD
I 1 T_

CL o CD X uo d- CD O CO 1"- uo C0 CD O CO Is- uo CD y_ o 00 CD uo CD _ CT) CO CD UO CD ,_ co en ^_ CD

LU o D cn D cn cn cn co 00 00 CO CO co h- Is- Is- r- n- r- CD CD CD CD CD uo uo uo uo UO LO O' ^r O' CD
o T_

CO
CD n- o e- CD O CD CD CT) CD CD O h- O' __ h- CD
co 00 CO e- r- e- CO CD uo uo UO UO O' O' D" CO CD
o
00 C0 o C- CO o X CD o e- O r- D" o X ^r o
^_ ^_
TABLE B-8

co un en
cc T— 1— o o o CT) CT) CT) 00 CO CO e- e- e- CD CD CD uo UO LO O' ■cr O' CD
ll 1- T— T— T— -T—
X
z CD o Is- LO CD o Is- UO CD o e- uo CD o e- uo CD CD o Is- LO CD cn un
< cn cn co CO CO CO e- X e- e— co co CD CD uo uo uo uo uo O' O' "Cf co CD
111
cc
§ o
< o CD uo D’ CD CM o cn CO e- CD uo D" CD CD o cn 00 Is- CD uo CD CD O CT) CO r- CD uo O' CO CD o
a; co CD CD CD CD cn CD CD CD CD CD CD CD CD CD CD CD C\J 1— i— T— T- T— T— T- T“

394
 APPENDIX C

MMPI-2: Critical Items

TABLE C — 1 Koss-Butcher Critical Items Sets, Revised

Acute Anxiety State Depressed Suicidal Ideation

2F 28T 208F 463T 9F 92T 233T 41 IT 520T

3F 39T 218T 469T 38T 95F 273T 454T 524T
5T 59T 223F 65T 130T 303T 485T
10F 140F 30 IT 71T 146T 306T 506T
15T 172T 444T 75F 215T 388F 518T

Threatened Assault Situational Stress Due to Alcoholism

37T 125F 51 IT
85T 264T 518T
134T 487T
213T 489T
389T 502T

Mental Confusion Persecutory Ideas

24T 180T 325T 17T 144T 241T

31T 198T 42T 145T 251T
32T 299T 99T 162T 259T
72T 31 IT 124T 216T 314F
96T 316T 138T 228T 333T

Source: From the Minnesota Muitiphasic Personality Inventory-2 Manual by J. N. Butcher, W. G.

Dahlstrom, J. R. Graham, A. Tellegen, & B. Kaemmer, Appendix C, pp. 85-86. Copyright © the
University of Minnesota 1 942, 1 943, 1 951, 1 967 (renewal 1 970), 1 989. This Manual, 1 989.
Reproduced by permission.
395
396 Appendix C

TABLE C-2 Lachar-Wrobel Critical Item Sets

Anxiety and Tension Depression and Worry

15T 261F 463T 2F 75F 273T

17T 299T 3F 130T 303T
172T 301T 10F 150T 339T
218T 320T 65T 165F 41 IT
223F 405F 73T 180T 415T

Sleep Disturbance Deviant Beliefs

5T 47 IT 42T 162T 333T

30T 99T 216T 336T
39T 106F 228T 355T
140F 138T 259T 361T
328T 144T 314F 466T

Deviant Thinking and Experience Substance Abuse

32T 298T 168T

60T 307T 264T
96T 316T 429F
122T 319T
198T 427F

Antisocial Attitude Family Conflict

27T 240T 21T

35T 254T 83F
84T 266F 125F
105T 324T 288T
227T
 MMPI-2: Critical Items 397

TABLE C-2 continued

Problematic Anger Sexual Concern and Deviation

85T 12F 268T

134T 34 F
213T 62T/F
389T 121F
166T

Somatic Symptoms

18T 47F HIT 176F 255F

28T 53T 142F 182T 295F
33F 57F 159F 224F 464T
40T 59T 164F 229T
44T 101T 175T 247T
 APPENDIX D

Prototypic Scores
for Specific Codetypes
in Psychiatric Settings

399
400 Appendix D

TABLE D-1 Prototypic Scores for Spike 1 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 76 119
Age 46.5 14.1 45.4 14.3
Men 86.8% 82.4%
Women 13.2 17.6

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 2.1%
Total (Obvious-Subtle)3 12.9 54.1 24.6 53.3
Critical items13 24.4 8.9 23.5 8.2
Overreportedc 1.3% 1.7%
Underreported0 34.2% 33.6%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.3 5.9 4.2 5.6
L 55.4 9.3 59.1 12.2
F 56.3 6.1 51.8 7.7
K 54.7 8.8 50.4 9.8
KHs) 74.0 4.7 69.3 4.0
2(D) 61.0 6.7 57.3 6.3
3(Hy) 62.8 4.4 56.3 5.6
4(Pd) 60.5 6.5 53.4 7.6
5(Mf) 51.9 8.4 43.6 8.8
6 (Pa) 55.1 6.6 49.0 7.8
7(Pt) 53.6 6.7 47.5 7.0
8(Sc) 55.8 7.3 49.4 6.8
9(Ma) 56.2 7.7 48.7 6.7
0(Si) 52.8 7.4 49.8 6.9

Supplementary Scales
M SD M SD
A 48.2 8.3 49.6 8.5
R 64.8 12.3 54.3 10.1
MAC/MAC-Rd
men 21A 4.5 27.0 4.5
women 23.6 2.6 22.8 3.6

Codetype Concordance
Men Women Men Women
MMPI-2 Spike 1 88.1% 52.2% MMPI Spike 1 64.1% 60.0%
1-2/2-1 21.7 Spike 4 30.0
1 -3/3-1 13.0

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 401

TABLE D-2 Prototypic Scores for 1-2/2-1 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 347 309
Age 46.8 1 1.7 46.5 12.3
Men 85.9% 70.2%
Women 14.1 29.8

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 12.2% 5.3%
Total (Obvious-Subtle)3 78.6 59.6 98.2 64.9
Critical itemsb 40.0 12.7 38.6 13.2
Overreportedc 17.7% 33.1%
Underreported0 8.5% 7.1%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.7 6.2 4.6 6.0
L 52.5 8.7 55.7 1 1.9
F 62.1 9.2 62.4 13.7
K 49.6 8.2 43.4 9.8
KHs) 86.9 10.6 78.8 9.1
2(D) 88.3 1 1.0 80.0 9.8
3(Hy) 75.6 9.4 68.2 10.6
4(Pd) 69.3 10.4 59.5 10.6
5(Mf) 56.4 9.0 47.2 9.6
6(Pa) 62.5 10.3 56.6 1 1.9
7(Pt) 71.5 12.1 64.3 12.0
8(Sc) 68.3 13.5 59.7 12.4
9(Ma) 56.4 10.1 49.4 9.5
0(Si) 63.3 9.4 62.1 9.6

Supplementary Scales
M SD M SD
A 62.0 10.5 64.0 11.1
R 71.4 12.7 60.7 1 1.2
MAC/MAC-Rd
men 26.9 4.8 26.8 4.9
women 21.4 4.1 21.3 4.5

Codetype Concordance
Men Women Men Women
MMPI-2 1-2/2-1 81.7% 44.1% MMPI 1-2/2-1 59.5% 82.0%
Spike 2 14.0 1-3/3-1 24.1
2-4Z4-2 1 1.8

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1 979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
402 Appendix D

TABLE D-3 Prototypic Scores for 1-3 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 164 241
Age 46.6 13.1 45.8 12.4
Men 62.8% 68.9%
Women 37.2 31.1

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 6.5% 5.7%
Total (Obvious-Subtle)8 25.3 61.0 61.2 71.0
Critical items6 35.1 12.7 35.9 13.6
Overreportedc 3.9% 1 5.3%
Underreported0 35.0% 22.5%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 5.1 5.8 5.0 6.0
L 56.1 9.6 60.0 13.5
F 59.6 8.4 60.4 13.3
K 56.2 10.0 50.3 1 1.7
KHs) 88.4 9.5 84.2 8.7
2(D) 70.8 9.4 68.4 9.6
3(Hy) 81.6 7.0 78.8 8.3
4(Pd) 68.1 9.3 60.6 10.5
5(Mf) 55.3 9.9 48.0 10.0
6(Pa) 62.4 9.3 57.5 1 1.9
7(Pt) 64.8 9.1 60.5 10.8
8(Sc) 67.7 9.8 61.5 10.8
9(Ma) 59.1 10.0 52.0 9.6
0(Si) 55.1 9.0 54.3 9.1

Supplementary Scales
M SD M SD
A 52.6 10.7 55.4 1 1.8
R 72.0 14.4 60.3 12.8
MAC/MAC-Rd
men 26.3 5.1 26.5 4.9
women 22.1 3.7 22.8 3.7

Codetype Concordance
Men Women Men Women
MMPI-2 1-3 61.5% 78.7% MMPI 1-3 38.3% 64.0%
3-1 36.5 16.4 1-2 28.1
1-4 12.0

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
cPercentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 403

TABLE D-4 Prototypic Scores for 1-4/4-1 Code Types in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 140 99
Age 41.3 13.5 39.9 13.1
Men 80.0% 71.8%
Women 20.0 28.2

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 10.0% 10.7%
Total (Obvious-Subtle)3 14.8 60.3 40.4 67.9
Critical itemsb 31.4 12.5 32.4 13.5
Overreportedc 2.6% 8.1%
Underreported0 30.1% 27.3%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 5.5 6.9 5.0 6.7
L 54.1 9.9 57.6 13.2
F 59.9 8.8 61.6 13.7
K 57.3 10.0 52.0 1 1.2
1(Hs) 79.9 8.2 73.3 7.1
2(D) 67.5 9.3 63.1 8.2
3(Hy) 71.5 7.7 65.1 7.9
4(Pd) 81.1 6.5 74.4 6.8
5(Mf) 54.0 8.7 46.7 8.9
6 (Pa) 60.7 8.5 56.3 10.6
7(Pt) 62.3 8.3 57.1 9.2
8(Sc) 65.4 8.5 59.6 9.2
9(Ma) 60.5 10.3 55.6 10.4
0(Si) 52.8 7.7 51.1 8.8

Supplementary Scales
M SD M SD
A 49.9 10.3 53.1 11.1
R 69.2 13.3 56.2 1 1.5
MAC/MAC-R6
men 28.3 5.2 29.2 4.5
women 24.1 4.4 24.0 3.5

Codetype Concordance
Men Women Men Women
MMPI-2 1 -4/4-1 41.1% 53.6% MMPI 1-4/4-1 64.8% 53.6%
1-3/3-1 28.6 25.0 Spike 4 1 7.9
Spike 1 14.3 4-8Z8-4 12.7

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 7 5th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
404 Appendix D

TABLE D-5 Prototypic Scores for 1-5/5-1 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 13 18
Age 37.1 1 1.4 40.9 13.4
Men 76.9% 22.2%
Women 23.1 77.8

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 14.3%
Total (Obvious-Subtle)3 41.0 59.4 58.3 60.7
Critical itemsb 33.8 9.9 29.9 10.9
Overreportedc 0.0% 5.6%
Underreported0 12.5% 16.7%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 2.5 2.4 7.2 9.0
L 49.8 9.4 56.9 10.4
F 60.5 9.5 56.9 1 1.5
K 51.4 9.5 47.4 1 1.7
1(Hs) 76.8 5.0 71.1 5.4
2(D) 62.7 8.4 58.5 5.5
3(Hy) 67.5 6.1 58.7 7.5
4(Pd) 63.0 10.4 53.7 9.2
5(Mf) 76.7 3.5 73.1 6.4
6(Pa) 60.9 5.8 51.4 8.7
7(Pt) 60.9 9.1 53.9 6.9
8(Sc) 63.3 6.7 55.3 5.1
9(Ma) 60.3 6.6 51.2 6.5
0(Si) 55.0 9.2 51.7 7.4

Supplementary Scales
M SD M SD
A 55.8 9.4 55.7 10.1
R 58.9 8.6 56.8 12.0
MAC/MAC-Rd
men 25.2 5.8 26.0 3.5
women 29.7 3.2 24.7 3.7

Codetype Concordance
Men Women Men Women
MMPI-2 1-3/3-1 30.0% MMPI 1-5/5-1 75.0% 14.3%
1-5/5-1 30.0 100.0% Spike 4 21.4
Spike 2 14.3
Spike 5 14.3

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1 979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 405

TABLE D-6 Prototypic Scores for 1-6/6-1 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 26 96
Age 44.4 13.2 42.6 12.1
Men 61.5% 61.5%
Women 38.5 38.5

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 14.6% 15.7%
Total (Obvious-Subtle)3 101.9 61.6 128.6 68.3
Critical itemsb 46.5 13.6 47.6 14.9
Overreportedc 22.0% 49.5%
Underreported0 4.9% 3.1%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 5.9 6.9 5.6 6.4
L 51.3 6.9 53.7 10.6
F 70.8 11.0 11A 16.8
K 47.8 6.5 41.6 9.3
1(Hs) 84.9 10.5 79.3 9.8
2(D) 69.2 11.0 67.1 10.3
3(Hy) 71.3 9.2 66.8 12.5
4(Pd) 68.8 8.3 62.9 10.3
5(Mf) 53.7 11.3 47.4 9.7
6(Pa) 85.1 10.9 83.4 1 1.4
7(Pt) 67.0 12.6 63.8 12.8
8(Sc) 73.5 12.9 68.6 12.7
9(Ma) 61.8 10.7 58.9 11.1
0(Si) 61.0 7.3 56.6 8.8

Supplementary Scales
M SD M SD
A 62.7 10.1 64.4 1 1.0
R 64.8 12.9 52.0 10.9
MAC/MAC-Ra
men 27.6 5.3 28.2 5.0
women 24.0 4.5 24.9 3.6

Codetype Concordance
Men Women Men Women
MMPI-2 1-6/6-1 100.0% 80.0% MMPI 1-6/6-1 26.7% 21.6%
1-8/8-1 21.7
1-2/2-1 20.0
4-6Z6-4 27.0
6-8/8-6 21.6
6-9Z9-6 10.8

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
c Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
406 Appendix D

TABLE D-7 Prototypic Scores for 7-7/7-1 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 29 41
Age 40.9 13.5 40.2 12.7
Men 89.7% 80.5%
Women 10.3 19.5

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 17.1% 5.5%
Total (Obvious-Subtle)3 91.7 64.7 126.3 63.2
Critical itemsb 45.3 13.5 46.0 13.3
Overreportedc 26.8% 50.0%
Underreported0 7.3% 0.0%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 5.6 6.5 5.5 6.8
L 48.2 7.0 54.6 1 1.8
F 64.9 10.6 71.1 14.2
K 50.2 8.4 44.3 8.6
1(Hs) 86.0 9.6 80.8 10.5
2(D) 77.5 9.9 71.7 1 1.5
3(Hy) 75.5 9.5 70.3 12.3
4(Pd) 70.5 1 1.3 63.2 10.0
5(Mf) 61.1 12.6 48.8 12.1
6(Pa) 66.8 12.0 62.4 12.4
7(Pt) 85.1 8.9 79.8 9.8
8(Sc) 75.8 1 1.2 70.8 1 1.4
9(Ma) 62.3 10.7 55.5 9.1
0(Si) 62.6 9.2 60.2 10.2

Supplementary Scales
M SD M SD
A 66.2 10.7 69.2 9.9
R 65.0 15.0 53.9 12.9
MAC/MAC-Rd
men 27.1 5.0 27.3 4.5
women 25.0 5.0 24.5 5.2

Codetype Concordance
Men Women Men Women
MMPI-2 7-7/7- 7 55.6% 100.0% MMPI 7-7/7- 7 47.1% 37.5%
1-6/6-1 14.8 1-8/8-1 23.5
3-7/7-3 14.8 4-7/7-4 25.0

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
cPercentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 407

TABLE D-8 Prototypic Scores for 1-8/8-1 Codetypes in Psychiatric Settings

MMPI MMPI-2
Demographics
M SD M SD
N 186 117
Age 38.5 14.1 37.5 1 5.0
Men 84.9% 70.1 %
Women 15.1 29.9

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 36.8% 30.7%
Total (Obvious-Subtle)3 134.8 72.5 1 55.0 73.8
Critical items'3 58.6 16.0 55.4 1 5.6
Overreported0 41.3% 65.6%
Underreported0 5.2% 1.7%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 6.3 7.0 5.2 6.7
L 51.3 9.1 54.7 12.1
F 80.4 14.4 85.2 19.2
K 47.0 9.5 40.8 10.0
KHs) 92.4 10.7 82.9 10.6
2(D) 80.7 1 2.7 71.8 1 1.0
3(Hy) 78.1 10.1 71.1 12.9
4(Pd) 76.9 1 1.0 66.0 1 1.8
5(Mf) 59.4 10.8 50.0 9.0
6 (Pa) 76.0 1 2.8 68.7 13.5
7(Pt) 79.9 12.2 71.4 13.3
8(Sc) 95.6 13.8 83.5 13.1
9(Ma) 70.6 10.6 61.9 1 1.2
0(Si) 63.0 9.2 61.9 9.1

Supplementary Scales
M SD M SD
A 68.3 10.6 68.7 9.9
R 65.0 13.3 55.4 1 1.2
MAC/MAC-Rd
men 27.7 4.8 28.0 4.9
women 23.8 4.7 24.7 3.7

Codetype Concordance
Men Women Men Women
MMPI-2 1-8/8-1 84.3% 50.0% MMPI 1-8/8-1 43.8% 64.3%
4-8Z8-4 19.4 1-3/3-1 13.1 14.3

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 7 5th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
408 Appendix D

TABLE D-9 Prototypic Scores for 1-9/9-1 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 46 68
Age 40.3 12.1 40.6 12.2
Men 82.6% 75.0%
Women 17.4 25.0

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 4.2% 4.9%
Total (Obvious-Subtle)3 61.0 62.7 88.3 61.4
Critical itemsb 42.7 12.3 42.6 12.5
Overreportedc 12.6% 26.1%
Underreported0 12.6% 5.8%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.7 6.0 4.3 5.2
L 48.4 6.0 50.8 10.0
F 65.0 10.8 67.6 1 5.4
K 47.5 8.7 41.4 9.3
1(Hs) 80.6 8.1 74.9 6.7
2(D) 62.1 1 1.2 57.4 9.5
3(Hy) 68.5 7.2 61.5 9.0
4(Pd) 67.9 7.9 60.2 9.1
5(Mf) 55.9 8.8 45.8 9.2
6(Pa) 62.0 7.9 56.6 9.4
7(Pt) 61.2 7.7 56.4 8.3
8(Sc) 67.7 8.8 62.0 8.1
9(Ma) 80.8 6.9 11A 8.2
0(Si) 49.1 7.7 49.1 8.6

Supplementary Scales
M SD M SD
A 59.2 10.0 60.8 9.9
R 53.2 10.2 45.8 9.3
MAC/MAC-Rd
men 31.9 4.1 32.2 4.2
women 25.4 3.9 27.7 4.0

Codetype Concordance
Men Women Men Women
MMPI-2 1-9/9-1 81.6% 87.5% MMPI 1-9/9-1 60.8% 38.9%
8-9Z9-8 17.6 16.7
4-9Z9-4 1 1.8 22.2
Spike 9 16.7

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
cPercentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 409

TABLE D-10 Prototypic Scores for 1-0/0-1 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 5 19
Age 53.6 16.0 42.1 13.8
Men 60.0% 84.2%
Women 40.0 1 5.8

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 0.0%
Total (Obvious-Subtle)3 39.0 18.9 85.7 41.9
Critical itemsb 27.6 2.2 33.1 6.6
Overreportedc 5.3% 21.1%
Underreported0 10.5% 0.0%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 3.2 4.7 4.9 6.5
L 53.0 7.7 52.9 1 1.6
F 58.2 3.9 58.8 6.8
K 49.8 7.0 40.4 7.5
1(Hs) 73.4 6.5 71.3 5.8
2(D) 64.8 9.1 62.9 6.6
3(Hy) 61.6 6.7 57.6 7.8
4(Pd) 54.4 10.2 51.0 7.3
5(Mf) 54.6 12.3 46.1 8.9
6 (Pa) 53.4 10.6 50.9 10.1
7(Pt) 56.8 7.8 54.4 1 1.0
8(Sc) 59.4 4.8 51.2 8.0
9(Ma) 55.2 7.1 46.5 6.7
0( Si) 74.6 3.3 71.5 3.8

Supplementary Scales
M SD M SD
A 51.4 2.5 60.4 9.1
R 78.8 8.5 59.5 9.4
M A C/M AC-Rd
men 22.7 1.5 26.9 5.1
women 22.0 2.8 22.0 6.2

Codetype Concordance
Men Women Men Women
MMPI-2 1-0/0-1 100.0% 50.0% MMPI Spike 1 31.8%
1-2/2-1 50.0 1-2/2-1 27.3
1-0/0-1 13.6
Spike 0 50.0%

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 7 5th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
410 Appendix D

TABLE D-11 Prototypic Scores for Spike 2 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 228 217
Age 44.5 13.2 42.3 12.9
Men 81.6% 74.2%
Women 18.4 25.8

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 3.0%
Total (Obvious-Subtle)3 23.0 47.5 36.4 50.8
Critical itemsb 23.3 8.3 23.7 8.0
Overreportedc 1.8% 4.1%
Underreported0 26.8% 20.6%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 5.7 6.4 4.8 5.7
L 52.3 8.0 56.2 1 1.2
F 55.0 5.8 52.4 8.2
K 50.1 7.9 44.6 8.9
KHs) 56.0 8.0 52.9 8.4
2(D) 74.1 4.2 69.1 3.5
3(Hy) 58.3 6.4 51.5 7.7
4(Pd) 60.2 7.6 53.7 7.6
5(Mf) 53.4 7.4 45.1 7.9
6(Pa) 54.6 7.1 49.4 8.2
7 (Ft) 58.8 6.2 53.3 6.9
8(Sc) 53.8 8.1 47.4 7.6
9(Ma) 51.3 8.9 45.5 7.4
0(Si) 59.5 6.8 56.1 6.4

Supplementary Scales
M SD M SD
A 55.3 9.4 57.4 9.7
R 67.8 12.0 56.6 10.5
MAC/MAC-Rd
men 25.3 4.5 25.8 4.8
women 20.7 4.0 21.4 4.1

Codetype Concordance
Men Women Men Women
MMPI-2 Spike 2 53.1% 81.6% MMPI Spike 2 48.2% 40.8%
2-474-2 35.9 2-0/0-2 44.7 23.7
1-2/2-1 17.1

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
cPercentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 411

TABLE D-12 Prototypic Scores for 2-3Z3-2 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 172 264
Age 45.0 12.9 42.9 13.0
Men 54.1% 50.8%
Women 45.9 49.2

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 7.3% 5.0%
Total (Obvious-Subtle)3 22.5 62.1 68.4 67.9
Critical itemsb 30.9 1 1.6 35.5 13.0
Overreported0 5.8% 20.2%
Underreported0 35.6% 1 1.9%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.8 6.2 4.7 6.3
L 55.4 8.9 59.3 12.5
F 58.7 7.2 61.7 12.3
K 55.4 9.3 47.7 10.7
KHs) 72.2 8.3 71.9 8.9
2(D) 84.9 8.9 83.0 9.7
3(Hy) 80.2 6.9 80.8 9.2
4(Pd) 69.4 8.5 63.2 9.3
5(Mf) 53.2 10.3 47.0 9.5
6 (Pa) 63.6 8.6 59.6 10.8
7(Pt) 70.3 8.9 68.6 10.5
8(Sc) 65.2 9.9 62.4 1 1.8
9(Ma) 52.4 10.7 48.3 9.6
O(Si) 59.5 9.6 59.6 9.8

Supplementary Scales
M SD M SD
A 56.3 10.1 61.9 10.9
R 79.1 13.8 65.3 12.7
MAC/MAC-Rd
men 24.0 4.8 24.0 4.9
women 19.8 4.4 20.1 4.2

Codetype Concordance
Men Women Men Women
MMPI-2 2-3Z3-2 53.7% 53.8% MMPI 2-3Z3-2 77.4% 93.7%
2-4Z4-2 15.7 10.0
1 -2/2-1 12.7

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1 979) critical items endorsed.
Percentage of patients within this codetype scoring above the 7 5th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
412 Appendix D

TABLE D-13 Prototypic Scores for 2-4Z4-2 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 606 343
Age 38.1 12.5 36.9 12.4
Men 78.1% 66.5%
Women 21.9 33.5

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 6.3% 3.2%
Total (Obvious-Subtle)3 38.6 59.9 68.3 62.3
Critical itemsb 32.9 12.1 34.4 1 1.9
Overreportedc 5.9% 16.3%
Underreported0 22.8% 13.7%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.2 5.5 3.4 4.7
L 50.9 8.0 52.5 10.5
F 62.2 9.0 64.9 13.9
K 51.5 8.8 45.3 9.3
1(Hs) 61.1 10.8 58.1 10.3
2(D) 81.8 9.1 76.1 8.4
3(Hy) 65.7 8.8 60.1 10.1
4(Pd) 83.1 8.3 77.6 7.9
5(Mf) 56.3 10.1 48.1 10.1
6(Pa) 63.5 9.6 59.3 1 1.0
7(Pt) 68.5 9.6 63.8 10.3
8(Sc) 65.8 1 1.6 60.6 1 1.7
9(Ma) 57.8 9.7 51.3 9.0
0(Si) 60.0 9.7 58.9 9.4

Supplementary Scales
M SD M SD
A 60.1 10.8 62.7 10.8
R 68.1 13.1 57.4 11.5
MAC/MAC-Rd
men 26.9 4.7 26.8 4.9
women 22.4 4.5 22.7 4.4

Codetype Concordance
Men Women Men Women
MMPI-2 2-4Z4-2 45.9% 53.0% MMPI 2-4Z4-2 92.1% 61.7%
2-7/7-2 1 5.7 Spike 4 1 5.6
Spike 4 10.4 4-8Z8-4 13.0

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 413

TABLE D-14 Prototypic Scores for 2-575-2 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 43 26
Age 37.6 14.7 42.0 15.1
Men 93.0% 34.6%
Women 7.0 65.4

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 1.1% 3.7%
Total (Obvious-Subtle)3 47.8 60.8 71.3 55.2
Critical itemsb 32.3 12.2 30.1 1 1.2
Overreportedc 7.4% 19.2%
Underreported0 25.5% 3.9%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 3.2 3.9 3.7 4.0
L 48.6 6.0 57.0 13.0
F 58.9 7.1 60.0 8.8
K 47.3 8.2 44.8 7.3
1(Hs) 58.9 12.2 56.2 9.8
2(D) 79.8 9.5 73.9 7.6
3(Hy) 62.4 8.8 56.0 10.4
4(Pd) 64.0 8.8 53.4 8.8
5(Mf) 79.0 7.6 73.8 6.8
6(Pa) 60.7 9.1 54.4 1 1.3
7 (Ft) 67.4 9.7 60.7 1 1.8
8(Sc) 62.3 12.3 60.2 9.7
9(Ma) 53.7 10.0 45.8 7.4
O(Si) 61.8 9.3 60.5 8.4

Supplementary Scales
M SD M SD
A 63.3 10.3 62.2 1 1.4
R 64.0 1 1.3 61.5 9.9
MAC/MAC-Rd
men 22.2 5.1 19.1 4.0
women 24.7 2.3 21.6 3.4

Codetype Concordance
Men Women Men Women
MMPI-2 2-575-2 22.2% 100.0% MMPI 2-575-2 88.9% 17.6%
2-0/0-2 22.2 Spike 2 29.4
Spike 2 11.1 Spike 5 17.6
2-676-2 1 1.1

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
414 Appendix D

TABLE D-15 Prototypic Scores for 2-6Z6-2 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 86 192
Age 40.0 12.8 39.4 12.7
Men 66.3% 55.2%
Women 33.7 44.8

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 10.5% 10.4%
Total (Obvious-Subtle)8 89.4 65.4 1 19.8 63.4
Critical items*3 41.8 13.7 43.3 13.3
Overreportedc 19.3% 42.5%
Underreported0 1 5.8% 1.0%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 6.1 7.2 4.7 5.9
L 50.4 6.9 53.9 10.6
F 68.4 12.2 74.4 19.5
K 45.7 6.9 39.8 8.0
KHs) 63.1 13.7 61.8 13.3
2(D) 84.1 10.0 80.5 10.8
3(Hy) 64.9 10.5 61.8 13.2
4(Pd) 72.1 9.4 65.9 10.6
5(Mf) 56.0 10.2 47.7 9.3
6(Pa) 82.1 10.2 81.9 1 1.7
7(Pt) 71.9 10.3 69.1 1 1.0
8(Sc) 70.9 14.0 66.9 14.1
9(Ma) 58.2 9.5 51.3 8.9
O(Si) 66.0 9.0 64.7 8.7

Supplementary Scales
M SD M SD
A 66.7 8.6 69.6 8.7
R 70.8 14.3 58.5 12.3
MAC/MAC-Rd
men 25.8 5.2 25.8 5.1
women 20.7 4.9 21.3 4.3

Codetype Concordance
Men Women Men Women
MMPI-2 2-6Z6-2 34.6% 32.6% MMPI 2-6Z6-2 81.0% 96.6%
2-4Z4-2 21.5 10.5
2-8Z8-2 1 5.0 14.0
6-8Z8-6 18.6

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 415

TABLE D-16 Prototypic Scores for 2-7/7-2 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 391 418
Age 40.3 12.7 38.8 12.8
Men 75.2% 58.2%
Women 24.8 41.8

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 10.1% 4.5%
Total (Obvious-Subtle)3 82.1 50.6 111.1 55.2
Critical items'3 38.3 10.5 40.3 10.9
Overreportedc 21.4% 34.8%
Underreported0 4.7% 0.7%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.6 6.1 4.3 5.9
L 49.8 7.3 51.8 10.1
F 62.9 8.4 66.8 14.3
K 48.9 7.2 42.7 7.8
KHs) 65.9 1 1.5 63.5 11.1
2(D) 89.8 1 1.3 84.3 9.7
3(Hy) 68.0 9.7 64.5 11.5
4(Pd) 70.5 9.8 63.8 10.2
5(Mf) 57.7 10.4 48.9 10.4
6(Pa) 66.1 9.4 63.0 1 1.1
7(Pt) 85.2 9.2 82.0 9.2
8(Sc) 74.4 1 1.2 69.6 1 1.0
9(Ma) 54.4 9.7 49.1 8.8
01 Si) 68.9 8.2 67.8 9.0

Supplementary Scales
M SD M SD
A 69.9 8.1 72.8 8.1
R 71.7 13.0 60.6 1 1.3
MAC/M AC-Rd
men 24.5 4.6 24.5 4.8
women 19.6 3.5 20.0 3.7

Codetype Concordance
Men Women Men Women
MMPI-2 2-777-2 65.4% 93.9% MMPI 2-777-2 78.9% 52.0%
2-8/Q-2 12.0 10.7
2-414-2 10.2

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
c Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
416 Appendix D

TABLE D-17 Prototypic Scores for 2-8Z8-2 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SO
N 461 219
Age 35.8 12.8 35.7 12.7
Men 73.8% 53.0%
Women 26.2 47.0

Test-Taking Scales/Indexes
M SO M SO
Inconsistent 22.2% 11.1%
Total (Obvious-Subtle)3 137.2 68.0 155.0 73.0
Critical items*3 54.7 16.1 52.5 16.4
Overreportedc 45.2% 56.1%
Underreported0 3.3% 1.9%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SO
?d 5.5 6.7 5.2 6.1
L 49.1 7.7 51.7 10.6
F 79.9 14.6 88.4 20.7
K 45.7 8.0 39.5 8.4
KHs) 75.1 14.8 69.7 13.4
2(D) 95.4 12.1 86.3 1 1.5
3(Hy) 72.4 11.1 66.4 14.3
4(Pd) 79.2 1 1.4 69.2 12.4
5(Mf) 60.8 1 1.9 52.0 10.2
6(Pa) 76.6 13.5 70.7 14.9
7(Pt) 84.9 1 1.9 77.2 12.3
8(Sc) 97.2 14.2 86.8 13.4
9(Ma) 63.2 10.5 56.6 9.6
O(Si) 71.8 8.3 69.2 9.5

Supplementary Scales
M SO M SO
A 72.3 9.3 73.0 9.5
R 72.2 13.3 59.6 1 1.6
MAC/MAC-Rd
men 24.6 4.9 24.4 4.7
women 21.0 4.4 21.7 4.3

Codetype Concordance
Men Women Men Women
MMPI-2 2-8Z8-2 32.5% 53.3% MMPI 2-8Z8-2 94.8% 63.1%
6-8Z8-6 21.0 4-8Z8-4 22.3
2-7/7-2 20.5

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
c Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 417

TABLE D-18 Prototypic Scores for 2-9Z9-2 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 18 25
Age 39.3 1 1.0 39.5 12.2
Men 72.2% 64.0%
Women 27.8 36.0

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 8.8% 3.6%
Total (Obvious-Subtle)3 73.1 33.8 86.1 49.7
Critical itemsb 38.9 9.6 40.1 10.0
Overreportedc 2.2% 12.0%
Underreported0 24.2% 8.0%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.9 6.6 4.4 5.8
L 49.0 5.7 51.6 8.2
F 64.4 9.4 66.6 11.1
K 44.4 7.6 38.8 8.6
KHs) 58.2 9.3 56.1 10.2
2(D) 77.1 5.3 73.0 5.3
3(Hy) 62.5 9.7 57.9 10.7
4(Pd) 69.1 9.0 62.4 8.6
5(Mf) 56.1 9.2 50.0 9.7
6(Pa) 61.5 9.8 57.1 1 1.1
7(Pt) 66.6 8.4 62.6 7.6
8(Sc) 65.5 8.3 59.5 8.7
9(Ma) 79.5 6.8 74.9 7.1
O(Si) 54.5 8.3 53.2 8.7

Supplementary Scales
M SD M SD
A 63.4 7.7 66.5 9.5
R 58.9 10.2 48.2 10.7
MAC/MAC-Rd
men 29.5 2.3 28.6 2.9
women 25.8 4.0 25.6 4.1

Codetype Concordance
Men Women Men Women
MMPI-2 2-9Z9-2 83.3% 100.0% MMPI 2-9Z9-2 62.5% 55.6%
2-4Z4-2 25.0
8-9Z9-8 22.2

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
418 Appendix D

TABLE D-19 Prototypic Scores for 2-0/0-2 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 1 1 1 243
Age 40.7 13.0 41.5 13.0
Men 50.5% 72.4%
Women 49.5 27.6

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 1.4% 2.6%
Total (Obvious-Subtle)8 69.2 39.0 93.0 46.1
Critical items13 31.5 8.4 31.5 9.7
Overreportedc 5.1 % 20.5%
Underreported0 8.3% 1.6%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.1 6.4 4.0 5.4
L 49.9 6.1 51.4 8.9
F 59.7 7.4 59.4 1 1.6
K 44.5 5.8 38.9 6.8
1(Hs) 56.7 9.3 55.3 10.3
2(D) 79.4 6.9 75.3 7.3
3(Hy) 58.6 7.4 52.4 9.4
4(Pd) 63.1 8.3 56.2 9.1
5(Mf) 52.5 10.4 47.3 9.1
6(Pa) 59.7 9.2 53.6 9.7
7(Pt) 65.7 6.9 61.6 8.9
8(Sc) 61.0 9.4 55.0 1 1.0
9(Ma) 48.5 8.7 43.5 7.4
O(Si) 76.0 3.9 73.2 5.3

Supplementary Scales
M SD M SD
A 64.9 6.9 67.1 8.6
R 74.0 10.3 60.6 9.8
MAC/MAC-Rd
men 23.7 4.4 24.2 4.4
women 19.1 3.8 19.2 3.9

Codetype Concordance
Men Women Men Women
MMPI-2 2-0/0-2 82.1% 80.0% MMPI 2-0/0-2 21.9% 51.2%
Spike 0 10.7 Spike 2 30.0 20.9
Spike 0 1 7.4
2-4Z4-2 16.2
2-7/7-2 1 1.4

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 419

TABLE D-20 Prototypic Scores for Spike 3 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 44 66
Age 41.1 12.0 42.5 12.1
Men 40.9% 56.1%
Women 59.1 43.9

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 0.0%
Total (Obvious-Subtle)3 -39.4 41.4 -18.3 51.4
Critical itemsb 19.6 8.3 18.3 8.2
Overreported0 0.0% 0.0%
Underreported0 70.5% 60.6%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 6.7 7.0 5.0 5.7
L 54.0 8.9 59.1 1 2.1
F 51.8 4.5 49.5 8.2
K 57.6 8.1 54.4 10.4
KHs) 63.1 4.4 60.5 3.7
2(D) 59.1 6.9 56.4 6.5
3(Hy) 72.6 2.8 67.6 2.7
4(Pd) 60.7 6.9 53.8 7.2
5(Mf) 52.0 8.4 45.9 8.4
6(Pa) 56.0 7.4 49.7 7.7
7(Pt) 54.1 7.1 49.5 7.3
8(Sc) 54.7 7.2 48.8 7.5
9(Ma) 53.7 7.6 47.2 6.4
O(Si) 50.6 7.0 47.7 7.0

Supplementary Scales
M SD M SD
A 45.8 7.9 47.0 8.1
R 67.7 1 1.0 57.2 1 1.2
MAC/MAC-Rd
men 24.5 3.9 24.4 4.7
women 22.4 3.3 22.8 4.4

Codetype Concordance
Men Women Men Women
MMPI-2 Spike 3 100% 57.1 % MMPI Spike 3 48.7% 82.8%
1-3/3-1 28.5 3-4Z4-3 29.7
2-3Z3-2 13.5
Spike 4 13.8

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1 979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
420 Appendix D

TABLE D-21 Prototypic Scores for Spike 3-1 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 76 215
Age 44.6 1 1.7 43.3 12.8
Men 52.6% 59.1%
Women 47.4 40.9

Test-Taking Scales/indexes
M SD M SD
Inconsistent 6.5% 5.7%
Total (Obvious-Subtle)3 1.8 56.7 48.7 70.9
Critical items'3 30.4 1 1.9 35.7 14.9
Overreported0 0.0% 1 5.3%
Underreported0 56.6% 22.5%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 3.9 5.2 4.1 5.5
L 55.1 9.6 57.5 12.3
F 56.8 7.0 58.9 12.3
K 56.1 9.5 49.7 1 1.2
KHs) 77.0 6.2 79.0 9.1
2(D) 69.1 7.6 69.1 10.6
3(Hy) 82.0 6.8 85.1 9.9
4(Pd) 66.7 8.4 63.4 10.2
5(Mf) 51.9 10.0 47.6 10.1
6(Pa) 61.0 7.6 59.5 1 1.8
7(Pt) 62.7 8.3 62.6 1 1.8
8(Sc) 62.5 9.6 61.3 12.7
9(Ma) 57.3 9.6 52.2 9.7
0(Si) 52.3 7.3 52.4 9.6

Supplementary Scales
M SD M SD
A 51.8 10.5 57.6 12.1
R 71.6 12.3 59.5 1 1.4
MAC/MAC-Rd
men 26.5 3.8 26.5 5.2
women 22.5 3.8 22.7 4.1

Codetype Concordance
Men Women Men Women
MMPI-2 Spike 1 88.1% 52.2% MMPI Spike 1 64.1 % 60.0%
1-2/2-1 21.7 Spike 4 30.0
1-3/3-1 13.0

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1 979) critical items endorsed.
c Percentage of patients within this codetype scoring above the 7 5th percentile or below the 2 5th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 421

TABLE D-22 Prototypic Scores for 3-4Z4-3 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 168 1 74
Age 39.1 13.1 36.4 1 2.6
Men 52.4% 61.5%
Women 47.6 38.5

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 2.7% 0.5%
Total (Obvious-Subtle)3 -21.8 58.8 18.7 66.0
Critical itemsb 27.6 12.9 31.1 14.0
Overreported0 0.6% 7.5%
Underreported0 50.2% 36.2%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.4 5.7 4.0 5.3
L 53.9 8.9 55.9 1 1.7
F 57.4 7.1 59.5 13.0
K 59.4 10.0 53.1 10.8
KHs) 67.1 7.9 66.3 8.5
2(D) 66.2 7.9 65.3 9.2
3(Hy) 77.2 6.2 76.8 9.0
4(Pd) 82.2 7.4 78.7 8.1
5(Mf) 53.3 10.1 48.8 10.0
6(Pa) 62.5 8.2 59.0 1 1.3
7(Pt) 63.6 7.8 61.5 9.9
8(Sc) 64.4 8.8 60.8 10.7
9(Ma) 59.2 9.1 54.4 10.3
O(Si) 49.8 7.6 48.2 7.8

Supplementary Scales
M SD M SD
4 49.6 9.9 54.5 1 1.3
R 68.9 13.4 56.0 10.9
MAC/MAC-R6
men 26.8 4.7 26.9 4.1
women 23.2 3.6 24.3 4.1

Codetype Concordance
Men Women Men Women
MMPI-2 3-4Z4-3 65.9% 56.3% MMPI 3-4Z4-3 54.2% 67.2%
Spike 3 12.5 2-4Z4-2 1 5.0
1-3Z3-1 10.2 22.5 Spike 4 1 1.9
4-8/8-4 10.4

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1 979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
422 Appendix D

TABLE D-23 Prototypic Scores for 3-5Z5-3 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 12 9
Age 35.8 14.4 33.4 1 5.0
Men 91.7% 77.8%
Women 8.3 22.2

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 1.4% 0.0%
Total (Obvious-Subtle)8 - 14.3 50.3 -7.6 49.9
Critical itemsb 26.7 9.3 24.6 8.7
Overreportedc 0.0% 0.0%
Underreported0 64.8% 55.6%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 2.0 2.9 1.7 2.7
L 51.3 5.9 55.2 12.4
F 54.7 8.1 53.1 10.0
K 54.2 7.0 51.0 8.5
1(Hs) 66.4 4.8 63.0 5.2
2(D) 64.6 7.3 62.6 5.7
3(Hy) 73.3 4.5 70.4 6.4
4(Pd) 67.8 7.3 61.8 6.8
5(Mf) 78.4 5.8 70.6 4.1
6(Pa) 60.6 6.9 53.9 9.1
7(Pt) 59.7 6.4 53.0 6.7
8(Sc) 59.0 8.9 49.1 9.0
9(Ma) 60.8 8.0 48.0 6.4
O(Si) 49.4 9.4 45.3 6.8

Supplementary Scales
M SD M SD
A 51.6 7.5 51.6 7.3
R 63.4 9.9 54.2 12.1
MAC/MA C-Rd
men 24.8 3.9 24.4 4.4
women 25.0 — 27.5 3.5

Codetype Concordance
Men Women Men Women
MMPI-2 3-5Z5-3 54.5% 100.0% MMPI 3-5Z5-3 85.7%
Spike 3 50.0%
Spike 4 50.0%

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
c Percentage of patients within this codetype scoring above the 7 5th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 423

TABLE D-24 Prototypic Scores for 3-6Z6-3 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 22 64
Age 40.1 14.5 36.9 12.7
Men 31.8% 53.1%
Women 68.2 46.9

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 6.3% 12.2%
Total (Obvious-Subtle)8 37.5 60.7 85.9 74.2
Critical itemsb 36.5 1 1.3 43.6 16.7
Overreportedc 6.3% 28.1%
Underreported0 41.3% 12.5%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 5.3 7.4 5.3 6.9
L 52.9 6.7 55.3 1 1.2
F 59.1 8.5 72.0 20.0
K 49.4 8.9 44.5 9.5
1(Hs) 66.0 6.8 71.5 10.8
2(D) 65.9 6.6 67.7 12.2
3(Hy) 75.7 4.6 80.7 10.6
4(Pd) 68.6 7.7 68.1 10.3
5(Mf) 51.7 12.6 51.4 12.6
6 (Pa) 78.9 5.9 83.6 9.8
7(Pt) 61.8 7.0 64.3 13.2
8(Sc) 62.0 8.8 68.5 14.7
9(Ma) 58.6 6.9 56.5 10.3
O(Si) 57.0 8.0 53.1 8.6

Supplementary Scales
M SD M SD
A 58.4 9.6 62.5 1 1.5
R 66.1 10.0 55.3 10.3
MAC/MAC-R6
men 27.7 5.2 26.3 5.7
women 22.5 5.6 23.7 4.9

Codetype Concordance
Men Women Men Women
MMPI-2 3-6Z6-3 100.0% 80.0% MMPI 3-6Z6-3 20.6% 40.0%
1-3/3-1 20.0 2-3Z3-2 1 1.8
4-6Z6-4 26.7
6-8/8-6 20.0

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 7 5th percentile or below the 2 5th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
424 Appendix D

TABLE D-25 Prototypic Scores for 3-7/7-3 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 16 51
Age 37 A 12.8 36.3 13.0
Men 25.0% 49.0%
Women 75.0 51.0

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 3.8% 1.6%
Total (Obvious-Subtle)3 43.5 50.0 85.1 57.8
Critical itemsb 40.8 9.8 42.0 1 1.9
Overreportedc 0.0% 27.5%
Underreported0 34.6% 9.8%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.8 6.1 5.8 6.1
L 50.3 8.8 51.2 10.3
F 60.4 6.3 62.7 1 1.4
K 53.0 6.4 47.1 8.4
KHs) 74.6 8.9 72.6 10.4
2(D) 74.0 10.7 74.3 9.9
3(Hy) 83.8 7.6 83.2 10.1
4(Pd) 70.7 9.4 68.3 10.3
5(Mf) 49.1 1 2.0 47.9 10.6
6 (Pa) 67.6 10.4 66.1 12.0
7(Pt) 82.9 7.5 82.9 9.8
8(Sc) 75.3 8.5 73.0 10.3
9(Ma) 58.2 9.5 54.3 10.2
O(Si) 63.1 1 1.4 58.9 1 1.0

Supplementary Scales
M SD M SD
A 65.3 7.2 69.4 9.0
R 73.8 13.2 58.0 10.3
MAC/M AC-FId
men 26.3 4.5 27.1 5.7
women 20.4 4.1 21.2 3.6

Codetype Concordance
Men Women Men Women
MMPI-2 3-7/7-3 75.0% 66.7% MMPI 2-7/7-2 24.0%
2-4Z4-2 25.0 4-7/7-4 20.0 26.9%
1 -7/7-1 16.0
7-8/8-7 16.0
3-7/7-3 12.0 30.8
3-8Z8-3 19.2

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 425

TABLE D-26 Prototypic Scores for 3-8Z8-3 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 41 46
Age 34.2 12.6 32.7 1 1.7
Men 31.7% 69.6%
Women 68.3 30.4

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 15.7% 26.9%
Total (Obvious-Subtle)3 53.7 56.5 122.5 87.5
Critical itemsb 46.0 13.4 53.9 17.6
Overreportedc 10.0% 51.1%
Underreported0 24.3% 8.5%

Standard Validity and Clinical Scales (K-Corrected}

M SD M SD
?d 5.1 6.6 5.9 7.4
L 53.1 9.6 57.9 14.5
F 67.7 8.4 85.9 20.5
K 50.8 7.8 45.6 10.9
1(Hs) 74.9 9.1 80.1 12.2
2(D) 73.8 9.0 74.6 1 1.8
3(Hy) 82.6 7.1 87.8 1 1.5
4(Pd) 72.1 9.6 71.5 1 3.2
5(Mf) 53.1 13.0 54.2 1 2.5
6(Pa) 71.2 9.1 74.4 13.0
7(Pt) 74.2 8.4 76.6 12.6
8(Sc) 85.1 8.1 89.5 14.2
9(Ma) 65.7 7.6 65.1 1 1.0
O(Si) 57.9 7.6 58.0 10.2

Supplementary Scales
M SD M SD
A 61.7 9.9 66.5 1 2.7
R 67.4 12.2 57.0 1 1.4
MAC/MAC-R6
men 25.8 7.4 26.1 3.9
women 23.1 3.3 23.3 3.0

Codetype Concordance
Men Women Men Women
MMPI-2 3-8Z8-3 46.2% 39.3% MMPI 1-8/8-1 36.4%
3-6Z6-3 23.1 2-8/8-2 18.2
1-3Z3-1 32.1 3-8Z8-3 18.2 73.3%
3-7Z7-3 17.9 4-8Z8-4 13.3

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
426 Appendix D

TABLE D-27 Prototypic Scores for 3-9Z9-3 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 20 27
Age 35.0 10.3 37.9 12.7
Men 55.0% 70.4%
Women 45.0 29.6

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 8.8%
Total (Obvious-Subtle)3 14.5 48.7 42.7 46.0
Critical items'3 35.1 10.3 35.5 10.4
Overreportedc 0.0% 3.7%
Underreported0 62.0% 22.2%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 5.3 5.7 5.8 7.1
L 51.8 6.1 54.0 8.8
F 59.2 6.0 59.8 9.4
K 50.0 5.9 44.8 6.7
1(Hs) 66.9 6.7 64.4 6.3
2(D) 61.3 8.3 58.6 9.2
3(Hy) 75.6 4.8 74.1 7.0
4(Pd) 67.8 5.1 63.1 7.0
5(Mf) 55.8 9.6 51.4 9.7
6 (Pa) 58.9 7.5 55.8 9.2
7(Pt) 58.9 7.4 55.0 8.5
8(Sc) 64.1 7.9 59.1 7.6
9(Ma) 79.4 6.0 75.0 6.2
O(Si) 47.2 8.1 45.4 7.5

Supplementary Scales
M SD M SD
A 53.8 7.3 57.0 7.4
R 58.4 8.6 48.6 6.9
MAC/MAC-Rd
men 30.4 2.7 29.3 3.9
women 25.0 5.7 25.1 6.2

Codetype Concordance
Men Women Men Women
MMPI-2 3-9Z9-3 90.9% 44.4% MMPI 3-9Z9-3 52.6% 50.0%
1 -3/3-1 33.3 3-4Z4-3 10.5
4-9Z9-4 10.5 37.5

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 427

TABLE D-28 Prototypic Scores for 3-0/0-3 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M
N 4 2
Age 39.0 12.6 36.5
Men 0.0% 50.0%
Women 100.0 50.0

Test-Taking Scales/Indexes
M SD M
Inconsistent 0.0% 0.0%
Total (Obvious-Subtle)3 60.5 56.9 49.5
Critical itemsb 32.3 10.8 29.0
Overreportedc 0.0% 0.0%
Underreported0 45.5% 50.0%

Standard Validity and Clinical Scales (K-Corrected)

M SD M
?d 8.8 1A 9.5
L 52.3 1 1.0 61.5
F 62.5 5.3 63.0
K 48.8 6.2 46.5
UHs) 65.5 3.0 65.5
2(D) 69.8 4.3 67.0
3(Hy) 73.0 2.9 69.5
4(Pd) 57.3 5.3 55.0
5(Mf) 48.8 12.2 53.0
6(Pa) 59.8 3.8 60.0
7(Pt) 64.3 3.5 65.0
8(Sc) 63.8 7.9 59.0
9(Ma) 44.0 1 1.6 42.5
O(Si) 75.0 2.2 73.5

Supplementary Scales
M SD M
A 60.0 6.1 63.0
R 82.3 3.5 69.0
MAC/MAC-R6
men — — 16.0
women 19.3 5.3 1 7.0

Codetype Concordance
Men Women Men Women
MMPI-2 1-3/3-1 none 25.0% MMPI 2-0/0-2 100.0%
2-3Z3-2 25.0 3-0/0-3 100.0%
2-0/0-2 25.0
3-0/0-3 25.0

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1 979) critical items endorsed.
Percentage of patients within this codetype scoring above the 7 5th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
428 Appendix D

TABLE D-29 Prototypic Scores for Spike 4 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 752 529
Age 34.0 12.6 32.7 1 1.8
Men 74.6% 82.2%
Women 25.4 17.8

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 2.0%
Total (Obvious-Subtle)3 -36.0 52.2 -15.0 54.4
Critical items5 18.4 8.2 19.4 8.2
Overreportedc 0.1% 0.6%
Underreported0 71.4% 59.6%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.6 6.0 3.9 5.2
L 53.0 8.0 56.0 10.5
F 55.1 6.2 53.4 8.9
K 58.5 9.2 53.7 10.5
1(Hs) 53.1 7.0 50.2 7.8
2(D) 57.3 7.0 54.5 6.5
3(Hy) 58.0 6.6 51.0 7.3
4(Pd) 76.6 5.2 71.3 5.2
5(Mf) 52.3 8.2 44.3 8.7
6(Pa) 55.5 7.1 50.4 8.2
7{Pt) 55.2 6.6 50.0 6.7
8(Sc) 56.7 6.8 51.0 6.9
9(Ma) 57.6 7.3 51.2 6.8
O(Si) 49.3 7.5 47.2 7.4

Supplementary Scales
M SD M SD
A 46.0 8.3 48.4 8.7
R 62.8 1 1.2 52.1 9.4
MAC/M AC-Rd
men 27.6 4.1 27.7 4.1
women 23.9 3.7 24.1 3.7

Codetype Concordance
Men Women Men Women
MMPI-2 2-7/7-2 65.4% 93.9% MMPI 2-7/7-2 78.9% 52.0%
2-878-2 12.0 10.7
•
2-474-2 10.2

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1 979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 429

TABLE D-30 Prototypic Scores for 4-5Z5-4 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 1 13 45
Age 32.9 1 1.7 28.4 10.2
Men 87.6% 51.1%
Women 1 2.4 48.9

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 1.3% 3.8%
Total (Obvious-Subtle)3 -14.7 56.6 -2.0 53.8
Critical itemsb 24.7 10.2 23.3 9.7
Overreportedc 0.7% 2.2%
Underreported0 48.5% 55.6%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.8 6.6 3.8 6.2
L 50.1 8.0 53.7 1 1.5
F 58.9 7.1 56.6 11.1
K 55.8 9.4 52.1 9.6
1(Hs) 54.7 8.2 49.6 7.9
2(D) 60.1 8.7 53.2 8.7
3(Hy) 61.3 6.5 52.3 7.3
4(Pd) 79.8 6.1 73.2 5.9
5(Mf) 76.7 5.1 73.0 6.0
6 (Pa) 60.2 8.2 51.8 10.0
7(Pt) 60.7 7.2 52.9 8.2
8(Sc) 62.2 7.9 54.3 7.2
9(Ma) 61.6 9.0 55.0 9.0
O(Si) 49.5 8.1 45.5 7.3

Supplementary Scales
M SD M SD
A 50.9 10.2 51.0 9.3
R 59.6 12.3 48.7 11.1
MAC/MAC-Rd
men 25.9 4.2 27.3 4.4
women 26.8 4.1 25.8 4.9

Codetype Concordance
Men Women Men Women
MMPI-2 4-5Z5-4 26.1 % 85.7% MMPI 4-5Z5-4 100.0% 54.5%
Spike 4 25.0 Spike 4 27.3
4-6Z6-4 1 2.5
Spike 5 10.2

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
cPercentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
430 Appendix D

TABLE D-31 Prototypic Scores for 4-6Z6-4 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 358 410
Age 34.1 13.2 32.1 12.6
Men 53.6% 67.6%
Women 46.4 32.4

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 8.8% 7.4%
Total (Obvious-Subtle)8 47.1 62.8 82.6 68.8
Critical items6 37.5 13.0 39.5 13.6
Overreportedc 8.5% 23.6%
Underreported0 21.2% 13.4%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.9 6.0 4.5 6.0
L 51.5 7.8 52.6 10.7
F 68.6 1 1.6 73.8 18.1
K 49.9 8.2 43.4 9.4
1[Hs) 56.4 10.4 54.0 11.1
2(D) 64.2 10.2 62.8 10.1
3(Hy) 61.2 8.6 55.5 10.9
4(Pd) 83.7 8.3 79.0 9.1
5(Mf) 54.9 10.8 49.3 10.1
6(Pa) 81.0 8.6 80.6 10.8
7(Pt) 63.4 9.6 61.1 10.8
8(Sc) 70.1 10.2 65.4 1 1.2
9(Ma) 64.9 9.0 59.1 10.2
0(Si) 55.8 9.2 54.1 9.4

Supplementary Scales
M SD M SD
A 58.6 10.2 62.8 10.7
R 60.9 13.2 49.8 10.5
MAC/M AC-R6
men 28.7 4.4 28.7 4.3
women 24.5 4.0 24.9 4.5

Codetype Concordance
Men Women Men Women
MMPI-2 4-6Z6-4 81.8% 68.7% MMPI 4-6Z6-4 52.4% 85.7%
4-8Z8-4 18.2 12.0

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 7 5th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 431

TABLE D-32 Prototypic Scores for 4-7/7-4 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 161 102
Age 32.3 12.0 30.1 10.6
Men 87.0% 81.4%
Women 13.0 18.6

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 7.4% 4.2%
Total (Obvious-Subtle)3 49.8 53.7 76.6 60.7
Critical itemsb 34.9 1 1.2 36.9 1 1.8
Overreportedc 6.3% 21.4%
Underreported0 16.3% 8.7%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 5.8 7.1 4.6 6.3
L 46.9 6.5 47.5 8.0
F 61.8 7.8 65.0 12.8
K 51.4 8.5 46.0 8.8
1(Hs) 59.0 1 1.6 55.1 10.5
2(D) 70.3 7.7 67.0 8.1
3(Hy) 63.5 9.5 57.6 10.0
4(Pd) 83.1 7.9 79.9 8.5
5(Mf) 56.7 9.6 49.2 8.8
6(Pa) 64.7 9.2 61.4 1 1.2
7(Pt) 79.7 7.1 76.4 8.1
8(Sc) 70.4 9.5 65.7 10.1
9(Ma) 62.4 10.5 57.0 10.0
O(Si) 58.9 9.9 57.3 9.8

Supplementary Scales
M SD M SD
A 63.7 10.2 67.1 10.8
R 61.0 13.3 51.2 1 1.5
MAC/MAC-Rd
men 28.7 4.1 28.3 3.8
women 22.2 4.0 24.3 3.5

Codetype Concordance
Men Women Men Women
MMPI-2 4-7Z7-4 47.8% 38.1 % MMPI 4-7/7-4 76.2% 42.1%
3-7/7-3 33.3 4-8Z8-4 31.6
Spike 4 1 5.8

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 7 5th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
432 Appendix D

TABLE D-33 Prototypic Scores for 4-8Z8-4 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 664 238
Age 29.1 1 1.3 27.8 10.5
Men 66.3% 66.0%
Women 33.7 34.0

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 19.3% 15.5%
Total (Obvious-Subtle)3 84.6 11.1 104.7 77.9
Critical itemsb 47.1 17.6 45.9 17.5
Overreportedc 25.9% 40.8%
Underreported0 14.1 % 8.3%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 5.1 6.3 4.4 5.9
L 49.2 8.2 51.3 1 1.2
F 77.8 14.3 85.1 19.8
K 49.6 10.0 44.7 1 1.2
1(Hs) 64.0 12.5 61.2 1 1.9
2(D) 72.8 12.1 68.3 10.2
3(Hy) 65.9 10.2 60.0 12.4
4(Pd) 88.7 9.4 83.0 9.6
5(Mf) 58.1 10.6 51.2 10.7
6(Pa) 74.1 12.0 69.2 13.2
7(Pt) 74.4 11.5 69.5 12.1
8(Sc) 89.6 13.7 82.1 12.9
9(Ma) 69.7 10.8 63.2 1 1.6
O(Si) 61.1 10.7 59.1 10.7

Supplementary Scales
M SD M SD
A 64.2 12.4 65.6 12.8
R 62.7 13.0 52.1 1 1.3
MAC/MAC-Rd
men 27.5 4.4 21A 4.6
women 24.0 4.5 24.0 4.5

Codetype Concordance
Men Women Men Women
MMPI-2 4-8Z8-4 35.4% 35.3% MMPI 4-8Z8-4 97.5% 98.8%
6-8Z8-6 16.0
4-6Z6-4 12.3
7-8/8-7 1 2.4
2-8Z8-2 10.2

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 433

TABLE D-34 Prototypic Scores for 4-9Z9-4 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 522 318
Age 30.6 12.3 29.9 1 1.8
Men 75.1% 73.0%
Women 24.9 27.0

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 3.0% 2.9%
Total (Obvious-Subtle)3 20.2 56.9 46.6 60.0
Critical itemsb 32.2 12.1 33.7 1 2.4
Overreportedc 2.7% 9.4%
Underreported0 28.3% 20.1%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.5 5.9 3.8 5.0
L 48.9 7.0 50.8 9.9
F 63.6 10.5 67.3 16.1
K 51.3 8.0 45.9 8.7
1(Hs) 54.5 9.5 52.6 10.0
2(D) 56.0 9.9 53.2 9.6
3(Hy) 58.0 8.6 52.2 10.0
4(Pd) 81.3 7.3 76.1 7.7
5(Mf) 55.3 9.6 47.2 9.6
6 (Pa) 62.4 9.9 57.6 10.9
7(Pt) 59.8 8.7 55.5 9.5
8(Sc) 65.4 9.5 60.1 10.0
9(Ma) 79.3 7.2 75.7 8.0
O(Si) 47.7 7.1 45.4 7.7

Supplementary Scales
M SD M SD
A 54.5 9.9 57.2 10.3
R 52.5 10.2 43.4 8.9
MAC/MA C-Rd
men 31.0 3.8 31.4 3.7
women 28.1 4.1 27.9 3.8

Codetype Concordance
Men Women Men Women
MMPI-2 4-9Z9-4 55.8% 64.6% MMPI 4-9Z9-4 89.7% 96.6%
Spike 9 1 5.8

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1 979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
434 Appendix D

TABLE D-35 Prototypic Scores for 4-0/0-4 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 31 42
Age 29.4 9.7 30.6 10.1
Men 45.2% 76.2%
Women 54.8 23.8

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.8% 2.2%
Total (Obvious-Subtle)8 56.7 45.7 77.8 46.5
Critical itemsb 31.5 9.1 31.5 9.4
Overreportedc 1.6% 1 1.9%
Underreported0 16.5% 7.1%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.2 5.4 3.4 3.8
L 48.6 6.6 50.0 9.2
F 62.4 9.1 64.7 1 1.8
K 45.8 7.5 40.2 8.0
UHs) 51.8 9.7 46.7 9.7
2(D) 63.9 7.0 62.5 6.2
3(Hy) 54.9 7.7 46.4 8.3
4(Pd) 77.3 5.4 72.4 5.4
5(Mf) 48.0 10.0 46.1 8.2
6(Pa) 59.9 8.0 56.1 10.5
7(Pt) 62.4 6.6 57.7 8.4
8(Sc) 59.7 8.3 54.4 9.1
9(Ma) 56.0 6.7 49.3 7.6
O(Si) 73.2 2.3 70.9 3.7

Supplementary Scales
M SD M SD
A 62.1 9.5 64.8 9.9
R 65.9 1 1.6 54.4 9.5
MAC/M AC-Rd
men 21A 3.4 27.0 4.3
women 21.1 4.2 20.0 3.3

Codetype Concordance
Men Women Men Women
MMPI-2 4-0/0-4 75.0% 50.0% MMPI Spike 4 39.2% 43.8%
Spike 0 12.5 2-4Z4-2 17.6
2-4Z4-2 1 1.1 4-0/0-4 1 5.7 56.3
8-0Z0-8 11.1

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1 979) critical items endorsed.
Percentage of patients within this codetype scoring above the 7 5th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 435

TABLE D-36 Prototypic Scores for Spike 5 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 1 13 135
Age 34.8 14.1 35.2 13.6
Men 61.9% 37.0%
Women 38.1 63.0

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 3.6%
Total (Obvious-Subtle)3 -23.9 53.6 .8 54.3
Critical itemsb 18.1 8.8 1 7.8 8.4
Overreported0 0.0% 0.7%
Underreported0 65.5% 45.9%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.4 5.7 3.9 6.2
L 53.0 7.5 58.5 1 1.9
F 55.5 7.2 54.9 1 1.0
K 54.8 8.4 49.9 10.0
1(Hs) 51.2 7.4 48.4 8.0
2(D) 55.3 7.4 50.4 7.2
3(Hy) 55.9 7.4 46.9 8.3
4(Pd) 60.6 6.0 51.8 7.3
5(Mf) 75.2 4.8 70.9 5.5
6(Pa) 54.8 7.7 47.0 8.7
7(Pt) 53.3 7.1 46.7 8.0
8(Sc) 54.1 7.1 47.7 7.8
9(Ma) 57.9 7.3 51.1 6.5
O(Si) 49.8 8.5 48.1 8.0

Supplementary Scales
M SD M SD
A 47.3 8.9 48.9 9.1
R 60.9 10.7 50.2 10.0
MAC/M AC-Rd
men 24.2 4.7 24.0 4.2
women 23.5 3.8 24.3 3.6

Codetype Concordance
Men Women Men Women
MMPI-2 Spike-5 83.3% 76.4% MMPI Spike 5 53.2% 73.3%
5-0/0-5 16.7 13.9 4-5Z5-4 19.2
Spike 4 18.7

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
436 Appendix D

TABLE D-37 Prototypic Scores for 5-6Z6-5 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 21 26
Age 36.5 14.7 36.2 14.2
Men 90.5% 57.7%
Women 9.5 42.3

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 8.2% 26.1%
Total (Obvious-Subtle)3 18.8 58.1 70.5 70.9
Critical items6 32.2 1 1.2 35.4 13.5
Overreportedc 3.3% 23.1 %
Underreported0 31.1% 19.2%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.0 5.0 3.5 4.4
L 51.9 7.5 54.3 12.3
F 63.6 8.7 72.4 1 7.3
K 51.0 8.8 42.7 10.9
1(Hs) 57.8 9.7 52.3 13.4
2(D) 58.0 10.7 55.2 10.7
3(Hy) 61.0 1 1.5 50.0 1 5.7
4(Pd) 66.4 10.0 57.7 9.5
5(Mf) 79.0 8.8 73.8 7.9
6 (Pa) 78.3 6.3 78.2 10.1
7(Pt) 62.5 7.8 56.3 1 1.0
8(Sc) 66.1 9.6 64.2 9.0
9(Ma) 62.9 8.1 58.3 9.4
O(Si) 53.3 8.9 53.2 8.6

Supplementary Scales
M SD M SD
A 56.2 10.3 60.0 1 1.7
R 60.5 10.7 48.8 12.6
MAC/MAC-Rd
men 25.3 5.2 24.4 3.5
women 25.0 2.8 26.1 5.1

Codetype Concordance
Men Women Men Women
MMPI-2 5-6Z6-5 47.4% 100.0% MMPI 5-6Z6-5 60.0% 18.2%
Spike 6 31.6 4-5Z5-4 20.0
Spike 6 45.5
6-8Z8-6 27.3

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 437

TABLE D-38 Prototypic Scores for 5-7/7-5 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 16 10
Age 35.7 13.5 33.2 12.2
Men 93.7% 70.0%
Women 6.3 30.0

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 10.0%
Total (Obvious-Subtle)3 41.4 48.8 83.1 59.4
Critical items'3 31.2 9.7 36.5 1 1.3
Overreportedc 4.0% 10.0%
Underreported0 16.0% 10.0%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.1 4.1 4.4 5.1
L 48.1 5.5 51.7 14.2
F 59.4 5.1 63.9 8.0
K 50.1 6.8 42.7 8.8
KHs) 58.3 11.1 56.0 9.5
2(D) 63.5 10.3 59.9 8.6
3(Hy) 58.8 9.9 52.8 9.0
4(Pd) 63.8 8.5 57.8 9.9
5(Mf) 80.8 7.2 76.4 8.8
6 (Pa) 63.3 5.5 59.5 10.7
7(Pt) 77.1 5.6 72.6 5.5
8(Sc) 68.3 7.4 64.5 7.1
9(Ma) 62.0 9.1 59.2 8.0
O(Si) 60.4 9.2 58.1 7.8

Supplementary Scales
M SD M SD
A 64.6 6.8 67.4 9.1
R 64.5 8.4 50.6 6.5
MAC/MAC-Rd
men 25.4 5.2 24.3 6.0
women 30.0 — 28.7 1.5

Codetype Concordance
Men Women Men Women
MMPI-2 5-7/7-5 35.7% 100.0% MMPI 5-7/7-5 71.4% 33.3%
Spike 5 21.4 5-8Z8-5 28.6
Spike 4 33.3
Spike 5 33.3

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
438 Appendix D

TABLE D-39 Prototypic Scores for 5-8Z8-5 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 39 21
Age 31.1 1 1.4 29.4 7.7
Men 92.3% 61.9%
Women 7.7 38.1

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 16.2% 41.3%
Total (Obvious-Subtle)3 67.1 52.5 82.8 45.2
Critical itemsb 40.6 12.4 41.6 9.9
Overreported0 13.2% 18.2%
Underreported0 7.4% 0.0%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?6 4.9 6.3 3.4 5.5
L 48.6 8.7 54.1 1 2.1
F 70.3 12.1 75.7 17.6
K 47.6 7.8 43.0 8.7
1(Hs) 57.3 12.7 57.1 12.3
2(D) 65.3 1 1.9 58.2 10.5
3(Hy) 60.3 10.7 54.2 1 1.9
4(Pd) 68.7 7.4 60.1 7.8
5(Mf) 82.6 7.6 74.9 7.2
6(Pa) 68.9 8.2 63.9 8.7
7(Pt) 69.4 9.5 61.2 8.8
8(Sc) 83.9 1 1.2 75.6 10.7
9(Ma) 69.0 8.2 61.2 9.6
O(Si) 59.6 9.1 55.7 9.6

Supplementary Scales
M SD M SD
A 65.3 10.1 65.3 1 1.2
R 58.5 10.7 49.3 9.5
MAC/MAC-R6
men 25.3 3.4 25.7 4.0
women 23.7 7.8 24.9 3.9

Codetype Concordance
Men Women Men Women
MMPI-2 5-8-Z8-5 35.1% 50.0% MMPI 5-8Z8-5 100.0% 22.2%
Spike 5 10.8 4-8Z8-4 22.2
Spike 8 22.2

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 439

TABLE D-40 Prototypic Scores for 5-9Z9-5 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 57 45
Age 31.7 1 1.0 30.2 1 1.3
Men 86.0% 44.4%
Women 14.0 55.6

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 1.7% 7.7%
Total (Obvious-Subtle)3 -0.5 55.0 27.6 56.7
Critical items'3 26.9 1 1.0 27.1 10.7
Overreported0 0.6% 6.5%
Underreported0 35.3% 30.4%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.1 5.3 5.1 6.3
L 50.1 7.3 54.2 1 1.3
F 59.8 7.9 61.2 1 5.7
K 51.2 8.5 46.2 9.2
1(Hs) 50.4 9.4 47.1 10.8
2(D) 49.4 8.9 45.8 8.6
3(Hy) 56.4 8.9 47.4 10.5
4(Pd) 63.0 9.2 56.0 7.6
5(Mf) 76.5 5.9 72.1 5.6
6 (Pa) 59.6 8.1 52.5 10.8
7(Pt) 56.6 7.7 49.6 6.9
8(Sc) 62.4 8.5 56.8 8.1
9(Ma) 76.8 5.4 73.0 6.3
O(Si) 44.5 7.8 43.1 8.0

Supplementary Scales
M SD M SD
A 52.7 10.9 53.7 10.3
R 49.1 10.1 40.0 9.2
M AC/M AC-Rd
men 28.0 4.6 28.1 5.4
women 21A 5.0 27.7 4.3

Codetype Concordance
Men Women Men Women
MMPI-2 5-9Z9-5 44.2% 100.0% MMPI 5-9Z9-5 95.0% 30.8%
Spike 9 34.9 Spike 9 42.3

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
440 Appendix D

TABLE D-41 Prototypic Scores for 5-0/0-5 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 4 10
Age 37.5 9.5 34.4 13.2
Men 50.0% 40.0%
Women 50.0 60.0

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 0.0%
Total (Obvious-Subtle)3 70.5 44.1 47.1 40.0
Critical items’3 24.8 6.9 22.1 9.1
Overreported0 2.3% 0.0%
Underreported0 27.9% 20.0%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 14.5 1 1.3 4.8 5.0
L 44.0 4.9 49.5 7.7
F 52.0 2.4 53.6 7.2
K 40.8 7.1 42.0 10.3
1(Hs) 43.0 9.1 43.3 1 1.3
2(D) 61.3 9.6 58.2 7.6
3(Hy) 44.0 6.5 40.4 6.8
4(Pd) 58.5 7.2 50.2 6.1
5(Mf) 71.5 1.3 69.3 3.3
6 (Pa) 50.8 5.7 47.0 8.3
7 (Ft) 54.8 7.5 56.2 5.3
8(Sc) 53.0 5.6 50.6 8.4
9(Ma) 58.0 4.1 46.3 5.9
0(Si) 74.0 2.8 68.9 3.8

Supplementary Scales
M SD M SD
A 60.3 12.0 57.8 10.9
R 60.0 20.5 59.5 10.2
MAC/M A C-Rd
men 25.0 2.8 19.8 4.6
women 23.0 4.2 21.0 3.5

Codetype Concordance
Men Women Men Women
MMPI-2 Spike 0 100.0% MMPI Spike 5 71.4% 62.5%
5-0/0-5 100.0% Spike 0 25.0
5-0/0-5 1 2.5

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 7 5th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 441

TABLE D-42 Prototypic Scores for Spike 6 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 102 126
Age 39.5 13.4 36.9 13.6
Men 47.1% 66.7%
Women 52.9 33.3

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 5.2%
Total (Obvious-Subtle)3 12.8 52.7 34.7 55.4
Critical items'3 25.5 8.4 25.9 8.9
Overreportedc 2.0% 2.4%
Underreported0 38.2% 26.2%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 6.8 7.7 5.9 6.7
L 53.2 9.1 56.2 1 1.8
F 59.6 7.8 59.9 1 1.9
K 50.5 8.4 44.9 9.2
KHs) 50.2 8.2 47.8 8.7
2(D) 57.2 7.1 52.9 6.9
3(Hy) 53.7 7.7 46.7 7.6
4(Pd) 61.5 5.7 54.5 5.9
5(Mf) 53.0 9.3 47.3 8.5
6(Pa) 74.0 5.3 73.1 6.5
7(Pt) 55.6 7.6 50.5 8.2
8(Sc) 57.3 7.3 52.1 7.9
9(Ma) 57.4 7.6 51.8 6.5
O(Si) 55.2 8.6 50.7 7.4

Supplementary Scales
M SD M SD
A 53.1 9.2 55.5 9.6
R 61.2 10.6 48.4 9.2
MAC/M AC-Rd
men 26.8 4.4 21A 4.4
women 22.7 3.6 23.8 3.5

Codetype Concordance
Men Women Men Women
MMPI-2 Spike 6 81.6% 49.1 MMPI Spike 6 50.0% 75.0%
6-0/0-6 16.3 14.6 6-8Z8-6 1 5.0
4-6Z6-4 25.0

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1 979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
442 Appendix D

TABLE D-43 Prototypic Scores for 6-7/7-6 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 35 109
Age 34.9 1 1.0 34.3 1 1.8
Men 74.3% 75.2%
Women 25.7 24.8

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 12.5% 8.5%
Total (Obvious-Subtle)8 83.8 56.9 139.1 62.0
Critical itemsb 42.2 12.4 48.0 13.0
Overreportedc 23.2% 57.3%
Underreported0 8.9% 0.9%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 7A 7.1 5.8 7.2
L 47.5 7.3 48.7 8.5
F 66.8 12.6 75.6 1 7.9
K 47.2 7.0 39.4 6.6
KHs) 62.6 13.3 62.2 1 2.7
2(D) 71.5 12.0 70.4 10.9
3(Hy) 63.7 1 1.4 61.3 13.5
4(Pd) 71.2 9.7 66.6 10.6
5(Mf) 54.9 10.3 51.0 10.3
6(Pa) 81.9 9.3 85.6 1 2.5
7(Pt) 82.9 8.4 81.8 10.3
8(Sc) 74.7 1 1.3 74.4 12.4
9(Ma) 65.1 9.3 57.8 9.2
O(Si) 63.4 8.6 64.3 9.6

Supplementary Scales
M SD M SD
A 69.9 7.7 74.7 8.3
R 62.8 14.5 53.3 1 1.7
MAC/M A C-Rd
men 29.7 5.9 27.8 5.3
women 21.1 2.6 23.4 4.1

Codetype Concordance
Men Women Men Women
MMPI-2 6-7/7-6 77.8% 88.9% MMPI 7-8/8-7 36.1%
6-0/0-6 1 1.1 6-777-6 25.3 29.6%
6-878-6 44.4
4-676-4 11.1

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 443

TABLE D-44 Prototypic Scores for 6-8Z8-6 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 678 863
Age 30.4 12.2 29.8 12.0
Men 67.9% 74.3%
Women 36.1 25.7

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 43.0% 28.7%
Total (Obvious-Subtle)3 1 59.6 68.3 202.5 70.8
Critical items13 64.0 1 7.2 66.2 16.5
Overreportedc 41.1% 84.5%
Underreported0 1.1% 0.5%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 5.1 6.3 4.8 5.9
L 49.4 7.9 50.5 10.5
F 92.5 14.9 105.0 1 7.3
K 42.8 7.2 36.3 7.3
KHs) 70.3 1 5.1 69.4 14.3
2(D) 75.8 14.9 73.2 13.2
3(Hy) 67.1 12.1 64.6 1 5.3
4(Pd) 79.4 12.2 73.7 12.9
5(Mf) 60.3 1 1.3 54.0 10.5
6 (Pa) 95.3 1 1.4 97.4 13.5
7(Pt) 80.7 13.4 78.0 13.7
8(Sc) 100.5 1 5.1 96.0 14.5
9(Ma) 75.2 10.9 69.6 12.3
O(Si) 65.5 9.4 65.7 10.2

Supplementary Scales
M SD M SD
A 72.0 9.5 76.1 9.6
R 59.2 14.0 49.9 12.2
MAC/MAC-Rd
men 28.6 4.4 28.0 4.4
women 25.5 4.4 25.8 4.2

Codetype Concordance
Men Women Men Women
MMPI-2 6-8Z8-6 86.1% 75.2% MMPI 6-8Z8-6 58.0% 82.6%
7-8/8-7 1 2.2
4-8Z8-4 1 1.6
2-8Z8-2 10.9

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
444 Appendix D

TABLE D-45 Prototypic Scores for 6-979-6 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 131 260
Age 33.9 13.0 31.2 12.4
Men 51.9% 66.5%
Women 48.1 33.5

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 13.9% 15.0%
Total (Obvious-Subtle)3 74.6 54.0 1 14.9 62.6
Critical itemsb 42.5 12.3 46.3 14.3
Overreportedc 12.7% 41.8%
Underreported0 12.0% 3.5%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.0 5.5 4.4 6.0
L 49.5 7.4 50.1 9.9
F 73.2 12.3 82.6 19.4
K 43.4 7.1 37.2 7.2
1(Hs) 53.0 1 1.0 52.7 12.5
2(D) 52.0 8.7 51.7 9.6
3(Hy) 52.6 9.7 48.3 1 1.7
4(Pd) 66.7 8.5 62.8 10.2
5(Mf) 57.0 9.1 51.4 9.4
6(Pa) 80.6 8.6 81.2 1 1.3
7 (Ft) 59.6 8.0 58.2 10.6
8(Sc) 70.8 9.4 67.9 1 1.5
9(Ma) 82.5 8.6 80.5 10.3
O(Si) 51.8 8.1 49.5 8.2

Supplementary Scales
M SD M SD
A 61.2 9.1 65.9 9.9
R 48.6 1 1.4 38.9 10.2
MAC/MAC-Rd
men 30.4 4.2 32.0 4.1
women 27.6 4.0 27.9 4.2

Codetype Concordance
Men Women Men Women
MMPI-2 6-979-6 84.2% 88.2% MMPI 6-979-6 3 8.5% 71.9%
Spike 6 1 5.8 6-878-6 2 3.1% 12.3%
1-6/6-1 1 1.8

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 445

TABLE D-46 Prototypic Scores for 6-0/0-6 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 13 50
Age 39.8 1 5.1 36.5 13.5
Men 23.1 % 82.0%
Women 76.9 18.0

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 1.8% 8.6%
Total (Obvious-Subtle)3 60.2 39.6 124.3 49.4
Critical itemsb 32.2 8.0 37.0 11.1
Overreportedc 1.8% 44.0%
Underreported0 1 7.5% 0.0%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.4 4.6 4.4 5.9
L 54.2 7.0 50.5 10.0
F 65.4 9.0 71.2 13.4
K 47.2 7.5 36.4 6.8
KHs) 57.8 9.3 49.4 1 1.2
2(D) 65.3 4.6 63.3 7.7
3(Hy) 57.5 7.0 46.2 8.8
4(Pd) 59.8 7.9 57.5 9.0
5(Mf) 46.4 5.8 47.3 8.4
6 (Pa) 75.3 5.8 75.9 7.1
7(Pt) 59.7 7.4 59.5 9.8
8(Sc) 63.9 5.5 61.1 10.3
9(Ma) 54.4 8.8 51.4 9.3
O(Si) 73.2 3.2 72.0 5.5

Supplementary Scales
M SD M SD
A 59.0 7.4 69.7 10.6
Ft 77.3 12.7 57.1 9.8
MAC/MAC-Rd
men 25.7 5.9 25.5 5.3
women 22.1 4.6 19.9 1.8

Codetype Concordance
Men Women Men Women
MMPI-2 6-0/0-6 100.0% 50.0% MMPI 6-8/8-6 17.8%
1-6/6-1 20.0 Spike 6 1 7.8 53.3%
6-0/0-6 33.3

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
446 Appendix D

TABLE D-47 Prototypic Scores for Spike 7 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 33 43
Age 32.5 14.0 31.4 12.2
Men 72.7% 69.8%
Women 27.3 30.2

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 6.0%
Total (Obvious-Subtle)3 28.5 52.7 50.2 hi.2
Critical items13 27.4 8.9 28.1 9.4
Overreportedc 6.1 % 7.0%
Underreported0 36.4% 27.9%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.9 5.7 3.2 3.8
L 49.8 7.6 53.0 1 1.4
F 57.9 7.1 57.9 9.9
K 51.4 8.6 46.1 10.8
1(Hs) 53.3 8.2 51.1 8.0
2(D) 61.4 5.2 58.3 5.2
3(Hy) 56.6 7.6 49.3 8.3
4(Pd) 61.1 7.8 53.9 7.3
5(Mf) 54.2 8.8 46.4 6.9
6 (Pa) 59.2 5.8 52.4 6.7
7(Pt) 74.3 3.7 69.1 3.7
8(Sc) 62.5 5.9 56.4 5.8
9(Ma) 57.2 7.9 50.3 7.4
O(Si) 57.5 6.6 55.2 6.8

Supplementary Scales
M SD M SD
A 61.4 8.9 62.4 9.5
R 60.2 1 1.3 49.3 7.6
MAC/MAC-Rd
men 26.6 4.0 26.8 3.8
women 18.4 3.4 22.0 4.3

Codetype Concordance
Men Women Men Women
MMPI-2 Spike 7 61.1% 63.2% MMPI Spike 7 45.8% 100.0%
7-0/0-7 38.3 4-777-4 33.3
2-777-2 10.5

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 7 5th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 447

TABLE D-48 Prototypic Scores for 7-8/8-7 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 443 323
Age 32.1 12.3 32.9 12.6
Men 7 5.8% 67.5%
Women 24.2 32.5

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 27.5% 20.2%
Total (Obvious-Subtle)8 137.6 62.0 164.8 63.0
Critical itemsb 56.1 1 5.2 55.0 14.0
Overreported0 44.5% 70.3%
Underreported0 1.9% 1.2%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 5.0 6.1 5.1 6.0
L 47.1 7.0 48.0 9.4
F 80.2 1 5.2 85.9 18.3
K 45.0 6.7 39.1 7.4
1(Hs) 70.5 14.4 67.7 12.5
2(D) 79.3 12.1 74.8 10.6
3(Hy) 68.8 1 1.0 64.7 13.0
4(Pd) 77.2 1 1.2 69.4 1 1.8
5(Mf) 60.5 1 1.6 52.2 10.5
6(Pa) 76.6 12.3 71.8 1 1.8
7(Pt) 90.8 10.5 86.7 10.4
8(Sc) 97.9 14.2 88.8 12.6
9(Ma) 70.6 10.5 63.0 1 1.4
O(Si) 67.6 8.8 67.5 9.7

Supplementary Scales
M SD M SD
A 74.7 1A 77.3 7.8
R 61.0 13.2 51.8 1 1.7
MAC/MAC-Rd
men 21A 4.7 26.8 4.5
women 22.3 4.2 22.9 4.5

Codetype Concordance
Men Women Men Women
MMPI-2 7-8/8-7 52.1% 71.6% MMPI 7-878-7 81.9% 74.3%
6-878-6 23.8 2-878-2 10.4
2-777-2 1 5.6 4-878-4 13.3

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1 979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
448 Appendix D

TABLE D-49 Prototypic Scores for 7-9Z9-7 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 36 46
Age 36.8 12.3 34.9 14.9
Men 88.9% 73.9%
Women 1 1.1 26.1

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 8.8% 3.6%
Total (Obvious-Subtle)8 67.9 53.1 94.6 54.4
Critical itemsb 38.5 1 1.8 40.0 1 1.4
Overreportedc 6.3% 23.9%
Underreported0 7.5% 4.4%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.5 5.5 4.4 5.7
L 45.5 6.3 46.6 8.4
F 62.3 9.0 65.3 13.6
K 45.4 6.8 39.0 6.7
KHs) 54.1 1 1.4 52.0 1 1.2
2(D) 64.0 10.2 59.1 9.0
3(Hy) 57.3 7.5 49.5 8.8
4(Pd) 69.3 6.4 61.0 8.0
5(Mf) 59.3 10.2 49.6 8.8
6(Pa) 62.7 7.5 57.7 9.2
7(Pt) 76.8 4.7 72.8 5.5
8(Sc) 67.4 8.5 63.9 9.3
9(Ma) 80.5 7.0 76.1 7.4
0(Si) 55.7 10.1 54.8 9.6

Supplementary Scales
M SD M SD
A 67.4 8.6 69.5 7.9
R 49.8 9.8 43.0 7.2
MAC/MAC-Rd
men 29.4 3.7 29.4 3.5
women 25.3 7.2 24.7 3.9

Codetype Concordance
Men Women Men Women
MMPI-2 7-9/9-7 66.7% 50.0% MMPI 7-9/9-7 64.7% 16.7%
Spike 9 1 2.1 7- 8/8-7 1 1.8
8- 9Z9-8 33.0
Spike 9 25.0

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 449

TABLE D-50 Prototypic Scores for 7-0/0-7 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 17 57
Age 34.3 12.0 36.1 14.1
Men 64.7% 77.2%
Women 35.3 22.8

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 0.0%
Total (Obvious-Subtle)3 75.8 48.9 106.3 49.6
Critical items'3 34.1 8.8 34.8 10.2
Overreportedc 13.2% 26.3%
Underreported0 10.5% 1.8%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 2.1 3.1 3.6 4.9
L 48.9 7.0 49.0 7.9
F 62.1 6.6 61.8 1 1.8
K 47.5 8.5 40.2 8.6
1(Hs) 57.8 10.2 54.1 10.2
2(D) 70.4 6.8 67.0 7.1
3(Hy) 57.2 7.9 50.1 10.8
4(Pd) 66.0 6.6 60.0 7.7
5(Mf) 56.4 8.6 47.6 9.9
6 (Pa) 60.6 7.0 55.7 8.9
7(Pt) 11 .A 5.7 74.6 6.2
8(Sc) 69.5 6.1 61.7 7.6
9(Ma) 55.1 8.1 48.5 9.3
0( Si) 77.2 4.9 73.5 5.8

Supplementary Scales
M SD M SD
A 67.6 8.8 72.8 9.1
R 71.8 13.0 57.9 1 1.0
MAC/MAC-R6
men 25.0 4.6 26.3 4.7
women 19.7 2.0 19.6 2.7

Codetype Concordance
Men Women Men Women
MMPI-2 7-0/0-7 54.5% 100.0% MMPI 2-777-2 26.0%
Spike 0 18.2 4-777-4 26.0
7-878-7 16.0
Spike 7 12.0
7-070-7 42.9%
8-0/0-8 14.3
Spike 0 14.3

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 7 5th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
450 Appendix D

TABLE D-51 Prototypic Scores for Spike 8 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 54 41
Age 35.6 1 5.4 35.3 1 5.7
Men 64.8% 53.7%
Women 35.2 46.3

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 4.9%
Total (Obvious-Subtle)3 1 5.9 62.1 33.4 62.2
Critical itemsb 27.4 1 1.5 26.5 10.5
Overreportedc 5.6% 7.3%
Underreported0 33.3% 26.8%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.9 6.1 4.5 5.9
L 54.0 9.6 58.4 1 2.1
F 62.1 8.2 66.3 14.9
K 53.5 10.7 49.3 1 1.4
1(Hs) 56.6 6.7 51.8 6.6
2(D) 58.1 5.7 54.7 5.9
3(Hy) 56.0 7.2 49.3 8.6
4(Pd) 61.5 7.6 55.3 7.0
5(Mf) 55.6 8.1 48.9 8.2
6 (Pa) 59.7 6.0 55.5 7.2
7(Pt) 59.9 5.8 55.8 6.2
8(Sc) 73.7 4.3 67.7 4.0
9(Ma) 61.3 6.1 54.7 5.9
0(Si) 55.6 8.2 52.2 6.2

Supplementary Scales
M SD M SD
A 54.4 10.0 55.3 8.9
R 61.5 10.9 51.3 1 1.8
MAC/M AC-Rd
men 25.0 5.5 26.5 4.6
women 20.8 2.7 21.9 3.9

Codetype Concordance
Men Women Men Women
MMPI-2 Spike 8 66.7% 46.4% MMPI Spike 8 36.4 46.4
Spike 0 16.7 4-8Z8-4 36.4
1-8/8-1 14.3 8-9Z9-8 13.6
8-0/0-8 10.7 8-0/0-8 10.7

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
cPercentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 451

TABLE D-52 Prototypic Scores for 8-9Z9-8 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 305 210
Age 31.3 12.7 30.5 1 1.8
Men 71.1% 69.5%
Women 28.9 30.5

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 32.8% 22.9%
Total (Obvious-Subtle)3 92.1 63.5 121.5 63.5
Critical itemsb 49.2 14.8 49.8 13.8
Overreportedc 18.7% 45.8%
Underreported0 6.3% 2.8%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d A.l 5.8 4.6 5.9
L 41.7 8.2 48.2 1 1.2
F 78.6 14.4 86.6 18.8
K 45.1 8.4 39.5 8.4
1(Hs) 61.4 12.3 59.8 1 1.9
2(D) 58.6 1 1.6 55.9 1 1.3
3(Hy) 58.3 10.6 52.8 12.7
4(Pd) 71.0 10.1 64.0 11.1
5(Mf) 59.3 9.8 51.7 10.4
6(Pa) 71.1 10.7 67.1 12.6
7(Pt) 69.7 10.0 65.9 10.1
8(Sc) 87.4 1 2.0 80.7 10.8
9(Ma) 86.9 8.5 84.4 10.9
O(Si) 54.6 9.0 52.9 9.0

Supplementary Scales
M SD M SD
A 64.7 10.3 67.4 10.3
R 48.4 12.9 40.3 10.9
MAC/MAC-Rd
men 31.0 4.5 30.6 4.2
women 27.4 4.9 27.7 4.6

Codetype Concordance
Men Women Men Women
MMPI-2 8-9Z9-8 57.9% 63.3% MMPI 8-9Z9-8 83.9% 87.7%
6-9/9-6 1 7.6 11.1

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPf 2 and MMPI,
respectively).
dRaw score.
452 Appendix D

TABLE D-53 Prototypic Scores for 8-0/0-8 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 24 36
Age 30.9 1 2.5 33.6 1 1.2
Men 29.2% 58.3%
Women 70.8 41.7

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 6.3% 4.9%
Total (Obvious-Subtle)3 1 10.6 41.4 142.4 45.7
Critical itemsb 43.2 10.1 44.5 10.9
Overreportedc 25.0% 64.9%
Underreported0 8.3% 0.0%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 5.9 7.1 5.3 5.7
L 47.9 6.0 49.0 10.3
F 71.5 1 1.8 78.5 16.7
K 42.1 5.1 36.1 6.2
1(Hs) 56.9 10.1 54.7 10.2
2(D) 67.5 8.5 64.8 1 1.4
3(Hy) 55.9 8.6 49.5 10.1
4(Pd) 66.8 9.2 60.5 10.2
5(Mf) 52.6 8.4 50.7 9.5
6(Pa) 65.5 7.9 60.9 1 1.5
7(Pt) 70.0 8.9 66.5 10.5
8(Sc) 82.2 9.0 77.9 10.4
9(Ma) 59.7 7.9 54.6 8.9
O(Si) 78.2 5.2 75.9 7.2

Supplementary Scales
M SD M SD
A 70.6 7.8 73.6 9.2
R 68.6 9.5 55.0 9.5
MAC/MAC-Rd
men 22.1 4.2 24.3 4.3
women 20.5 3.3 21.1 3.4

Codetype Concordance
Men Women Men Women
MMPI-2 8-0/0-8 71.4% 64.7% MMPI 2-8Z8-2 28.6%
Spike 0 28.6 8-0/0-8 23.8 61.1%
4-8Z8-4 19.0
Spike 8 16.7

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 453

TABLE D-54 Prototypic Scores for Spike 9 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 347 348
Age 34.3 13.8 33.3 1 3.4
Men 73.2% 76.4%
Women 26.8 23.6

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 1.8%
Total (Obvious-Subtle)3 12.6 48.2 34.7 50.0
Critical itemsb 25.7 8.6 26.6 8.4
Overreportedc 0.9% 2.9%
Underreported0 38.9% 23.1%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 4.2 5.2 3.4 4.9
L 49.4 7.3 51.3 9.9
F 58.4 7.5 57.6 10.6
K 48.9 7.4 43.2 8.0
1(Hs) 49.3 7.6 46.0 8.5
2(D) 49.7 8.4 46.0 8.2
3(Hy) 51.1 7.3 43.6 7.6
4(Pd) 61.0 6.3 53.9 7. 1
5(Mf) 53.6 8.7 45.8 9.0
6 (Pa) 55.3 7.5 49.9 8.5
7(Pt) 53.6 7.1 47.9 7.5
8(Sc) 56.6 7.1 50.9 7.3
9(Ma) 75.5 5.8 72.1 6.6
O(Si) 47.3 7.6 44.4 8.0

Supplementary Scales
M SD M SD
A 52.6 9.0 54.9 9.5
R 50.2 10.3 40.6 8.6
MAC/MAC-Rd
men 30.0 3.8 30.2 3.8
women 26.4 3.4 26.6 3.6

Codetype Concordance
Men Women Men Women
MMPI-2 Spike 9 92.1% 70.7% MMPI Spike 9 63.7% 70.7%
3-9Z9-3 21.5

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
454 Appendix D

TABLE D-55 Prototypic Scores for 9-0/0-9 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 3 7
Age 43.0 16.5 38.1 13.5
Men 66.7% 100.0%
Women 33.3 0.0

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 5.3% 10.0%
Total (Obvious-Subtle)3 71.3 29.7 104.7 48.3
Critical itemsb 32.8 8.4 35.7 5.4
Overreportedc 0.0% 14.3%
Underreported0 1 5.8% 0.0%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 1.0 1.7 2.6 3.6
L 43.3 3.1 46.4 4.9
F 62.0 10.1 73.4 6.3
K 41.0 8.7 35.3 5.7
1(Hs) 46.3 8.7 48.0 9.8
2(D) 58.3 12.7 58.3 9.2
3(Hy) 47.3 10.6 43.6 7.3
4(Pd) 60.0 7.0 56.7 9.5
5(Mf) 52.3 1 1.0 48.9 10.8
6 (Pa) 59.0 6.0 58.6 6.2
7(Pt) 60.3 8.5 53.4 7.2
8(Sc) 61.3 7.2 57.1 8.7
9(Ma) 72.7 2.5 68.6 3.9
O(Si) 71.3 1.2 68.9 2.5

Supplementary Scales
M SD M SD
A 66.3 8.6 68.3 8.0
R 62.3 5.1 51.1 5.8
MAC/MAC-Rd
men 26.5 .7 28.0 3.2
women 28.0 -

Codetype Concordance
Men Women Men Women
MMPI-2 9-0/0-9 50.0% 100.0% MMPI Spike 9 68.4% 100.0%
Spike 0 50.0

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1979) critical items endorsed.
Percentage of patients within this codetype scoring above the 75th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
 Prototypic Scores for Specific Codetypes in Psychiatric Settings 455

TABLE D-56 Prototypic Scores for Spike 0 Codetypes in Psychiatric Settings

MMPI MMPI-2

Demographics
M SD M SD
N 54 123
Age 40.5 14.0 39.6 14.3
Men 33.3% 78.0%
Women 66.7 22.0

Test-Taking Scales/Indexes
M SD M SD
Inconsistent 0.0% 4.4%
Total (Obvious-Subtle)3 30.3 43.1 62.4 44.0
Critical items6 23.9 7.6 23.9 7.8
Overreportedc 1.9% 4.0%
Underreported0 20.4% 5.7%

Standard Validity and Clinical Scales (K-Corrected)

M SD M SD
?d 5.2 4.8 4.6 5.3
L 51.8 7.3 53.6 9.3
F 56.5 7.2 55.0 9.7
K 46.9 6.9 39.5 7.4
KHs) 49.5 8.2 46.2 8.6
2(D) 61.7 7.2 57.7 5.9
3(Hy) 52.1 8.0 42.3 6.4
4(Pd) 58.6 7.6 51.5 7.6
5(Mf) 51.0 8.1 45.0 7.3
6(Pa) 56.5 6.9 50.0 8.4
7(Pt) 57.8 6.0 50.3 7.2
8(Sc) 53.4 8.9 46.1 8.4
9(Ma) 50.6 10.0 45.4 7.2
0(Si) 72.5 2.5 68.7 3.5

Supplementary Scales
M SD M SD
A 57.7 7.8 60.6 8.9
R 71.4 1 1.5 55.4 8.8
MAC/MA C-Rd
men 24.7 5.6 25.4 4.7
women 19.7 3.6 21.3 3.3

Codetype Concordance
Men Women Men Women
MMPI-2 Spike 0 100.0% 50.0% MMPI Spike 0 48.5% 92.6%
2-0/0-2 22.0 Spike 2 10.3
Spike 4 10.3

aSee Chapter 3, Table 3-24, for explanation of how this index is computed.
bThe total number of Lachar and Wrobel (1 979) critical items endorsed.
Percentage of patients within this codetype scoring above the 7 5th percentile or below the 25th
percentile on the total T score difference on the Wiener and Harmon Obvious and Subtle subscales
for all patients (see Tables 3-45 and 3-46 for the cutting scores for the MMPI-2 and MMPI,
respectively).
dRaw score.
V
 APPENDIX E

MMPI-2: Item Overlap

among MMPI-2 Scales

457
 Appendix E

True False

1 Mfm Mff R OH GF
2 Es Hs D Hy Mt PK
3 VRIN TRIN Hs Hy Pt Mt PK PS DEP
4 Mfm Mff GM
5 D
6 VRIN F Sc VRIN
7 MAC-R Hy R Re
8 GM Hs Hy VRIN
9 D Hy Pd Pt Sc Mt PK PS DEP TRIN
10 Hs D Hy R Mt WRK
1 1 Hy Pt HEA
12 TRIN F Pd Sc
13 Ma
14 Hy R
15 D Ma Mt ANX WRK OH
16 Pa Pt Sc Mt PK L
17 Pd Pa Sc PK PS WRK
18 F Hs D Hy HEA
19 Mfm Mff
20 GM Hs D Mt HEA
21 Pd Sc Ma PS FAM
22 Pd Pa Sc PK PS TRT
23 Pa Pt Sc Ma PK Es GM
24 F Pa MAC-R BIZ
25 Mfm Mff Si
26 ASP Hy Mfm Mff
27 TPA Mfm Mff Re GF
28 Hs Mt HEA VRIN
29 ANG L K D Hy OH Re
30 F PK ANX
31 D Hy Pd Pt Sc Si A Mt PK PS ANX WRK VRIN Es Do
32 Pd Sc PK PS BIZ VRIN Si Es Re
33 Es D Pt HEA
34 Pd Sc
35 Pd Sc ASP
36 F MAC-R HEA Es
37 PK PS ANG K D R
38 D Pt Sc A Mt PS DEP
39 Hs D Hy PK ANX VRIN TRIN Es
40 Hy HEA VRIN TRIN
41 L
42 F Pd Pa Sc
43 D Mt
44 Hy Sc PS HEA GM
45 Es Hs D Hy R PS HEA
46 D Sc SOD VRIN
47 Hs Hy HEA
48 F Sc PK PS VRIN TRIN
49 MAC-R VRIN D Si PK SOD
50 Ma CYN
51 L
52 Pd MAC-R PK DEP Do
 MMPI-2: Item Overlap among MMPI-2 Scales 459

Item True False

53 Hs HEA Es
54 F Pd FAM WRK
55 Ma Do OBS D
56 D Pd Pt Si A PK PS DEP VRIN TRIN
57 Hs HEA
58 CYN K Hy
59 Hs PK PS HEA VRIN
60 F BIZ Es
61 Ma LSE
62 Mfm Mff GF
63 TRIN Mfm Mff GF
64 Mfm Mff GM
65 Hy Pt Sc A PK PS DEP TRIN TRIN
66 F ASP
67 Mfm Mff OH GF
68 D Mfm Mff GF
69 MAC-R Mfm Mff R OH
70 Si LSE Pd Do Es GM
71 Pd Mt DEP
72 F MAC-R
73 D Pt Mt LSE WRK VRIN TRIN MAC-R Do GM
74 Mfm Mff GM
75 D PK PS DEP
76 CYN K D Hy Mfm Mff
77 L OH
78 F LSE
79 OH Pd Si GF
80 Mfm Mff GM
81 Mt CYN ASP VRIN Hy Pa VRIN
82 Pd Pt A MAC-R Mt PK DEP Es Do
83 K VRIN TRIN Pd FAM
84 F MAC-R ASP VRIN Re GF
85 Sc Ma PK PS
86 VRIN Mfm Mff Si SOD
87 Ma OBS Es
88 Ma
89 Pd Pt OH
90 VRIN F Sc VRIN
91 Hs Hy Sc HEA
92 DScPKDEPTRT
93 L Ma
94 Pd Pt PK PS
95 TRIN D Hy Pd Pa Mt PK PS DEP VRIN TRIN
96 F BIZ
97 Hs HEA
98 Ma Es WRK Hy Pa OH
99 Pd Pa TRIN VRIN
100 Si Re Pa Ma GM
101 Hs Hy PK HEA
102 L F
103 MAC-R VRIN Mfm Mff Re
104 Si CYN ASP Mfm Mff Pa
460 Appendix E

Item True False

105 Pd MAC-R ASP Re GF VRIN

106 Sc Ma Si
107 L Mfm Mff Ma MAC-R
108 F WRK
109 D Pt LSE
110 Si Mt CYN ASP VRIN K Hy Pa VRIN
111 Hs HEA
1 12 Mfm Mff Si R
113 Pd Pa Ma MAC-R
114 F
115 MAC-R Hy FRS
116 PS ANG VRIN K Hy OH
117 D Hs MAC-R OH HEA
118 D R HEA
119 Mfm Mff GF Es
120 F Mfm Mff R
121 Mff GF Mfm
122 Mfm Mff Ma K Pd
123 ASP L GF
124 CYN Hy
125 VRIN TRIN Hy Pd PK FAM VRIN TRIN
126 F
127 D Si A TRIN K
128 Mfm Mff MAC-R GF R Es
129 Hy Pd OH
130 D Pt Mt DEP LSE K
131 Ma Si Mt
132 F Mfm Mff
133 Mfm Mff GF
134 ANG D R
135 Si A PK PS OBS WRK Hy VRIN
136 TPA VRIN K Ma VRIN
137 Mfm Mff MAC-R GM
138 F Pa Sc BIZ VRIN
139 L
140 D Pt Mt PK ANX TRIN
141 Es Hs D Hy PS HEA
142 D R HEA
143 GM Hs D Pd
144 F Pa
145 Pa Sc Ma PS FAM Re
146 D Pa DEP GM
147 D Pt Sc
148 K D Hy Mt
149 Hs HEA
150 F PK PS
151 TPA Hy
152 GM Hs Hy Mt VRIN TRIN
1 53 L OH
154 FRS Ma
155 Ma GF
156 F
 MMPI-2: Item Overlap among MMPI-2 Scales 461

Item True False

1 57 K Hy Pd
1 58 Si SOD K Pd Ma
1 59 Es GM Hy HEA
160 Re Pd MAC-R
161 Si VRIN Hy VRIN
162 F Pa BIZ
163 GM Mfm Mff FRS
164 Hs Hy Re HEA
165 D Pt Sc PS VRIN TRIN
166 Hy Mfm Sc VRIN TRIN Mff MAC-R VRIN
167 Si SOD VRIN K Hy Pd Ma VRIN
168 F Sc Ma MAC-R PK PS R
169 Ma Es OH Re
170 D Pt Sc PK PS ANX
171 K Pd OH
1 72 Hy MAC-R Do
1 73 Hs Hy
1 74 F Pt
175 Hs D Hy Pt HEA Es
1 76 GM VRIN TRIN Hs Hy HEA
1 77 Mfm Mff Es Sc
1 78 D R
1 79 Es Hs Hy Sc HEA
180 F Sc PS
181 D Si HEA
182 Sc Ma
183 L
1 84 VRIN TRIN Mfm Mff
1 85 Si SOD VRIN Hy Pd VRIN
186 F FRS
187 Mfm Mff GM
188 D
189 Es D Si R
190 Sc Ma FAM
191 Mfm Mff
192 F Sc
193 Hy Mfm Mff
194 Mfm Mff HEA
195 Pd FAM VRIN TRIN VRIN TRIN
196 Mfm Mff Pt PK ANX OBS K Es VRIN TRIN
197 Mfm Mff R GF
198 F BIZ
199 Es Re GM VRIN Mfm Mff R VRIN
200 Ma
201 Mfm Mff Do Re GF
202 Pd MAC-R Do Re
203 L GF
204 F HEA
205 Mfm Mff Ma FAM
206 Ma
207 OH Do Mfm Mff Si
208 Hs Hy PS ANX
462 Appendix E

Item True False

209 Mfm Es TRIN Pd Mff Si

210 F Sc
21 1 Ma
212 Ma TPA D
213 Es K Hy
214 MAC-R GM Pd
215 D Si A Mt DEP Es
216 F
217 Pd FAM
218 Hy Pt Sc Ma Mt PS
219 Pd Mfm Mff
220 Ma Do GF
221 Sc PK PS D Es
222 F
223 D Mt PS ANX
224 MAC-R Hs Hy HEA
225 Pd CYN Es
226 VRIN D Pd
227 Ma ASP Do
228 F BIZ
229 Sc Ma MAC-R Es
230 Hy Es
231 Mfm Mff Si GF
232 Do L
233 D Sc A Mt WRK
234 F Pa Sc DEP
235 LSE Mfm Mff Re
236 Mfm Mff Es
237 GM Mfm Mff Si
238 Ma MAC-R D GF
239 TRIN Mfm Mff GF
240 F ASP
241 CYN Hy
242 Pt Sc Ma
243 Si A WRK K Hy Pd Ma Do
244 Ma Pa Do
245 Do Es D
246 F DEP Es
247 Hs Sc HEA
248 Ma ASP D R
249 Hs Hy HEA
250 Ma ASP GF
251 Mfm Mff Si A MAC-R
252 F Sc
253 Ma Hy
254 CYN ASP Mfm Mff
255 Hs Pa Sc Si R HEA
256 Mfm Mff Sc FAM R
257 MAC-R Mfm Mff GF
258 F
259 Pd Pa BIZ VRIN
260 L D
 MMPI-2: Item Overlap among MMPI-2 Scales 463

Item True False

261 Pd
262 Si SOD VRIN TRIN
263 GF Hy Pd Ma
264 F Pd GF
265 Si SOD Hy VRIN TRIN
266 Re GF Pd Pa MAC-R ASP
267 K D Pd Si VRIN
268 Mfm Sc VRIN Mff VRIN
269 Ma Mt ASP
270 F
271 Mfm Mff Pa
272 Mfm Mff GF
273 Pt Sc A Mt PS ANX
274 Sc PK PS TRT
275 Pt Si SOD Do Re VRIN TRIN
276 F Sc
277 Pa Pt Sc A PK PS DEP
278 Sc
279 Sc
280 MAC-R Sc Si PS SOD VRIN
281 Sc Fb SOD
282 F
283 CYN ASP Pa
284 Si CYN ASP VRIN K Pa VRIN
285 Pa Pt
286 OH CYN Pa
287 Sc MAC-R GF
288 F Pd VRIN TRIN
289 Pt Si A GM
290 ANX VRIN K Sc VRIN
291 Sc Fb
292 Sc FAM
293 Pt OH
294 F
295 Sc HEA
296 Sc Si
297 Pa R
298 Sc BIZ
299 Sc Mt PS ANX WRK MAC-R VRIN
300 F FAM
301 Pt A PS ANX
302 Pt Si Mt PK ANG TPA WRK
303 Sc Fb PK DEP
304 Pt PS
305 Pa OH PK PS ANX
306 F DEP TRT
307 Pa Sc Es
308 Pt Si
309 Pt A OBS Do
310 Pt A Es
31 1 Sc A Fb PS BIZ
312 F
464 Appendix E

Item True False

313 Pt OBS
314 TRIN Pa
315 CYN Pa
316 Pt Sc PK PS BIZ VRIN Es
317 Pt Fb FRS
318 F WRK
319 Sc Fb PK PS BIZ
320 Pt Sc
321 GM Pt Si SOD
322 Sc Fb FRS
323 Sc Es Fb FAM
324 F
325 Pt Sc A Mt PS MAC-R Do
326 Pt Si LSE
327 Pt PK OBS
328 Pt A PK PS OBS Si Es
329 Pt Sc Fb FRS
330 F K D R
331 Pt Mt GM DEP
332 Sc Fb
333 Pa Sc Fb BIZ VRIN
334 Pa Fb FRS
335 Si
336 F Pa BIZ
337 Si SOD
338 Si A K
339 A Mt PK ANX WRK K VRIN
340 Si SOD
341 A K
342 MAC-R Si
343 F Sc
344 MAC-R Si OH VRIN
345 Si
346 CYN K R
347 Si A PK
348 Si K
349 F PK SOD VRIN VRIN
350 GM Si R VRIN
351 Si TRIN GM
352 Si CYN
353 GF VRIN Si R SOD VRIN
354 Si R
355 F Pa Sc BIZ
356 K
357 Si Mt
358 CYN TPA Si Re
359 TRIN Si R SOD VRIN TRIN
360 Si SOD TRIN
361 F Pa BIZ
362 Si
363 Si R SOD
364 Si WRK TRT GM VRIN
 MMPI-2: Item Overlap among MMPI-2 Scales 465

Item True False

365 K R
366 Si
367 Si PK SOD TRIN TRIN
368 Si WRK TRT
369 Si LSE VRIN
370 VRIN Si SOD VRIN
371
372 VRIN PS VRIN
373 TRT
374 CYN ASP VRIN VRIN
375 TRT
376 LSE TRT
377 PS DEP TRT TRIN VRIN
378 FAM
379 FAM
380 LSE VRIN
381
382 FAM
383 Fb FAM
384 GF
385 Es GM FRS
386 Do PS
387 Fb MAC-R
388 GM VRIN A DEP
389 ANG
390 A OH
391 A TRT Es
392 FRS GM
393
394 A OBS WRK VRIN Es
395 Fb FRS VRIN GM VRIN
396 VRIN VRIN
397 FRS
398 OH
399 DEP CYN TRT Do
400 A PS DEP OH
401 GM FRS
402
403 CYN VRIN VRIN
404 Fb HEA
405 VRIN Mt PS ANX VRIN
406 Es GF
407 MAC-R Fb
408 A Mt ANX
409 WRK
410 ANG
41 1 A Mt DEP LSE VRIN VRIN
412 MAC-R ASP Do Re
413 Es FAM
414 MAC-R ANG TPA
415 A ANX VRIN
416 Do
466 Appendix E

Item True False

417 Re GF
418 ASP Re
419 ASP TPA
420 TPA OH
421 A LSE VRIN
422 MAC-R R
423 TPA R
424
425 Es
426 GF
427 BIZ
428 A WRK
429
430 ANG TPA R Re VRIN
431 Fb Re
432 R Re
433 OH
434 MAC-R
435 FRS VRIN GM VRIN
436
437 TPA
438 FRS GM
439 MAC-R
440 Re GM OH
441 FRS Es GM
442 A OBS
443
444
445 MAC-R CYN WRK
446
447 FRS Es
448 A
449 Mt GF FAM R
450 Fb LSE
451 A
452
453 FRS
454 Fb DEP
455 FAM
456 MAC-R GF R Re
457 LSE
458 FRS Es
459
460 OH
461 ANG
462 GM FRS
463 Fb PS ANX
464 A Mt PS WRK Es VRIN TRIN
465 R GF
466 BIZ
467 Re GM VRIN VRIN
468 Fb FRS Re
 MMPI-2: Item Overlap among MMPI-2 Scales 467

item True False

469 A Mt PS ANX Es GM
470 CYN Do Re
471 OH PS FRS Es GM
472 Mt VRIN VRIN
473 MAC-R GF Do
474 GM
475 PS LSE
476 Fb LSE
477 GF
478 Fb FAM
479 PS SOD
480 SOD
481
482 OBS TRT VRIN
483 LSE
484 Fb
485 LSE VRIN VRIN
486 ANG
487 GF
488 TRT
489 Fb
490 BIZ
491 OBS WRK TRT VRIN
492
493 TRT
494 TRT
495 TRT
496 ANX
497 OBS TRT
498 GM
499 TRT
500 TRT
501 Fb TRT
502 MAC-R
503 LSE
504 LSE TRT
505 WRK
506 MAC-R Fb DEP VRIN VRIN
507 TPA VRIN VRIN
508 BIZ
509 ANX OBS WRK GM VRIN
510 TPA GF
51 1 GF
512 DEP
513 ANG VRIN
514
515 PS SOD VRIN VRIN
516 Fb PS DEP
517 Fb WRK
518
519 LSE GM
520 Fb DEP VRIN VRIN
468 Appendix E

Item True False

521 VRIN WRK

522
523 TPA
524 Fb
525 Fb WRK
526 Fb LSE
527
528 Fb TRT
529
530 Fb
531 TPA
532 GM
533 VRIN VRIN
534 TRIN
535 TPA
536 GM
537 GF
538 CYN
539 Fb DEP TRT
540 Fb ANG
541 TPA
542 ANG VRIN
543 BIZ FAM
544 Fb
545 TPA WRK
546 DEP
547 OBS
548 ANG GF
549 MAC-R
550 FAM GF
551 BIZ
552 GF
553 OBS
554 DEP WRK TRT VRIN
555 Fb FRS
556 ANX VRIN TRIN VRIN
557
558
559 WRK
560 TRIN
561 WRK
562 LSE VRIN
563 FAM
564 PS ANG
565 PS VRIN TRIN
566 WRK
567 FAM
 APPENDIX F

MMPI-2 to MMPI
Conversion Tables
470 Appendix F

TABLE F-1 Conversion from MMPI Group Form to MMPI-2*

1-1 51-45 101-88 151-162 201-185 251-229 301-273

2-2 52-46 102-89 152-140 202-234 252-306 302-34
3-3 53- 103-91 153-141 203-187 253-230 303-274
4-4 54-78 104-92 154-142 204-191 254-231 304-275
5-5 55-47 105-93 155-143 205-240 255-232 305-277
6-7 56-84 106-94 156-168 206- 256-312 306-278
7-8 57-49 107-95 157-145 207-188 257-318 307-279
8-9 58- 108-97 158-146 208-189 258
- 308-21
9-10 59-50 109-98 159-147 209-246 259-233 309-280
10-11 60-51 110-99 160-148 210-252 260-235 310-12
11-13 61-52 111-100 161-149 211-258 261-236 311-35
12-14 62-53 112-120 162-151 212-190 262-237 312-281
13-15 63- 113-126 163-152 213-193 263-238 313-283
14- 64-55 114-101 164-174 214-194 264-239 314-16
15-16 65-90 115-132 165-153 215-264 265-241 315-17
16-17 66-96 116-103 166-154 216-195 266-242 316-284
17-6 67-56 117-104 167-155 217-196 267-243 317-285
18-20 68-57 118-105 168-180 218-270 268-244 318-9
19-19 69- 119-106 169-186 219-197 269-324 319-286
20-12 70- 120-107 170-157 220-276 270-245 320-287
21-21 71-58 121-138 171-158 221-199 271-248 321-289
22-23 72-59 122-109 172-161 222-200 272-330 322-290
23-18 73-61 123-144 173-160 223-201 273-247 323-32
24-22 74-62 124-110 174-159 224-202 274-249 324-291
25-25 75-102 125-111 175-164 225-203 275-336 325-292
26-26 76-65 126-112 176-163 226-205 276-343 326-23
27-24 77-64 127-113 177-192 227-282 277-250 327-297
28-27 78-67 128-115 178-165 228-206 278-251 328-31
29-28 79-63 129-116 179-166 229-207 279-253 329-293
30-29 80-68 130-117 180-167 230-208 280-254 330-295
31-30 81-69 131-118 181-169 231-209 281-255 331-42
32-31 82-70 132-119 182-170 232-211 282-256 332-296
33-32 83-108 133-121 183-171 233-212 283-257 333-22
34-36 84-71 134-122 184-198 234-213 284-259 334-298
35-42 85-114 135-123 185-204 235-214 285-260 335-299
36-33 86-73 136-124 186-172 236-215 286-349 336-302
37-34 87-74 137-125 187-177 237-217 287-261 337-301
38-35 88-75 138-127 188-173 238-218 288-18 338-305
39-37 89-76 139-150 189-175 239-219 289-263 339-303
40-48 90-77 140-128 190-176 240-220 290-15 340-304
41-38 91-79 141-129 191-178 241-221 291-355 341-307
42-54 92-80 142-130 192-179 242-223 292-265 342-308
43-39 93-81 143-131 193-181 243-224 293-361 343-309
44-40 94-82 144-133 194-182 244-225 294-266 344-310
45-41 95- 145-134 195-183 245-288 295- 345-311
46-43 96-83 146-156 196-210 246-294 296-267 346-313
47-44 97-85 147-135 197-216 247-300 297-268 347-314
48-60 98- 148-136 198-184 248-226 298-269 348-315
49-66 99-86 149-137 199-222 249- 299-271 349-316
50-72 100-87 150-139 200-228 250-227 300-272 350-319
 MMPI-2 to MMPI Conversion Tables 471

TABLE F-1 continued

351-317 382-390 413-407 444- 475-433 506-444 537-

352-320 383-338 414-408 445-422 476- 507-445 538-
353-321 384-391 415-350 446-344 477-434 508- 539-462
354-322 385-392 416-409 447-423 478- 509-446 540-
355-323 386-393 417-410 448-424 479-360 510-447 541-
356-325 387- 418-411 449-353 480-435 511-448 542-
357-326 388-395 419-412 450-359 481-362 512- 543-463
358-327 389-394 420- 451-363 482-342 513- 544-464
359-328 390-396 421-413 452- 483- 514-
545-
360-329 391-340 422- 453- 484- 515-
546-
361-331 392-397 423- 454- 485-436 516-449
547-370
362-285 393- 424- 455-357 486- 517-450
548-
363-332 394-398 425- 456- 487-364 518-451 RAO ouo
QAQ
364-333 395-399 426-414 457- 488- 519- ccn acc
ODU-4DO
365-334 396-400 427-351 458-425 489- 520-452
551-466
366-277 397-339 428- 459- 490- 521-262
552-467
367-385 398-348 429- 460- 491- 522-453
553-468
368-386 399-372 430- 461-356 492-438 523-
554-
369- 400-345 431-415 462- 493- 524-
370- 401-401 432-416 463-426 494-441 525-458 555-469
371-335 402-402 433- 464-427 495-437 526-454 556-
372- 403- 434-417 465-428 496- 527-455 557-
373- 404-403 435- 466-429 497- 528- 558-470
374-341 405-404 436-352 467- 498- 529-456 559-471
375- 406-346 437-418 468-430 499-442 530- 560-472
376- 407-405 438-419 469-358 500-439 531-457 561-
377-337 408- 439-420 470- 501-440 532-459 562-473
378- 409- 440-354 471-431 502-365 533- 563-474
379-388 410-406 441- 472-432 503- 534-460 564-369
380- 411-347 442- 473-367 504-443 535- 565-
381-389 412- 443-421 474- 505-366 536-461 566-

‘Missing numbers represent items from the original MMPI not included in the MMPI-2 booklet. The 16 items
duplicated in the original MMPI are not duplicated in MMPI-2; hence they occur in only one location in MMPI-2
(e.g., items 8 and 318 are now equivalent to item 9).
472 Appendix F

TABLE F-2 Conversion from MMPI-2 to MMPI Group Form*

1-1 65-76 129-141 193-213 257-283 321-353 385-367

2-2 66-49 130-142 194-214 258-211 322-354 386-368
3-3 67-78 131-143 195-216 259-284 323-355 387-
4-4 68-80 132-115 196-217 260-285 324-269 388-379
5-5 69-81 133-144 197-219 261-287 325-356 389-381
6-17 70-82 134-145 198-184 262-521 326-357 390-382
7-6 71-84 135-147 199-221 263-289 327-358 391-384
8-7 72-50 136-148 200-222 264-215 328-359 392-385
9-8,318 73-86 137-149 201-223 265-292 329-360 393-386
10-9 74-87 138-121 202-224 266-294 330-272 394-389
11-10 75-88 139-150 203-225 267-296 331-361 395-388
12-20,310 76-89 140-152 204-185 268-297 332-363 396-390
13-11 77-90 141-153 205-226 269-298 333-364 397-392
14-12 78-54 142-154 206-228 270-218 334-365 398-394
15-13,290 79-91 143-155 207-229 271-299 335-371 399-395
16-15,314 80-92 144-123 208-230 272-300 336-275 400-396
17-16,315 81-93 145-157 209-231 273-301 337-377 401-401
18-23,288 82-94 146-158 210-196 274-303 338-383 402-402
19-19 83-96 147-159 211-232 275-304 339-397 403-404
20-18 84-56 148-160 212-233 276-220 340-391 404-405
21-21,308 85-97 149-161 213-234 277-305,366 341-374 405-407
22-24,333 86-99 150-139 214-235 278-306 342-482 406-410
23-22,326 87-100 151-162 215-236 279-307 343-276 407-413
24-27 88-101 152-163 216-197 280-309 344-446 408-414
25-25 89-102 153-165 217-237 281-312 345-400 409-416
26-26 90-65 154-166 218-238 282-227 346-406 410-417
27-28 91-103 155-167 219-239 283-313 347-411 411-418
28-29 92-104 156-146 220-240 284-316 348-398 412-419
29-30 93-105 157-170 221-241 285-317,362 349-286 413-421-
30-31 94-106 158-171 222-199 286-319 350-415 414-426
31-32,328 95-107 159-174 223-242 287-320 351-427 415-431
32-33,323 96-66 160-173 224-243 288-245 352-436 416-432
33-36 97-108 161-172 225-244 289-321 353-449 417-434
34-37,302 98-109 162-151 226-248 290-322 354-440 418-437
35-38,311 99-110 163-176 227-250 291-324 355-291 419-438
36-34 100-111 164-175 228-200 292-325 356-461 420-439
37-39 101-114 165-178 229-251 293-329 357-455 421-443
38-41 102-75 166-179 230-253 294-246 358-469 422-445
39-43 103-116 167-180 231-254 295-330 359-450 423-447
40-44 104-117 168-156 232-255 296-332 360-479 424-448
41-45 105-118 169-181 233-259 297-327 361-293 425-458
42-35,331 106-119 170-182 234-202 298-334 362-481 426-463
43-46 107-120 171-183 235-260 299-335 363-451 427-464
44-47 108-83 172-186 236-261 300-247 364-487 428-465
45-51 109-122 173-188 237-262 301-337 365-502 429-466
46-52 110-124 174-164 238-263 302-336 366-505 430-468
47-55 111-125 175-189 239-264 303-339 367-473 431-471
48-40 112-126 176-190 240-205 304-340 368-549 432-472
49-57 113-127 177-187 241-265 305-338 369-564 433-475
50-59 114-85 178-191 242-266 306-252 370-547 434-477
51-60 115-128 179-192 243-267 307-341 371- 435-480
52-61 116-129 180-168 244-268 308-342 372-399 436-485
53-62 117-130 181-193 245-270 309-343 373- 437-495
54-42 118-131 182-194 246-209 310-344 374- 438-492
55-64 119-132 183-195 247-273 311-345 375- 439-500
56-67 120-112 184-198 248-271 312-256 376- 440-501
57-68 121-133 185-201 249-274 313-346 377- 441-494
58-71 122-134 186-169 250-277 314-347 378- 442-499
59-72 123-135 187-203 251-278 315-348 379- 443-504
60-48 124-136 188-207 252-210 316-349 380- 444-506
61-73 125-137 189-208 253-279 317-351 381- 445-507
62-74 126-113 190-212 254-280 318-257 382- 446-509
63-79 127-138 191-204 255-281 319-350 383- 447-510
64-77 128-140 192-177 256-282 320-352 384- 448-511
 MMPI-2 to MMPI Conversion Tables 473

TABLE F-2 continued

449-516 466-551 483- 500- 517- 534- 551-

450-517 467-552 484- 501- 518- 535- 552-
451-518 468-553 485- 502- 519- 536- 553-
452-520 469-555 486- 503- 520- 537- 554-
453-522 470-558 487- 504- 521- 538- 555-
454-526 471-559 488- 505- 522- 539- 556-
455-527 472-560 489- 506- 523- 540- 557-
456-529 473-562 490- 507- 524- 541- 558-
457-531 474-563 491- 508- 525- 542- 559-
458-525 475- 492- 509- 526- 543- 560-
459-532 476- 493- 510- 527- 544- 561-
460-534 477- 494- 511- 528- 545- 562-
461-536 478- 495- 512- 529- 546- 563-
462-539 479- 496- 513- 530- 547- 564-
463-543 480- 497- 514- 531- 548- 565-
464-544 481- 498- 515- 532- 549- 566-
465-550 482- 499- 516- 533- 550- 567-

‘Missing numbers represent items that did not appear in the original MMPI. Pairs of numbers refer
to the two locations in which duplicate items appeared in the original MMPI.
References

Aaronson, B. S. (1958). Age and sex influences on lescents. In C. D. Spielberger & J. N. Butcher
MMPI profile peak distributions in an ab¬ (Eds.), Advances in personality assessment
normal population. Journal of Consulting (Vol. 7, pp. 103-126). Hillsdale, NJ:
Psychology, 22, 203-206. Erlbaum.
Aaronson, B. S. (1959). A comparison of two Archer, R. P., Gordon, R. A., Giannetti, R., &
MMPI measures of masculinity-femininity. Singles, J. (1988). MMPI scale clinical corre¬
Journal of Clinical Psychology, 15, 48-50. lates for adolescent inpatients. Journal of
Aaronson, B. S., & Grumpelt, H. R. (1961). Ho¬ Personality Assessment, 52, 707-721.
mosexuality and some MMPI measures of Archer, R. P., Gordon, R. A., & Kirchner, F. H.
masculinity-femininity. Journal of Clinical (1987). MMPI response-set characteristics
Psychology, 17, 245-247. among adolescents. Journal of Personality
American Psychiatric Association (1980). Diag¬ Assessment, 51, 506-516.
nostic and statistical manual of mental disor¬ Archer, R. P., Pancoast, D. L., & Klinefelter, D.
ders (3rd ed.). Washington: Author. (1989). A comparison of MMPI code types
American Psychiatric Association (1987). Diag¬ produced by traditional and recent adoles¬
nostic and statistical manual of mental disor¬ cent norms. Psychological Assessment: A
ders (3rd ed. revised). Washington: Author. Journal of Consulting and Clinical Psychol¬
Anastasi, A. (1968). Psychological testing (3rd ogy, 1, 23-29.
ed.). New York: Macmillan. Armentrout, J. A., & Hauer, A. L. (1978).
Anderson, W. (1956). The MMPI: Low Pa scores. MMPIs of rapists of adults, rapists of chil¬
Journal of Counseling Psychology, 3, 226- dren, and non-rapist sex offenders. Journal
228. of Clinical Psychology, 34, 330-332.
Anderson, W. P., & Kunce, J. T. (1984). Diagnos¬ Arthur, G. (1944). An experience in examining an
tic implications of markedly elevated MMPI Indian twelfth-grade group with the MMPI.
Sc scores for nonhospitalized clients. Journal Mental Hygiene, 28, 243-250.
of Clinical Psychology, 40, 925-930. Astin, A. W. (1959). A factor study of the MMPI
Anthony, N. (1971). Comparison of clients’ stan¬ Psychopathic Deviate scale. Journal of Con¬
dard, exaggerated, and matching MMPI pro¬ sulting Psychology, 23, 550-554.
files. Journal of Consulting and Clinical Astin, A. W. (1961). A note on the MMPI Psycho¬
Psychology, 36, 100-103. pathic Deviate scale. Educational and Psy¬
Anthony, N. (1976). Malingering as role taking. chological Measurement, 21, 895-897.
Journal of Clinical Psychology, 32, 32-41. Ball, J. C. (1960). Comparison of MMPI profile
Apfeldorf, M. (1978). Alcoholism scales and the differences among Negro-white adolescents.
MMPI: Contributions and future directions. Journal of Clinical Psychology, 16, 304-307.
International Journal of the Addictions, 13, Ball, J. C. (1962). Social deviancy and adolescent
17-55. personality. Lexington: University of Ken¬
Archer, R. P. (1984). Use of the MMPI with ado¬ tucky Press.
lescents: A review of salient issues. Clinical Ball, J. C., & Carroll, D. (1960). Analysis of
Psychology Review, 4, 241 -251. MMPI Cannot Say scores in an adolescent
Archer, R. P. (1987). Using the MMPI with ado¬ population. Journal of Clinical Psychology,
lescents. Hillsdale, NJ: Erlbaum. 16, 30-31.
Archer, R. P. (1988). Using the MMPI with ado¬ Barley, W. D., Sabo, T. W. , & Greene, R. L.

475
476 References

(1986). MMPI normal K+ and other un¬ Blanchard, J. S. (1981). Readability of the MMPI.
elevated profiles. Journal of Consulting and Perceptual & Motor Skills, 52, 985-986.
Clinical Psychology, 54, 502-506. Block, J. (1965). The challenge of response sets:
Barron, F. (1953). An ego-strength scale which Unconfounding meaning, acquiescence, and
predicts response to psychotherapy. Journal social desirability in the MMPI. New York:
of Consulting Psychology, 17, 327-333. Appleton-Century-Crofts.
Baucom, D. H. (1976). Independent masculinity Block, J. (1977). An illusory interpretation of the
and femininity scales on the California Psy¬ first factor of the MMPI: A reply to Shwe-
chological Inventory. Journal of Consulting der. Journal of Consulting and Clinical Psy¬
and Clinical Psychology, 44, 876. chology, 45, 930-935.
Baughman, E. E., & Dahlstrom, W. G. (1968). Blumberg, S. (1967). MMPI Fscale as an indicator
Negro and white children: A psychological of severity of psychopathology. Journal of
study in the rural South. New York: Aca¬ Clinical Psychology, 23, 96-99.
demic Press. Boerger, A. R. (1975). The utility of some alterna¬
Ben-Porath, Y. S., & Butcher, J. N. (1989). Psy tive approaches to MMPI scale construction.
chometric stability of rewritten MMPI items. Unpublished doctoral dissertation, Kent
Journal of Personality Assessment, 53, 645- State University, Kent, OH.
653. Boerger, A. R., Graham, J. R., & Lilly, R. S.
Ben-Porath, Y. S., Hostetler, K., Butcher, J. N., (1974). Behavioral correlates of single-scale
& Graham, J. R. (1989). New subscales for MMPI code types. Journal of Consulting
the MMPI-2 Social Introversion (Si) scale. and Clinical Psychology, 42, 398-402.
Psychological Assessment, 1, 169-174. Bond, J. A. (1986). Inconsistent responding to re¬
Benarick, S. J., Guthrie, G. M., & Snyder, W. U. peated MMPI items: Is its major cause really
(1951). An interpretive aid for the Sc scale of carelessness? Journal of Personality Assess-
the MMPI. Journal of Consulting Psychol¬ ment, 50, 50-64.
ogy, 15, 142-144. Bradley, L. A., Prieto, E. J., Hopson, L., & Pro-
Berg, I. A. (1955). Response bias and personality: kop, C. K. (1978). Comment on “Personality
The deviation hypothesis. Journal of Psy¬ organization as an aspect of back pain in a
chology, 40, 61-72. medical setting.” Journal of Personality As¬
Berg, I. A. (1957). Deviant responses and deviant sessment, 42, 573-578.
people: The formulation of the deviation hy¬ Bradley, L. A., Prokop, C. K., Margolis, R., &
pothesis. Journal of Counseling Psychology, Gentry, W. D. (1978). Multivariate analyses
4, 154-161. of the MMPI profiles of low-back pain pa¬
Bernreuter, R. G. (1933). The theory and con¬ tients. Journal of Behavioral Medicine, 1,
struction of the personality inventory. Jour¬ 253-272.
nal of Social Psychology, 4, 387-405. Broughton, R. (1984). A prototype strategy for
Bernstein, I. H., & Garbin, C. P. (1985). A com¬ construction of personality scales. Journal of
parison of alternative proposed subscale Personality and Social Psychology, 47, 1334—
structures for MMPI Scale 2. Multivariate 1346.
Behavioral Research, 20, 223-235. Brower, D. (1947). The relation between intelli¬
Bertelson, A. D., Marks, P. A., & May, G. D. gence and MMPI scores. Journal of Social
(1982). MMPI and race: A controlled study. Psychology, 25, 243-245.
Journal of Consulting and Clinical Psychol¬ Brown, M. N. (1950). Evaluating and scoring the
ogy, 50, 316-318. MMPI “Cannot Say” items. Journal of
Bieliauskas, L. A., & Shekelle, R. B. (1983). Sta¬ Clinical Psychology, 6, 180-184.
ble behaviors associated with high-point D Brozek, J. (1955). Personality changes with age:
MMPI profiles in a nonpsychiatric popula¬ An item analysis of the MMPI. Journal of
tion. Journal of Clinical Psychology, 39, Gerontology, 10, 194-206.
422-426. Buck, J, A., & Graham, J. R. (1978). The 4-3
Bieliauskas, V. J. (1965). Recent advances in psy¬ MMPI profile type: A failure to replicate.
chology of masculinity and femininity. Jour¬ Journal of Consulting and Clinical Psychol¬
nal of Psychology, 60, 255-263. ogy, 46, 344.
Birtchnell, J., & Kennard, J. (1983). What does Buechley, R., & Ball, H. (1952). A new test of “va¬
the MMPI Dependency scale really measure? lidity” for the group MMPI. Journal of Con¬
Journal of Clinical Psychology, 39, 532-543. sulting Psychology, 16, 299-301.
 References 477

Burish, T. G., & Houston, B. K. (1976). Construct Butcher, J. N., Graham, J. R., Dahlstrom, W. G.
validity of the Lie scale as a measure of de¬ & Bowman, E. (1990). The MMPI-2 with col¬
fensiveness. Journal of Clinical Psychology, lege students. Journal of Personality Axsess-
52,310-314. ment, 54, 1-15.
Burkhart, B. R., Christian, W. L., & Gynther, M. Butcher, J. N., Graham, J. R., Williams, C. L., &
D. (1978). Item subtlety and faking on the Ben-Porath, Y. (1989). Development and use
MMPI: A paradoxical relationship. Journal of the MMPI-2 content scales. Minneapolis:
of Personality Assessment, 42, 76-80. University of Minnesota Press.
Burkhart, B. R., Gynther, M. D., & Fromuth, M. Butcher, J. N., & Hostetler, K. (1990). Abbreviat¬
E. (1980). The relative predictive validity of ing MMPI item administration: What can be
subtle versus obvious items on the MMPI de¬ learned from the MMPI for the MMPI-2?
pression scale. Journal of Clincical Psychol¬ Psychological Assessment: Journal of Con¬
ogy, 36, 748-751. sulting and Clinical Psychology, 2, 12-21.
Buros, O. K. (Ed.). (1978). The eighth mental Butcher, J. N., Kendall, P. C., & Hoffman, N.
measurements yearbook. Highland Park, (1980). MMPI short forms: Caution. Journal
NJ: Gryphon Press. of Consulting and Clinical Psychology, 48,
Burton, A. (1947). The use of the Masculinity- 275-278.
Femininity scale of the MMPI as an aid in the Butcher, J. N., & Pancheri, P. (1976). A hand¬
diagnosis of sexual inversion. Journal of book of cross-cultural MMPI research. Min¬
Psychology, 24, 161-164. neapolis: University of Minnesota Press.
Butcher, J. N. (1985). Current developments in Butcher, J. N., & Tellegen, A. (1966). Objections
MMPI use: An international perspective. In to MMPI items. Journal of Consulting Psy¬
C. D. Spielberger & J. N. Butcher (Eds.), Ad¬ chology, 30, 527-534.
vances in personality assessment (Vol. 4, pp. Butcher, J. N., & Tellegen, A. (1978). Common
83-94). Hillsdale, NJ: Erlbaum. methodological problems in MMPI research.
Butcher, J. N. (1990). Education level and MMPI- Journal of Consulting and Clinical Psychol¬
2 measured psychopathology: A case of neg¬ ogy, 46, 620-628.
ligible influence. MMPI-2 News and Pro¬ Byrne, D., Barry, J., & Nelson, D. (1963). Rela¬
files, 7,3. tion of the revised repression-sensitization
Butcher, J. N. (Ed.) (1972). Objective personality scale to measures of self-description. Psycho¬
assessment. New York: Academic Press. logical Reports, 13, 323-334.
Butcher, J. N. (Ed.) (1987). Computerized psy¬ Caldwell, A. B. (1969). MMPI critical items. Un¬
chological assessment: A practitioner's published mimeograph. (Available from
guide. New York: Basic. Caldwell Report, 1545 Sawtelle Boulevard,
Butcher, J. N., Ball, B., & Ray, E. (1964). Effects Los Angeles, CA 90025.)
of socio-economic level on MMPI differ¬ Caldwell, A. B. (1988). MMPI supplemental scale
ences in Negro-white college students. Jour¬ manual. Los Angeles: Caldwell Report.
nal of Counseling Psychology, 11, 83-87. Caldwell, A. B. (1990, August). MMPI, MMPI-2,
Butcher, J. N., Braswell, L., & Raney, D. (1983). and the measurement of the human condi¬
A cross-cultural comparison of American In¬ tion. Paper presented at the annual meeting
dian, black, and white inpatients on the of the American Psychological Association,
MMPI and presenting symptoms. Journal of Boston.
Consulting and Clinical Psychology, 51, Caldwell, M. G. (1953). The youthful male of¬
587-594. fender in Alabama: A study in delinquency
Butcher, J. N., & Clark, L. A. (1979). Recent causation. Sociology and Social Research,
trends in cross-cultural MMPI research and 37, 236-243.
application. In J. N. Butcher (Ed.), New de¬ Calsyn, D. A., Louks, J., & Freeman, C. W.
velopments in the use of the MMPI (pp. 69- (1976). The use of the MMPI with chronic
111). Minneapolis: University of Minnesota low back pain patients with a mixed diagno¬
Press. sis. Journal of Clinical Psychology, 32, 532—
Butcher, J. N., Dahlstrom, W. G., Graham, J. R., 536.
Tellegen, A. M., & Kaemmer, B. (1989). Carson, R. C. (1969). Interpretative manual to the
MMPI-2: Manual for administration and MMPI. In J. N. Butcher (Ed.), MMPI.: Re¬
scoring. Minneapolis: University of Minne¬ search development and clinical applications
sota Press. (pp. 279-296). New York: McGraw.
478 References

Christian, W. L., Burkhart, B. R., & Gynther, M. gree of profile distortion. Journal of Person¬
D. (1978). Subtle-obvious ratings of MMPI ality Assessment, 41, 511-513.
items: New interest in an old concept. Jour¬ Clopton, J. R., Pallis, D. J., & Birtchnell, J.
nal of Consulting and Clinical Psychology, (1979). MMPI profile patterns of suicide
46, 1178-1186. attempters. Journal of Consulting and Clini¬
Clark, C. G., & Miller, H. L. (1971). Validation of cal Psychology, 47, 135-139.
Gilberstadt and Duker’s 8-6 profile type on a Clopton, J. R., Post, R. D., & Larde, J. (1983).
black sample. Psychological Reports, 29, Identification of suicide attempters by means
259-264. of MMPI profiles. Journal of Clinical Psy¬
Clavelle, P. R., & Butcher, J. N. (1977). An adap¬ chology, 39, 868-871.
tive typological approach to psychiatric Coche, E., & Steer, R. A. (1974). The MMPI re¬
screening. Journal of Consulting and Clini¬ sponse consistencies of normal, neurotic and
cal Psychology, 45, 851-859. psychotic women. Journal of Clinical Psy¬
Clopton, J. R. (1974). A computer program for chology, 30, 194-195.
MMPI scale development with contrasted Cofer, C. N., Chance, J., & Judson, A. J. (1949).
groups. Educational and Psychological Mea¬ A study of malingering on the MMPI. Jour¬
surement, 34, 161-163. nal of Psychology, 27, 491-499.
Clopton, J. R. (1975). Automated MMPI inter¬ Colligan, R. C., Davis, Jr., L. J., Morse, R. M., &
pretation based on a modification of Offord, K. P. (1988). Screening medical pa¬
Gilberstadt’s codebook. Journal of Clinical tients for alcoholism with the MMPI: A com¬
Psychology, 31, 648-651. parison of seven scales. Journal of Clinical
Clopton, J. R. (1978a). Alcoholism and the Psychology, 44, 582-592.
MMPI: A review. Journal of Studies on Al¬ Colligan, R. C., & Offord, K. P. (1986). [MMPI
cohol, 39, 1540-1558. data research tape for Mayo Clinic medical
Clopton, J. R. (1978b). MMPI scale development outpatients referred for psychiatric evalua¬
methodology. Journal of Personality Assess- tions.] Unpublished raw data.
ment, 42, 148-151. Colligan, R. C., & Offord, K. P. (1987a). Resil-
Clopton, J. R. (1979a). Development of special ency reconsidered: Contemporary MMPI
MMPI scales. In C. S. Newmark (Ed.), normative data for Barron’s Ego Strength
MMPI: Clinical and research trends (pp. Scale. Journal of Clinical Psychology, 43,
354-372). New York: Praeger. 467-471.
Clopton, J. R. (1979b). The MMPI and suicide. In Colligan, R. C., & Offord, K. P. (1987b). The
C. S. Newmark (Ed.), MMPI: Clinical and MacAndrew alcoholism scale applied to a
research trends (pp. 149-166). New York: contemporary normative sample. Journal of
Praeger. Clinical Psychology, 43, 291-293.
Clopton, J. R. (1982). MMPI scale development Colligan, R. C., & Offord, K. P. (1988a). Changes
methodology reconsidered. Journal of Per¬ in MMPI factor scores: Norms for the Welsh
sonality Assessemnt, 46, 143-146. A and R dimensions from a contemporary
Clopton, J. R., & Baucom, D. H. (1979). MMPI normal sample. Journal of Clinical Psychol¬
ratings of suicide risk. Journal of Personality ogy, 44, 142-148.
Assessment, 43, 293-296. Colligan, R. C., & Offord, K. P. (1988b). Con¬
Clopton, J. R., & Jones, W. C. (1975). Use of the temporary norms for the Wiggins content
MMPI in the prediction of suicide. Journal scales: A 45-year update. Journal of Clinical
of Clinical Psychology, 31, 52-54. Psychology, 44, 23-32.
Clopton, J. R., & Klein, G. L. (1978). An initial Colligan, R. C., & Offord, K. P. (1989). The aging
look at the redundancy of specialized MMPI MMPI: Contemporary norms for contempo¬
scales. Journal of Consulting and Clinical rary adolescents. Mayo Clinic Proceedings,
Psychology, 46, 1436-1438. 64, 3-27.
Clopton, J. R., & Neuringer, C. (1977a). FOR¬ Colligan, R. C., Osborne, D., & Offord, K. P.
TRAN computer programs for the develop¬ (1980). Linear transformation and the inter¬
ment of new MMPI scales. Educational and pretation of MMPI T scores. Journal of
Psychological Measurement, 37, 783-786. Clinical Psychology, 36, 162-165.
Clopton, J. R., & Neuringer, C. (1977b). MMPI Colligan, R. C., Osborne, D., & Offord, K. P.
Cannot Say scores: Normative data and de¬ (1984). Normalized transformations and the
 References 479

interpretation of MMPI T scores: A reply to Cooke, G., Pogany, E., & Johnston, N. G. (1974).
Hsu. Journal of Consulting and Clinical Psy¬ A comparison of blacks and whites commit¬
chology, 52, 824-826. ted for evaluation of competency to stand
Colligan, R. C., Osborne, D., Swenson, W. M., & trial on criminal charges. Journal of Psychia¬
Offord, K. P. (1983). The MMPI: A contem¬ try and Law, 2, 319-337.
porary normative study. New York: Praeger. Costello, R. M. (1973). Item level racial differ¬
Colligan, R. C., Osborne, D., Swenson, W. M., & ences on the MMPI. Journal of Social Psy¬
Offord, K. P. (1984). The MMPI: Develop¬ chology, 91, 161-162.
ment of contemporary norms. Journal of Costello, R. M. (1977). Construction and cross-
Clinical Psychology, 40, 100-107. validation of an MMPI black-white scale.
Colligan, R. C., Osborne, D., Swenson, W. M., & Journal of Personality Assessment, 41, 514-
Offord, K. P. (1985). Using the 1983 norms 519.
for the MMPI: Code type frequencies in four Costello, R. VI., Fine, H. J., & Blau, B. I. (1973).
clinical samples. Journal of Clinical Psychol¬ Racial comparisons on the MMPI. Journal
ogy, 41, 629-633. of Clinical Psychology, 29, 63-65.
Colligan, R. C., Osborne, D., Swenson, W. M., & Costello, R. M., Tiffany, D. W., & Gier, R. H.
Offord, K. P. (1989). The MMPI: A contem¬ (1972). Methodological issues and racial
porary normative study of adults (2nd ed.). (black-white) comparisons on the MMPI.
Odessa, FL: Psychological Assessment Re¬ Journal of Consulting and Clinical Psychol¬
sources. ogy, 38, 161-168.
Comrey, A. L. (1957a). A factor analysis of items Cowan, M. A., Watkins, B. A., & Davis, W. E.
on the MMPI Hypochondriasis scale. Educa¬ (1975). Level of education, diagnosis and
tional and Psychological Measurement, 17, race-related differences in MMPI perfor¬
568-577. mance. Journal of Clinical Psychology, 31,
Comrey, A. L. (1957b). A factor analysis of items 442-444.
on the MMPI Depression scale. Educational Coyle, Jr., F. A., & Heap, R. F. (1965). Interpre¬
and Psychological Measurement, 17, 578- ting the MMPI L scale. Psychological Re¬
585. ports, 77, 722.
Comrey, A. L. (1957c). A factor analysis of items Craig, R. J. (1984). A comparison of MMPI pro¬
on the MMPI Hysteria scale. Educational files of heroin addicts based on multiple
and Psychological Measurement 17, 586- methods of classification. Journal of Person¬
592. ality Assessment, 48, 115-120.
Comrey, A. L. (1958a). A factor analysis of items Cuellar, I., Harris, L. C., & Jasso, R. (1980). An
on the MMPI Psychopathic Deviate scale. acculturation scale for Mexican-American
Educational and Psychological Measure¬ normal and clinical populations. Hispanic
ment, 18, 91-98. Journal of Behavioral Science, 2, 199-217.
Comrey, A. L. (1958b). A factor analysis of items Dahlstrom, W. G. (1954). Prediction of adjust¬
on the MMPI Paranoia scale. Educational ment after neurosurgery. American Psychol¬
and Psychological Measurement, 18, 99-107. ogist, 9, 353-354.
Comrey, A, L. (1958c). A factor analysis of items Dahlstrom, W. G., Lachar, D., & Dahlstrom, L.
on the MMPI Psychasthenia scale. Educa¬ E. (1986). MMPI patterns of American mi¬
tional and Psychological Measurement, 18, norities. Minneapolis: University of Minne¬
293-300. sota Press.
Comrey, A. L. (1958d). A factor analysis of items Dahlstrom, W. G., Welsh, G. S., & Dahlstrom, L.
on the MMPI Hypomania scale. Educational E. (1972). An MMPI handbook: Vol. I. Clin¬
and Psychological Measurement, 18, 312- ical interpretation (rev. ed.). Minneapolis:
323. University of Minnesota Press.
Comrey, A. L., & Marggraff, W. M. (1958). A Dahlstrom, W. G., Welsh, G. S., & Dahlstrom, L.
factor analysis of items on the MMPI Schizo¬ E. (1975). An MMPI handbook: Vol. II. Re¬
phrenia scale. Educational and Psychological search applications (rev. ed.). Minneapolis:
Measurement, 18, 301-311. University of Minnesota Press.
Constantinople, A. (1973). Masculinity-feminin¬ Dana, R. H. (1990). Cross-cultural and multi-eth¬
ity: An exception to a famous dictum? Psy¬ nic assessment. In J. N. Butcher & C. D.
chological Bulletin, 80, 389-407. Spielberger (Eds.), Advances in personality
480 References

assessment (\o 1. 8, pp. 1-26). Hillsdale, NJ: codebook for counselors. Minneapolis: Uni¬
Erlbaum. versity of Minnesota Press.
Davies, J., Nichols, D. S., & Greene, R. L. (1985, Dubinsky, S., Gamble, D. J., & Rogers, M. L.
March). A new schizophrenia scale for the (1985). A literature review of subtle-obvious
MMPI. Paper presented at the 20th Annual items on the MMPI. Journal of Personality
Symposium on Recent Advances in the Use Assessment, 49, 62-68.
of the MMPI, Honolulu. Dubro, A. F., Wetzler, S., & Kahn, M. W. (1988).
Davis, K. R., & Sines, J. O. (1971). An antisocial A comparison of three self-report question¬
behavior pattern associated with a specific naires for the diagnosis of DSM-III personal¬
MMPI profile. Journal of Consulting and ity disorders. Journal of Personality
Clinical Psychology, 36, 229-234. Disorders, 2, 256-266.
Davis, L. J., Colligan, R. C., Morse, R. M., & Duckworth, J. C., & Anderson, W. (1986). MMPI
Offord, K. P. (1987). Validity of the Mac- interpretation manual for counselors and cli¬
Andrew scale in a general medical population. nicians (3rd ed.). Muncie, IN: Accelerated
Journal of Studies on Alcohol, 48, 202-206. Development.
Davis, W. E. (1972). Age and the discriminative Duckworth, J. C., & Barley, W. D. (1988).
“power” of the MMPI with schizophrenic Within-Normal-Limit profiles. In R. L.
and nonschizophrenic patients. Journal of Greene (Ed.), The MMPI: Use in specific
Consulting and Clinical Psychology, 38, 151. populations (pp. 278-315). San Antonio:
Davis, W. E. (1975). Race and the differential Grune & Stratton.
“power” of the MMPI. Journal of Personal¬ Dye, C. J., Bohm, K., Anderten, P., and Won
ity Assessment, 39, 138-140. Cho, D. (1983). Age group differences in de¬
Davis, W. E., Beck, S. J., & Ryan, T. A. (1973). pression on MMPI D Scale. Journal of Clin¬
Race-related and educationally-related ical Psychology, 39, 227-234.
MMPI profile differences among hospital¬ Eaddy, M. L. (1962). An investigation of the Can¬
ized schizophrenics. Journal of Clinical Psy¬ not Say scale of the group MMPI. Disserta¬
chology, 29, 478-479. tion Abstracts, 23, 1070.
Davis, W. E., & Jones, M. H. (1974). Negro ver¬ Edinger, J. D. (1981). MMPI short forms: A clini¬
sus Caucasian psychological test perfor¬ cal perspective. Psychological Reports, 48,
mance revisited. Journal of Consulting and 627-631.
Clinical Psychology, 42, 675-679. Edwards, A. L. (1959). Edwards Personal Prefer¬
Dean, R. B., & Richardson, H. (1964). Analysis of ence Schedule manual. New York: Psycho¬
MMPI profiles of forty college-educated logical Corporation.
overt male homosexuals. Journal of Consult¬ Edwards, A. L. (1977). Comments on Shweder’s
ing Psychology, 28, 483-486. “Illusory correlation and the MMPI contro¬
Deiker, T. E. (1974). A cross-validation of MMPI versy.” Journal of Consulting and Clinical
scales of aggression on male criminal crite¬ Psychology, 45, 925-929.
rion groups. Journal of Consulting and Clin¬ Edwards, A. L., & Diers, C. J. (1962). Social de¬
ical Psychology, 42, 196-202. sirability and the factorial interpretation of
Denny, N., Robinowitz, R., & Penk, W. E. the MMPI. Educational and Psychological
(1987). Conducting applied research on Viet¬ Measurement, 22, 501-509.
nam combat-related Post-traumatic Stress Ehrenworth, N. V., & Archer, R. P. (1985). A
Disorder. Journal of Clinical Psychology, comparison of clinical accuracy ratings of in¬
43, 56-66. terpretive approaches for adolescent MMPI
Dolan, M. P., Roberts, W. R., Penk, W. E., responses. Journal of Personality Assess-
Robinowitz, R., & Atkins, H. G. (1983). Per¬ ment, 49, 413-421.
sonality differences among black, white, and Eichman, W. J. (1962). Factored scales for the
Hispanic-American male heroin addicts on MMPI: A clinical and statistical manual.
MMPI content scales. Journal of Clinical Journal of Clinical Psychology, 18, 363-395.
Psychology, 39, 807-813. Elion, V. H., & Megargee, E. I. (1975). Validity of
Drake, L. E. (1946). A social I-E scale for the the MMPI Pd scale among black males.
MMPI. Journal of Applied Psychology, 30, Journal of Consulting and Clinical Psychol¬
51-54. ogy, 43, 166-172.
Drake, L. E., & Oetting, E. R. (1959). An MMPI Endicott, N. A., Jortner, S., & Abramoff, E.
 References 481

(1969). Objective measures of suspicious¬ Faschingbauer, T. R. (1979). The future of the

ness. Journal of Abnormal Psychology, 74, MMPI. In C. S. Newmark (Ed.), MMPI:
26-32. Clinical and research trends (pp. 373-398).
Erdberg, S. P. (1975). MMPI differences associ¬ New York: Praeger.
ated with sex, race, and residence in a south¬ Faschingbauer, T. R., & Newmark, C. S. (1978).
ern sample. In W. G. Dahlstrom, G. S. Short forms of the MMPI. Lexington, MA:
Welsh, & L. E. Dahlstrom (Eds.), An MMPI Heath.
handbook: Vol. II. Research applications Fekken, G. C., & Holden, R. R. (1987). Assessing
(pp. 155-157). Minneapolis: University of the person reliability of an individual MMPI
Minnesota Press. protocol. Journal of Personality Assessment,
Erickson, W. D., Luxenberg, M. G., Walbek, N. 51, 123-132.
H., & Seely, R. K. (1987). Frequency of Ferguson, G. A. (1971). Statistical analysis in psy¬
MMPI two-point code types among sex of¬ chology and education (3rd ed.). New York:
fenders. Journal of Consulting and Clinical McGraw Hill.
Psychology, 55, 566-570. Fillenbaum, G. G., & Pfeiffer, E. (1976). The
Evans, C., & McConnell, T. R. (1941). A new Mini-Mult: A cautionary note. Journal of
measure of introversion-extroversion. Jour¬ Consulting and Clinical Psychology, 44,
nal of Psychology, 12, 111-124. 698-703.
Evans, R. G. (1984a). MMPI Dependency scale Finney, J. C., Smith, D. F., Skeeters, D. E., & Au-
norms for alcoholics and psychiatric pa¬ venshine, C. D. (1971). MMPI alcoholism
tients. Journal of Clinical Psychology, 40, scales: Factor structure and content analysis.
345-346. Quarterly Journal of Studies on Alcohol, 32,
Evans, R. G. (1984b). Normative data for two 1055-1060.
MMPI critical item sets. Journal of Clinical Fjordbak, T. (1985). Clinical correlates of high
Psychology, 40, 512-515. Lie scale elevations among forensic patients.
Evans, R. G. (1984c). Utility of the MMPI-168 Journal of Personality Assessment, 49, 252-
with men inpatient alcoholics. Journal of 255.
Studies on Alcohol, 45, 371-373. Flanagan, J., & Lewis, G. (1969). Comparison of
Evans, R. G., & Dinning, W. D. (1983). Response Negro and white lower class men on the gen¬
consistency among high F scale scorers on the eral aptitude test battery and the MMPI.
MMPI. Journal of Clinical Psychology, 39, Journal of Social Psychology, 78, 289-291.
246-248. Forgac, G. E., Cassel, C. A., & Michaels, E. J.
Exner, Jr., J. E., McDowell, E., Pabst, J., Stack- (1984). Chronicity of criminal behavior and
man, W. & Kirk, L. (1963). On the detection psychopathology in male exhibitionists.
of willful falsifications in the MMPI. Journal Journal of Clinical Psychology, 40, 827-832.
of Consulting Psychology, 27, 91-94. Forsyth, D. R. (1967). MMPI and college popula¬
Farberow, N. L. (1956). Personality patterns of tions. Journal of College Student Personnel,
suicidal mental hospital patients. In G. S. 8, 90-96,
Welsh & W. G. Dahlstrom (Eds.), Basic Fowler, R. D., & Coyle, F. A. (1968a). A compar¬
readings on the MMPI in psychology and ison of two MMPI actuarial systems used in
medicine (pp. 427-432). Minneapolis: Uni¬ classifying an alcoholic out-patient popula¬
versity of Minnesota Press. tion. Journal of Clinical Psychology, 24,
Farberow, N. L., & Devries, A. G. (1967). An item 434-435.
differentiation analysis of MMPIs of suicidal Fowler, R. D., & Coyle, F. A. (1968b). Scoring
neuropsychiatric hospital patients. Psycho¬ error on the MMPI. Journal of Clinical Psy¬
logical Reports, 20, 607-617. chology, 24, 68-69.
Farr, S. P. & Martin, P. (1988). Neuropsychologi¬ Fowler, R. D., & Coyle, F. A. (1969). Collegiate
cal dysfunction. In R. L. Greene (Ed.), The normative data on MMPI content scales.
MMPI: Use with specific populations (pp. Journal of Clinical Psychology, 25, 62-63.
214-245). San Antonio: Grune & Stratton. Freeman, C., Calsyn, D., & Louks, J. (1976). The
Faschingbauer, T. R. (1974). A 166-item written use of the MMPI with low back pain pa¬
short form of the group MMPI: The FAM. tients. Journal of Clinical Psychology, 32,
Journal of Consulting and Clinical Psychol¬ 294-298.
ogy, 42, 645-655. Friberg, R. R. (1967). Measures of homosexuality:
482 References

Cross-validation of two MMPI scales and Research Laboratory, Veterans Administra¬

implications for usage. Journal of Consult¬ tion Hospital.
ing Psychology, 31, 88-91. Gilberstadt, H., & Duker, J. (1965). A handbook
Fricke, B. G. (1956). Conversion hysterics and the for clinical and actuarial MMPI interpreta¬
MMPI. Journal of Clinical Psychology, 12, tion. Philadelphia: Saunders.
322-326. Goldberg, L. R. (1965). Diagnosticians vs. diag¬
Friedman, A. F., Webb, J. T., & Lewak, R. nostic signs: The diagnosis of psychosis vs.
(1989). Psychological assessment with the neurosis from the MMPI. Psychological
MMPI. Hillsdale, NJ: Erlbaum. Monographs, 79(9, Whole No. 602).
Friedrich, W. N. (1988). Child abuse and sexual Goldberg, L. R. (1969). The search for configural
abuse. In R. L. Greene (Ed.), The MMPI: relationships in personality assessment: The
Use in specific populations (pp. 246-258). diagnosis of psychosis vs. neurosis from the
San Antonio: Grune & Stratton. MMPI. Multivariate Behavioral Research, 4,
Fry, F. D. (1949). A study of the personality traits 523-536.
of college students and of state prison in¬ Golden, C, J., Sweet, J. J., & Osmon, D. C.
mates as measured by the MMPI. Journal of (1979). The diagnosis of brain-damage by the
Psychology, 28, 439-449. MMPI: A comprehensive evaluation. Jour¬
Gallucci, N. T. (1984). Prediction of dissimulation nal of Personality Assessment, 43, 138-142.
on the MMPI in a clinical field setting. Jour¬ Gonen, J. Y., & Lansky, L. M. (1968). Masculin¬
nal of Consulting and Clinical Psychology, ity, femininity, and masculinity-femininity:
52,917-918. A phenomenological study of the M/'scale of
Gallucci, N. T. (1985). Influence of dissimulation the MMPI. Psychological Reports, 23, 183—
on indexes of response consistency for the 194.
MMPI. Psychological Reports, 57, 1013-1014. Good, P. K., & Brantner, J. P. (1961). The
Gallucci, N. T., Kay, D. C., & Thornby, J. I. physician's guide to the MMPI. Minneapo¬
(1989). The sensitivity of 11 substance abuse lis: University of Minnesota Press.
scales from the MMPI to change in clinical Good, P. K., & Brantner, J. P. (1974). A practical
status. Psychology of Addictive Behaviors, guide to the MMPI. Minneapolis: University
3, 29-33. of Minnesota Press.
Gauron, E., Severson, R., & Englehart, R. (1962). Goodstein, L. D. (1954). Regional differences in
MMPI F scores and psychiatric diagnosis. MMPI responses among male college stu¬
Journal of Consulting Psychology, 26, 488. dents. Journal of Consulting Psychology, 18,
Gayton, W. F., Burchstead, G. N., & Matthews, 437-441.
G. R. (1986). An investigation of the utility Gordon, N. G., & Swart, E. C. (1973). A compar¬
of an MMPI Posttraumatic Stress Disorder ison of the Harris-Lingoes subscales between
subscale. Journal of Clinical Psychology, 42, the original standardization population and
916-917. an inpatient Veterans Administration hospi¬
Gearing, M. L. (1979). The MMPI as a primary tal population. VA Newsletter for Research
differentiator and predictor of behavior in in Mental Health and Behavioral Sciences,
prison: A methodological critique and review 15, 28-31.
of the recent literature. Psychological Bulle¬ Gottesman, I. I., Hanson, D. R., Kroeker, T. A.,
tin, 86, 929-963. & Briggs, P. F. (1987). New MMPI norma¬
Genthner, R. W., & Graham, J. R. (1976). Effects tive data and power-transformed T score ta¬
of short-term public psychiatric hospitaliza¬ bles for the Hathaway-Monachesi Minnesota
tion for both black and white patients. Jour¬ cohort of 14,019 15-year-olds and 3,674 18-
nal of Consulting and Clinical Psychology, year-olds. In R. P. Archer (Ed.), Using the
44, 118-124. MMPI with adolescents (pp. 241-297).
Giannetti, R. A., Johnson, J. H., Klingler, D. E., Hillsdale, NJ: Erlbaum.
& Williams, T. A. (1978). Comparison of lin¬ Gottesman, 1.1., & Prescott, C. A. (1989). Abuses
ear and configural MMPI diagnostic meth¬ of the MacAndrew alcoholism scale: A criti¬
ods with an uncontaminated criterion. cal review. Clinical Psychology Review, 9,
Journal of Consulting and Clinical Psychol¬ 223-242.
ogy, 46, 1046-1052. Gough, H. G. (1947). Simulated patterns on the
Gilberstadt, H. (1970). Comprehensive MMPI MMPI. Journal of Abnormal and Social
code book for males. Minneapolis: MMPI Psychology, 42, 215-225.
 References 483

Gough, H. G. (1950). The F minus K dissimula¬ characteristics of alcoholics: A review. Jour¬

tion index for the MMPI. Journal of Con¬ nal of Consulting and Clinical Psychology,
sulting Psychology, 14, 408-413. 56, 197-205.
Gough, H. G. (1954). Some common misconcep¬ Graham, J. R., & Tisdale, M. J. (1983, April). In¬
tions about neuroticism. Journal of Consult¬ terpretation of low Scale 5 scores for women
ing Psychology, 18, 287-292. of high educational levels. Paper presented at
Gough, H. G. (1957). California Psychological In¬ the 18th Annual Symposium on Recent De¬
ventory manual. Palo Alto, CA: Consulting velopments in the Use of the MMPI, Minne¬
Psychologists Press. apolis.
Gough, H. G., McClosky, H., & Meehl, P. E. Gravitz, M. A. (1967). Frequency and content of
(1951). A personality scale for dominance. test items normally omitted from MMPI
Journal of Abnormal and Social Psychology, scales. Journal of Consulting Psychology, 31,
46, 360-366. 642.
Gough, H. G., McClosky, H., & Meehl, P. E. Gravitz, M. A. (1968). Normative findings for the
(1952). A personality scale for social respon¬ frequency of MMPI critical items. Journal of
sibility. Journal of Abnormal and Social Clinical Psychology, 24, 220.
Psychology, 47, 73-80. Gravitz, M. A. (1970). Validity implications of
Graham, J. R. (1977). The MMPI: A practical normal adult MMPI “L” scale endorsement.
guide. New York: Oxford. Journal of Clinical Psychology, 26, 497-499.
Graham, J. R. (1978). Review of MMPI special Gravitz, M. A. (1971). Declination rates on the
scales. In P. McReynolds (Ed.), Advances in MMPI validity and clinical scales. Journal of
psychological assessment (Vol. IV, pp. 11- Clinical Psychology, 27, 103.
55). San Francisco: Jossey-Bass. Gravitz, M. A. (1987). An empirical study of
Graham, J. R. (1979). Using the MMPI in coun¬ MMPI F scale validity. Psychological Re¬
seling and psychotherapy. Clinical notes on ports, 60, 389-390.
the MMPI. No. 1. Minneapolis: National Gravitz, M. A., & Gerton, M. 1. (1976). An empir¬
Computer Systems. ical study of internal consistency in the
Graham, J. R. (1987). The MMPI: A practical MMPI. Journal of Clinical Psychology, 32,
guide (2nd ed.). New York: Oxford. 567-568.
Graham, J. R. (1990a). Congruence between Gray, W. S., & Robinson, H. M. (1963). Gray
MMPI and MMPI-2 codetypes. MMPI-2 Oral Reading Test. Indianapolis: Bobbs-
News and Profiles, 1, 1-2, 12. Merrill.
Graham, J. R. (1990b). MMPI-2: Assessing per¬ Grayson, H. M. (1951). A psychological admis¬
sonality and psychopathology. New York: sions testing program and manual. Los An¬
Oxford. geles: Veterans Administration Center,
Graham, J. R., & Mayo, M. A. (1985, March). A Neuropsychiatric Hospital.
comparison of MMPI strategies for identify¬ Grayson, H. M., & Olinger, L. B. (1957). Simula¬
ing black and white male alcoholic. Paper tion of “normalcy” by psychiatric patients
presented at the 20th Annual Symposium on on the MMPI. Journal of Consulting Psy¬
Recent Developments in the Use of the chology, 21, 73-77.
MMPI, Honolulu. Greene, R. L. (1977). Student acceptance of gener¬
Graham, J. R., & Schroeder, H. E. (1972). Abbre¬ alized personality interpretations: A reexam¬
viated Mf and Si scales for the MMPI. ination. Journal of Consulting and Clinical
Journal of Personality Assessment, 36, 436- Psychology, 45, 965-966.
439. Greene, R. L. (1978a). An empirically derived
Graham, J. R., Schroeder, H. E., & Lilly, R. S. MMPI carelessness scale. Journal of Clinical
(1971). Factor analysis of items on the Social Psychology, 34, 407-410.
Introversion and Masculinity-Femininity Greene, R. L. (1978b). Can clients provide valu¬
scales of the MMPI. Journal of Clinical Psy¬ able feedback to clinicians about their per¬
chology, 27, 367-370. sonality interpretations? Greene replies.
Graham, J. R., Smith, R. L., & Schwartz, G. F. Journal of Consulting and Clinical Psychol¬
(1986). Stability of MMPI configurations for ogy, 46, 1496-1497.
psychiatric inpatients. Journal of Consulting Greene, R. L. (1979). Response consistency on the
and Clinical Psychology, 54, 375-380. MMPI: The TR index. Journal of Personal¬
Graham, J. R., & Strenger, V. E. (1988). MMPI ity Assessment, 43, 69-71.
484 References

Greene, R. L. (1980). The MMPI: An interpretive ing of the American Psychological Associa¬
manual. New York: Grime & Stratton. tion, Atlanta.
Greene, R. L. (1982). Some reflections on “MMPI Greene, R. L., & Garvin, R. D. (1988). Substance
short forms: A literature review.” Journal of abuse/dependence. In R. L. Greene (Ed.),
Personality Assessment, 46, 486-487. The MMPI: Use in specific populations (pp.
Greene, R. L. (1986). [MMPI data research tape 159-197). San Antonio: Grune & Stratton.
for normal adults and college students]. Un¬ Griffith, A. V., & Fowler, R. D. (1960). Psychas¬
published raw data. thenic and Hypomanic scales of the MMPI
Greene, R. L. (1987). Ethnicity and MMPI perfor¬ and reaction to authority. Journal of Coun¬
mance: A review. Journal of Consulting and seling Psychology, 7, 146-147.
Clinical Psychology, 55, 497-512. Griffith, A, V., Upshaw, H. S., & Fowler, R. D.
Greene, R. L. (1988a). Summary. In R. L. Greene (1958). The Psychasthenic and Hypomanic
(Ed.), The MMPI: Use with specific popula¬ scales of the MMPI and uncertainty in judg¬
tions (pp. 316-321). San Antonio: Grune & ments. Journal of Clinical Psychology, 14,
Stratton. 385-386.
Greene, R. L. (1988b). The relative efficacy of F-K Gross, L. R. (1959). MMPI L-F-K relationships
and the obvious and subtle scales to detect with criteria of behavioral disturbance and
overreporting of psychopathology on the social adjustment in a schizophrenic popula¬
MMPI. Journal of Clinical Psychology, 44, tion. Journal of Consulting Psychology, 23,
152-159. 319-323.
Greene, R. L. (Ed.). (1988c). The MMPI: Use in Grossman, L. S., & Wasyliw, O. E. (1988). A psy¬
specific populations. San Antonio: Grune & chometric study of stereotypes: Assessment
Stratton. of malingering in a criminal forensic group.
Greene, R. L. (1989). Assessing the validity of Journal of Personality Assessment, 52, 549-
MMPI profiles in clinical settings. Clinical 563.
Notes on the MMPI, No. 11. Minneapolis: Gulas, L. (1974). MMPI low-point codes for a
National Computer Systems. “normal” young adult population: A nor¬
Greene, R. L. (1990). Stability of MMPI scale mative study. Journal of Clinical Psychol¬
scores across four codetypes over four de¬ ogy, 30, 77-78.
cades. Journal of Personality Assessment, Gynther, M. D. (1961). The clinical utility of “in¬
55, 1-6. valid” MMPI F scores. Journal of Consult¬
Greene, R. L. (1990, March). The effects of age, ing Psychology, 25, 540-542.
education, gender, and specific MMPI Gynther, M. D. (1972). White norms and black
codetypes on the MacAndrew Alcoholism MMPls: A prescription for discrimination?
Scale. Paper presented at the annual meeting Psychological Bulletin, 78, 386-402.
of the Society for Personality Assessment, Gynther, M. D. (1979). Ethnicity and personality:
San Diego. An update. In J. N. Butcher (Ed.), New de¬
Greene, R. L., Arredondo, R., & Davis, H. G. velopments in the use of the MMPI (pp. 113-
(1990, August). The comparability between 140). Minneapolis: University of Minnesota
the MacAndrew Alcoholism Scale—Revised Press.
(MMPI-2) and the MacAndrew Alcoholism Gynther, M. D. (1983). MMPI interpretation: The
Scale (MMPI). Paper presented at the annual effects of demographic variables. In C. D.
meeting of the American Psychological As¬ Spielberger & J. N. Butcher (Eds.), Advances
sociation, Boston. in personality assessment (Vol. 3, pp. 175—
Greene, R. L., & Brown, R. C. (1988). MMPI 193). New York: Lawrence Erlbaum.
adult interpretive system. Lutz, FL: Psycho¬ Gynther, M. D. (1989). MMPI comparisons of
logical Assessment Resources. blacks and whites: A review and commen¬
Greene, R. L., & Brown, R. C. (1990). MMPI-2 tary. Journal of Clinical Psychology, 45,
adult interpretive system. Lutz, FL: Psycho¬ 878-883.
logical Assessment Resources. Gynther, M. D., Altman, H., & Sletten, I. W.
Greene, R. L., Davis, H. G., & Welch, S. M. (1973). Replicated correlates of MMPI two-
(1988, August). Clinical correlates of self-de¬ point code types: The Missouri actuarial sys¬
ception and impression management on the tem. Journal of Clinical Psychology, 29,
MMPI. Paper presented at the annual meet¬ 263-289.
 References 485

Gynther, M. D., Altman, H., & Warbin, R. lessness and confusion for use with the Ad¬
(1973a). Behavioral correlates for the MMPI diction Research Center Inventory (ARCI).
4-9, 9-4 codetypes: A case of the emperor’s Journal of Clinical Psychology, 19, 407-412.
new clothes? Journal of Consulting and Clin¬ Hale, G., Zimostrad, S., Duckworth, J., & Nicho¬
ical Psychology, 40, 259-263. las, D. (1986). The abusive personality:
Gynther, M. D., Altman, H., & Warbin, R. MMPI profiles of male batterers. Paper pre¬
(1973b). Interpretation of uninterpretable sented at the 21th Annual Symposium on Re¬
MMPI profiles. Journal of Consulting and cent Advances in the Use of the MMPI,
Clinical Psychology, 40, 78-83. Clearwater, FL.
Gynther, M. D., & Brilliant, P. J. (1968). The Hall, G. C. N., Mauiro, R. D., Vitaliano, P. P., &
MMPI K+ profile: A reexamination. Journal Proctor, W. C. (1986). The utility of the
of Consulting and Clinical Psychology, 32, MMPI with men who have sexually assaulted
616-617. children. Journal of Consulting and Clinical
Gynther, M. D., & Burkhart, B. R. (1983). Are Psychology, 54, 493-496.
subtle MMPI items expendable? In J. N. Hampton, P. J. (1953). The development of a per¬
Butcher & C. D. Spielberger (Eds.), Ad¬ sonality questionnaire for drinkers. Genetic
vances in personality assessment (Vol. 2, pp. Psychological Monographs, 48, 55-115.
115-132). Hillsdale NJ: Lawrence Erlbaum. Hanvik, L. J. (1949). Some psychological dimen¬
Gynther, M. D., Burkhart, B. R., & Hovanitz, C. sions of low-back pain. Unpublished doc¬
(1979). Do face-valid items have more predic¬ toral dissertation, University of Minnesota,
tive validity than subtle items? The case of Minneapolis.
the MMPI Pd scale. Journal of Consulting Hanvik, L. J. (1951). MMPI profiles in patients
and Clinical Psychology, 47, 295-300. with low back pain. Journal of Consulting
Gynther, M. D., Fowler, R. D., & Erdberg, P. Psychology, 15, 350-353.
(1971). False positives galore: The applica¬ Harding, C. F., Holz, W. C., & Kawakami, D.
tion of standard MMPI criteria to a rural, (1958). The differentiation of schizophrenic
isolated, Negro sample. Journal of Clinical and superficially similar reactions. Journal
Psychology, 27, 234-237. of Clinical Psychology, 14, 147-149.
Gynther, M. D., & Green, S. B. (1980). Accuracy Hare, R. D. (1985). Comparison of procedures for
may make a difference, but does a difference the assessment of psychopathy. Journal of
make for accuracy?: A response to Pritchard Consulting and Clinical Psychology, 53, 7-
and Rosenblatt. Journal of Consulting and 16.
Clinical Psychology, 48, 268-272. Harrell, T. H. (1990, August). The MMPI-2: Do
Gynther, M. D., Lachar, D., & Dahlstrom, W. G. the old interpretations hold true? Paper pre¬
(1978). Are special norms for minorities sented at the annual meeting of the American
needed? Development of an MMPI F scale Psychological Association, Boston.
for blacks. Journal of Consulting and Clini¬ Harris, R. E., & Lingoes, J. C. (1955). Subscales
cal Psychology, 46, 1403-1408. for the MMPI: An aid to profile interpreta¬
Gynther, M. D., & Petzel, T. P. (1967). Differen¬ tion. Unpublished manuscript, University of
tial endorsement of MMPI F scale items by California.
psychotics and behavior disorders. Journal Harris, R. J., Wittner, W., Koppell, B., & Hilf, F.
of Clinical Psychology, 23, 185-188. D. (1970). MMPI scales vs. interviewer rat¬
Gynther, M. D., & Shimkunas, A. M. (1965a). ings of paranoia. Psychological Reports, 27,
Age, intelligence, and MMPI F scores. Jour¬ 447-450.
nal of Consulting Psychology, 29, 383-388. Harrison, R. H., & Kass, E. H. (1967). Differ¬
Gynther, M. D., & Shimkunas, A. M. (1965b). ences between Negro and white pregnant
More data on MMPI F > 16 scores. Journal women on the MMPI. Journal of Consulting
of Clinical Psychology, 21, 275-277. Psychology, 31, 454-463.
Gynther, M. D., & Witt, P. H. (1976). Wind¬ Harrison, R. H., & Kass, E. H. (1968). MMPI cor¬
storms and important persons: Personality relates of Negro acculturation in a northern
characteristics of black educators. Journal of city. Journal of Personality and Social Psy¬
Clinical Psychology, 32, 613-616. chology, 10, 262-270.
Haertzen, C. A., & Hill, H. E. (1963). Assessing Hartshorne, H., & May, M. A. (1928). Studies in
subjective effects of drugs: An index of care¬ deceit. New York: Macmillan.
486 References

Harvey, M. A., & Sipprelle, C. N. (1976). Demand their relationship to the effectiveness of a
characteristic effects on the subtle and obvi¬ low-back pain program. Newsletter for Re¬
ous subscales of the MMPI. Journal of Per¬ search in Psychology, 14, 31-33.
sonality Assessment, 40, 539-544. Hedlund, J. L. (1977). MMPI clinical scale corre¬
Hathaway, S. R. (1956). Scales 5 (Masculinity- lates. Journal of Consulting and Clinical
Femininity), 6 (Paranoia), and 8 (Schizo¬ Psychology, 45, 739-750.
phrenia). In G. S. Welsh & W. G. Dahlstrom Hedlund, J. L., & Won Cho, D. (1979). [MMPI
(Eds.), Basic readings on the MMPI in psy¬ data research tape for Missouri Department
chology and medicine (pp. 104-111). Minne¬ of Mental Health patients.] Unpublished raw
apolis: University of Minnesota Press. data.
Hathaway, S. R. (1972). Where have we gone Hedlund, J. L., Won Cho, D., & Powell, B. J.
wrong? The mystery of the missing progress. (1975). Use of MMPI short forms with psy¬
In J. N. Butcher (Ed.), Objective personality chiatric patients. Journal of Consulting and
assessment: Changing perspectives (pp. 21- Clinical Psychology, 43, 924.
43). New York: Academic. Heilbrun, A. B. (1961). The psychological signifi¬
Hathaway, S. R. (1980). Scales 5 (Masculinity- cance of the MMPI K scale in a normal pop¬
Femininity), 6 (Paranoia), and 8 (Schizo¬ ulation. Journal of Consulting Psychology,
phrenia). In W. G. Dahlstrom & L. 25, 486-491.
Dahlstrom (Eds.), Basic readings on the Heilbrun, A. B. (1963). Revision of the MMPI K
MMPI in psychology and medicine: A new correction procedure for improved detection
selection on personality measurement (pp. of maladjustment in a normal college popu¬
65-75). Minneapolis: University of Minne¬ lation. Journal of Consulting Psychology,
sota Press. 27, 161-165.
Hathaway, S. R., & Briggs, P. F. (1957). Some Heilbrun, A. B. (1979). Psychopathy and violent
normative data on new MMPI scales. Jour¬ crime. Journal of Consulting and Clinical
nal of Clinical Psychology, 13, 364-368. Psychology, 47, 509-516.
Hathaway, S. R., & McKinley, J. C. (1940). A Helmes, E., & McLaughlin, J. D. (1983). A com¬
multiphasic personality schedule (Minne¬ parison of three MMPI short forms: Limited
sota): I. Construction of the schedule. Jour¬ clinical utility in classification. Journal of
nal of Psychology, 10, 249-254. Consulting and Clinical Psychology, 51,
Hathaway, S. R., & McKinley, J. C. (1942). A 786-787.
multiphasic personality schedule (Minne¬ Henrichs, T. F. (1964). Objective configural rules
sota): III. The measurement of symptomatic for disciminating MMPI profiles in a psychi¬
depression. Journal of Psychology, 14, 73- atric population. Journal of Clinical Psy¬
84. chology, 20, 157-159.
Hathaway, S. R., & McKinley, J. C. (1951). Henrichs, T. F. (1966). A note on the extension of
MMPI manual. New York: Psychological MMPI configural rules. Journal of Clinical
Corporation. Psychology, 22, 51-52.
Hathaway, S. R., & McKinley, J. C. (1967). Herreid, C. F., & Herreid, J. R. (1966). Differ¬
MMPI manual (Rev. ed.). New York: Psy¬ ences in MMPI scores in native and nonna¬
chological Corporation. tive Alaskans. Journal of Social Psychology,
Hathaway, S. R., & McKinley, J. C. (1983). Man¬ 70, 191-198.
ual for administration and scoring of the Hibbs, B. J., Kobos, J. C., & Gonzalez, J. (1979).
MMPI. Minneapolis: National Computer Effects of ethnicity, sex, and age on MMPI
Systems. profiles. Psychological Reports, 45, 591 —
Hathaway, S. R., & Meehl, P. E. (1951). An atlas 597.
for the clinical use of the MMPI. Minneapo¬ Hill, H. E., Haertzen, C. A., & Glaser, R. (1960).
lis: University of Minnesota Press. Personality characteristics of narcotic ad¬
Hathaway, S. R., & Monachesi, E. D. (1963). Ad¬ dicts as indicated by the MMPI. Journal of
olescent personality and behavior: MMPI General Psychology, 62, 127-139.
patterns of normal, delinquent, dropout, and Hiner, D. L., Ogren, D. J., & Baxter, J. C. (1969).
other outcomes. Minneapolis: University of Ideal-self responding on the MMPI. Journal
Minnesota Press. of Projective Techniques and Personality As¬
Haven, G. A., & Cole, K. M. (1972). Psychologi¬ sessment, 33, 389-396.
cal correlates of low-back pain syndrome and Hodo, G. L., & Fowler, R. D. (1976). Frequency
 References 487

of MMPI two-point codes in a large alcoholic Honaker, L. M., Harrell, T. M., & Buffaloe, J. D.
sample. Journal of Clinical Psychology, 32, (1989). Equivalency of microtest computer
487-489. MMPI administration for standard and spe¬
Hoffmann, H., Loper, R. G., & Kammeier, M. L. cial scales. Computers in Human Behavior,
(1974). Identifying future alcoholics with 4, 323-337.
MMPI alcoholism scales. Quarterly Journal Hoppe, C. M., & Singer, R. D. (1976). Over¬
of Studies on Alcohol, 35, 490-498. controlled hostility, empathy, and egocentric
Hoffmann, N. G., & Butcher, J. N. (1975). Clini¬ balance in violent and nonviolent psychiatric
cal limitations of three MMPI short forms. offenders. Psychological Reports, 39, 1303—
Journal of Consulting and Clinical Psychol¬ 1308.
ogy, 43, 32-39. Horowitz, L. M., Wright, J. C., Lowenstein, E.,
Hokanson, J. E., & Calden, G. (1960). Negro- & Parad, H. W. (1981). The prototype as a
white differences on the MMPI. Journal of construct in abnormal psychology: 1. A
Clinical Psychology, 16, 32-33. method for deriving prototypes. Journal of
Holcomb, W. R., & Adams, N. (1982). Racial Abnormal Psychology, 90, 568-574.
influences on intelligence and personality Hovanitz, C. A., & Gynther, M. D. (1980). The
measures of people who commit murder. prediction of impulsive behavior: Compara¬
Journal of Clinical Psychology, 38, 793- tive validities of obvious versus subtle MMPI
796. hypomania (Ma) items. Journal of Clinical
Holcomb, W. R., Adams, N. A., & Ponder, H. Psychology, 36, 422-427.
M. (1984). Are separate black and white Hovey, H. B., & Lewis, E. G. (1967). Semi-auto¬
MMPI norms needed?: An IQ-controlled matic interpretation of the MMPI. Journal
comparison of accused murderers. Journal of Clinical Psychology, 23, 123-134.
of Clinical Psychology, 40, 189-193. Hoyt, D. P., & Sedlacek, G. M. (1958). Differenti¬
Holden, R. R., & Jackson, D. N. (1979). Item sub¬ ating alcoholics from normals and abnor¬
tlety and face validity in personality assess¬ mal with the MMPI. Journal of Clinical
ment. Journal of Consulting and Clinical Psychology, 14, 69-74.
Psychology, 47, 459-468. Hryckowian, M. J., & Gynther, M. D. (1988).
Holland, T. R. (1979). Ethnic group differences in MMPI item subtlety: Another look. Journal
MMPI profile pattern and factorial structure of Clinical Psychology, 44, 148-152.
among adult offenders. Journal of Personal¬ Hsu, L. M. (1984). MMPI Tscores: Linear versus
ity Assessment, 43, 72-77. normalized. Journal of Consulting and Clin¬
Holland, T. R., & Levi, M. (1983). Personality ical Psychology, 52, 821-823.
correlates of extent versus type of antisocial Hsu, L. M., & Betman, J. A. (1986). MMPI T
behavior among adult offenders: A multivar¬ score conversion tables, 1957-1983. Journal
iate analysis. Multivariate Behavioral Re¬ of Consulting and Clinical Psychology, 54,
search, 18, 391-400. 497-501.
Holmes, G. R., Sabalis, R. F., Chestnut, E., & Huber, N. A., & Danahy, S. (1975). Use of the
Khoury, L. (1984). Parent MMPI critical MMPI in predicting completion and evaluat¬
item and clinical scale changes in the 1970s. ing changes in a long-term alcoholism treat¬
Journal of Clinical Psychology, 40, 1194— ment program. Journal of Studies on
1198. Alcohol, 36, 1230-1237.
Holmes, W. O. (1953). The development of an em¬ Huesmann, L. R., Lefkowitz, M. M., & Eron, L.
pirical MMPI scale for alcoholism. Unpub¬ D. (1978). Sum of MMPI Scales F, 4, and 9
lished Master’s thesis, San Jose State as a measure of aggression. Journal of Con¬
College. sulting and Clinical Psychology, 46, 1071-
Honaker, L. M. (1988). The equivalency of com¬ 1078.
puterized and conventional MMPI adminis¬ Hunt, H. F. (1948). The effect of deliberate decep¬
tration: A critical review. Clinical tion on MMPI performance. Journal of
Psychology Review, 8, 561-577. Consulting Psychology, 12, 396-402.
Honaker, L. M. (1990, August). MMPI and Hunt, H. F., Carp, A., Cass, Jr., W. A., Winder,
MMPI-2: Alternate forms or different tests? C. L., & Kantor, R. E. (1948). A study of the
Paper presented at the annual meeting of the differential diagnostic efficiency of the
American Psychological Association, Bos¬ MMPI. Journal of Consulting Psychology,
ton. 12, 331-336.
488 References

Hyer, L., Boudewyns, P., Harrison, W. R., (1973). Personality characteristics of alcohol¬
O’Leary, W. C., Bruno, R. D., Saucer, R. ics as college freshmen and at time of treat¬
T., & Blount, J. B. (1988). Vietnam veterans: ment. Quarterly Journal of Studies on
Overreporting versus acceptable reporting of Alcohol, 34, 390-399.
symptoms. Journal of Personality Assess¬ Karol, R. L. (1985). MMPI omitted items: A
ment, 52, 475-486. method for quickly determining individual
Hyer, L., Harkey, B., & Harrison, W. R. (1986). scale impact. Journal of Consulting and
MMPI scales and subscales: Patterns of Clinical Psychology, 53, 134-135.
older, middle-aged, and younger inpatients. Kazan, A. T., & Sheinberg, I. M. (1945). Clinical
Journal of Clinical Psychology, 42, 596- note on the significance of the validity score
601. (F) in the MMPI. American Journal of Psy¬
Ingram, J. C., Marchioni, P., Hill, G., Caraveo- chiatry, 102, 181-183.
Ramos, E., & McNeil, B. (1985). Recidivism, Keane, T. M., Malloy, P. F., & Fairbank, J. A.
perceived problem-solving abilities, MMPI (1984). Empirical development of an MMPI
characteristics, and violence: A study of subscale for the assessment of combat-re¬
black and white incarcerated male adult of¬ lated posttraumatic stress disorder. Journal
fenders. Journal of Clinical Psychology, 41, of Consulting and Clinical Psychology, 52,
425-432. 888-891.
Jackson, D. N. (1967). Review of J. Block: The Kelley, C. K., & King, G. D. (1978). Behavioral
challenge of response sets. Educational correlates for within-normal-limit MMPI
and Psychological Measurement, 27, 207- profiles with and without elevated K in stu¬
219. dents at a university mental health center.
Jackson, D. N. (1968). Personality Research Form Journal of Clinical Psychology, 34, 695-699.
manual. Goshen, NY: Research Psycholo¬ Kelley, C. K., & King, G. D. (1979a). Behavioral
gists Press. correlates of infrequent two-point MMPI
Jackson, D. N. (1971). The dynamics of struc¬ code types at a university mental health cen¬
tured personality tests: 1971. Psychological ter. Journal of Clinical Psychology, 35, 576-
Review, 78, 229-248. 585.
Jarnecke, R. W., & Chambers, E. D. (1977). Kelley, C. K., & King, G. D. (1979b). Behavioral
MMPI content scales: Dimensional struc¬ correlates of the 2-7-8 MMPI profile type in
ture, construct validity, and interpretive students at a university mental health center.
norms in a psychiatric population. Journal Journal of Consulting and Clinical Psychol¬
of Consulting and Clinical Psychology, 45, ogy, 47, 679-685.
1126-1131. Kelley, C. K., & King, G. D. (1979c). Cross-vali¬
Jenkins, G. (1984). Response sets and personality dation of the 2-878-2 MMPI code type for
measures: The K scale of the MMPI. Unpub¬ young adult psychiatric outpatients. Journal
lished doctoral dissertation, Texas Tech Uni¬ of Personality Assessment, 43, 143-149.
versity, Lubbock. Kelley, C. K., & King, G. D. (1980). Two- and
Johnson, J. H., Klingler, D. E., & Williams, T. A. three-point classification of MMPI profiles
(1977). An external criterion study of the in which Scales 2, 7, and 8 are the highest ele¬
MMPI validity indices. Journal of Clinical vations. Journal of Personality Assessment,
Psychology, 33, 154-156. 44, 25-33.
Jones, E. E. (1978). Black-white personality dif¬ Kincannon, J. C. (1968). Prediction of the stan¬
ferences: Another look. Journal of Personal¬ dard MMPI scale scores from 71 items: The
ity Assessment, 42, 244-252. Mini-Mult. Journal of Consulting and Clini¬
Jones, F. W., Neuringer, C., & Patterson, T. W. cal Psychology, 32, 319-325.
(1976). An evaluation of an MMPI response King, G. D., & Kelley, C. K. (1977a). Behavioral
consistency measure. Journal of Personality correlates for Spike-4, Spike-9, and 4-979-4
Assessment, 40, 419-421. MMPI profiles in students at a university
Kamerown, D. B., Pincus, H. A., & Macdonald, mental health center. Journal of Clinical Psy¬
D. I. (1986). Alcohol abuse, other drug chology, 33, 718-724.
abuse, and mental disorders in medical prac¬ King, G. D., & Kelley, C. K. (1977b). MMPI be¬
tice. Journal of the American Medical Asso¬ havioral correlates of Spike-5 and two-point
ciation, 255, 2054-2057. code types with Scale 5 as one elevation.
Kammeier, M. L., Hoffmann, H., & Loper, R. G. Journal of Clinical Psychology, 33, 180-185.
 References 489

King, H. F., Carroll, J. L., & Fuller, G. B. (1977). Lachar, D., Klinge, V., & Grisell, J. L. (1976).
Comparison of nonpsychiatric blacks and Relative accuracy of automated MMPI nar¬
whites on the MMPI. Journal of Clinical ratives generated from adult norm and ado¬
Psychology, 33, 725-728. lescent norm profiles. Journal of Consulting
Kirk, A. R., & Zucker, R. A. (1979). Some and Clinical Psychology, 44, 20-24.
sociopsychological factors in attempted sui¬ Lachar, D., & Wrobel, T. A. (1979). Validating
cide among urban black males. Suicide and clinicians’ hunches: Construction of a new
Life Threatening Behavior, 9, 76-86. MMPI critical item set. Journal of Consult¬
Kline, J. A., Rozynko, V. V., Flint, G., & Rob¬ ing and Clinical Psychology, 47, 277-284.
erts, A. C. (1973). Personality characteristics Lacks, P. B., Rothenberg, P. J., & Unger, B. L.
of male native American alcoholic patients. (1970). MMPI scores and marital status in
International Journal of the Addictions, 8, male schizophrenics. Journal of Clinical Psy¬
729-732. chology, 26, 221-222.
Klinge, V., & Strauss, M. E. (1976). Effects of Lair, C. V., & Trapp, E. P. (1962). The differen¬
scoring norms on adolescent psychiatric tial diagnostic value of MMPI with somati¬
patients’ MMPI profiles. Journal of Person¬ cally disturbed patients. Journal of Clinical
ality Assessment, 40, 13-17. Psychology, 18, 146-147.
Kleinmuntz, B. (1960). Identification of malad¬ Landis, C., & Katz, S. E. (1934). The validity of
justed college students. Journal of Counsel¬ certain questions which purport to measure
ing Psychology, 7, 209-211. neurotic tendencies. Journal of Applied Psy¬
Kleinmuntz, B. (1961a). The college maladjust¬ chology, 18, 343-356.
ment scale (MT): Norms and predictive valid¬ Lane, P. J., & Kling, J. S. (1979). Construct vali¬
ity. Educational and Psychological Mea¬ dation of the Overcontrolled Hostility scale
surement, 21, 1029-1033. of the MMPI. Journal of Consulting and
Kleinmuntz, B. (1961b). Screening: Identification Clinical Psychology, 47, 781-782.
or prediction? Journal of Counseling Psy¬ Lanyon, R. I. (1967). Simulation of normal and
chology, 8, 279-280. psychopathic MMPI personality patterns.
Kleinmuntz, B. (1963). MMPI decision rules for Journal of Consulting Psychology, 31, 94-
the identification of college maladjustment: 97.
A digital computer approach. Psychological Lanyon, R. I. (1968). A handbook of MMPI
Monographs, 77, 14(Whole No. 577). group profiles. Minneapolis: University of
Knapp, R. R. (1960). A reevaluation of the valid¬ Minnesota Press.
ity of MMPI scales of dominance and social Lanyon, R. I., & Lutz, R. W. (1984). MMPI dis¬
responsibility. Educational and Psychologi¬ crimination of defensive and nondefensive
cal Measurement, 20, 381-386. felony sex offenders. Journal of Consulting
Koss, M. P,, & Butcher, J. N. (1973). A compari¬ and Clinical Psychology, 52, 841-843.
son of psychiatric patients’ self-report with Lavin, P. (1984). [Summary of Tennessee Valley
other sources of clinical information. Jour¬ Authority data on the MMPI used in nuclear
nal of Research in Personality, 7, 225-236. power plant personnel screening.] Unpub¬
Koss, M. P., Butcher, J. N., & Hoffmann, N. G. lished raw data.
(1976). The MMPI critical items: How well Lawson, H. H., Kahn, M. W., & Heiman, E. M.
do they work? Journal of Consulting and (1982). Psychopathology, treatment outcome
Clinical Psychology, 44, 921-928. and attitude toward mental illness in Mexican
Kroger, R. O., & Turnbull, W. (1975). Invalidity American and European patients. Interna¬
of validity scales: The case of the MMPI. tional Journal of Social Psychology, 28, 20-
Journal of Consulting and Clinical Psychol¬ 26.
ogy, 43, 48-55. Lawton, M. P., & Kleban, M. H. (1965).
Lachar, D. (1974). The MMPI: Clinical assess¬ Prisoners’ faking on the MMPI. Journal of
ment and automated interpretation. Los An¬ Clinical Psychology, 21, 269-271.
geles: Western Psychological Services. Leon, G. R., Gillum, B., Gillum, R., & Gouze, M.
Lachar, D., & Alexander, R. S. (1978). Veridical- (1979). Personality stability and change over
ity of self- report: Replicated correlates of a 30-year period-middle age to old age. Jour¬
the Wiggins MMPI content scales. Journal nal of Consulting and Clinical Psychology,
of Consulting and Clinical Psychology, 46, 47, 517-524.
1349-1356. Leonard, C. V. (1977). The MMPI as a suicide
490 References

predictor. Journal of Consulting and Clinical MacAndrew, C. (1986). Toward the psychometric
Psychology, 45, 367-377. detection of substance misuse in young men:
Lester, D., & Clopton, J. R. (1979). Suicide and The SAP scale. Journal of Studies on Alco¬
overcontrol. Psychological Reports, 44, 758. hol, 47, 161-166.
Levitt, E. E. (1978). A note on “MMPI scale de¬ MacAndrew, C. (1987). An examination of the ap¬
velopment methodology.” Journal of Per¬ plicability of the Substance Abuse Proclivity
sonality Assessment, 42, 503-504. scale to young adult males. Psychology of
Levitt, E. E. (1989). The clinical application of Addictive Behaviors, 1, 140-145.
MMPI special scales. Hillsdale, NJ: Erl- MacAndrew, C. (1988). Differences in the self-de¬
baum. pictions of female alcoholics and psychiatric
Levitt, E. E. (1990). A structural analysis of the outpatients: Towards a depiction of the
impact of MMPI-2 on the MMPI-1. Journal modal female alcoholic. Journal of Studies
of Personality Assessment, 55, 562-577. on Alcohol, 49, 71-77.
Levy, L. H. (1963). Psychological interpretation. MacAndrew, C., & Geertsma, R. H. (1964). A cri¬
New York: Holt, Rinehart, & Winston. tique of alcoholism scales derived from the
Lichenstein, E. & Bryan, J. H. (1966). Short-term MMPI. Quarterly Journal of Studies on Al¬
stability of MMPI profiles. Journal of Con¬ cohol, 25, 68-76.
sulting Psychology, 30, 172-174. Mallory, C. H., & Walker, C. E. (1972). MMPI
Lingoes, J. C. (1960). MMPI factors of the Harris O-H scale responses of assaultive and non-
and the Wiener subscales. Journal of Con¬ assaultive prisoners and associated life his¬
sulting Psychology, 24, 74-83. tory variables. Educational and Psycho¬
Liske, R., & McCormick, R. (1976). MMPI pro¬ logical Measurement, 32, 1125-1128.
files compared for black and white hospital¬ Maloney, M. P., Duvall, S. W., & Friesen, J.
ized veterans. Newsletter for Research in (1980). Evaluation of response consistency
Mental Health and Behavioral Sciences, 18, on the MMPI. Psychological Reports, 46,
30-32. 295-298.
Listiak, R. L., & Stone, L. A. (1971). Psychophys¬ Manosevitz, M. (1971). Education and MMPI Mf
ical approach to clinical judgment of low T scores in homosexual and heterosexual
scores on the MMPI. Journal of Consulting males. Journal of Consulting and Clinical
and Clinical Psychology, 36, 447. Psychology, 36, 395-399.
Little, K. B., & Fisher, J. (1958). Two new experi¬ Marks, P. A., & Seeman, W. (1963). The actuarial
mental scales of the MMPI. Journal of Con¬ description of personality: An atlas for use
sulting Psychology, 22, 305-306. with the MMPI. Baltimore: Williams & Wil¬
Loper, R. G., Kammeier, M. L., & Hoffmann, H. kins.
(1973). MMPI characteristics of college Marks, P. A., Seeman, W., & Haller, D. L.
freshman males who later became alcoholics. (1974). The actuarial use of the MMPI with
Journal of Abnormal Psychology, 82, 159— adolescents and adults. Baltimore: Williams
162. & Wilkins.
Louks, J. L., Freeman, C. W., & Calsyn, D. A. Marsella, A. J., Sanborn, K. O., Kameoka, V.,
(1978). Personality organization as an aspect Shizuru, L., & Brennan, J. (1975). Cross-val¬
of back pain in a medical setting. Journal of idation of self-report measures of depression
Personality Assessment, 42, 152-158. among normal populations of Japanese, Chi¬
Lubin, B., Larsen, R. M., Matarazzo, J. D., & nese, and Caucasian ancestry. Journal of
Seever, M. (1985). Psychological test usage Clinical Psychology, 31, 281-287.
patterns in five professional settings. Ameri¬ Martin, P. W., & Greene, R. L. (1979, March). A
can Psychologist, 40, 857-861. comparison of the Pepper and Strong and
MacAndrew, C. (1965). The differentiation of Serkownek subscales for the Masculinity-
male alcoholic outpatients from nonalco¬ Femininity scale of the MMPI. Paper pre¬
holic psychiatric outpatients by means of the sented at the meeting of the Society for
MMPI. Quarterly Journal of Studies on Al¬ Personality Assessment, Scottsdale, AZ.
cohol, 26, 238-246. Matarazzo, J. D. (1955). MMPI validity scores as
MacAndrew, C. (1981). What the MAC scale tells a function of increasing levels of anxiety.
us about men alcoholics: An interpretive re¬ Journal of Consulting Psychology, 19, 213—
view. Journal of Studies on Alcohol, 42, 217.
604-625. McAnulty, D. P., Rappaport, N. B., & McAnulty,
 References 491

R. D. (1985). An a posteriori investigation of normality or control factor in personality

standard MMPI validity scales. Psychologi¬ testing. Psychological Monographs, 4
cal Reports, 57, 95-98. (Whole No. 274).
McCreary, C. (1975). Personality profiles of per¬ Meehl, P. E. (1956). Wanted—A good cookbook.
sons convicted of indecent exposure. Journal American Psychologist, 11, 263-272.
of Clinical Psychology, 31, 260-262. Meehl, P. E. (1959). A comparison of clinicians
McCreary, C., & Padilla, E. (1977). MMPI differ¬ with five statistical methods of identifying
ences among Black, Mexican-American, and psychotic MMPI profiles. Journal of Coun¬
White male offenders. Journal of Clinical seling Psychology, 6, 102-109.
Psychology, 33, 171-177. Meehl, P. E., & Dahlstrom, W. G. (1960). Objec¬
McDonald, R. L., & Gynther, M. D. (1962). tive configural rules for discriminating psy¬
MMPI norms for southern adolescent Ne¬ chotic from neurotic MMPI profiles. Journal
groes. Journal of Social Psychology, 58, of Consulting Psychology, 24, 375-387.
277-282. Meehl, P. E., & Hathaway, S. R. (1946). The K
McDonald, R. L., & Gynther, M. D. (1963). factor as a suppressor variable in the MMPI.
MMPI differences associated with sex, race, Journal of Applied Psychology, 30, 525-564.
and class in two adolescent samples. Journal Megargee, E. I. (1972). The California Psycholog¬
of Consulting Psychology, 27, 112-116. ical Inventory handbook. San Francisco:
McGill, J. C. (1980). MMPI score differences Jossey-Bass.
among Anglo, Black, and Mexican-Ameri¬ Megargee, E. I. (1985). Assessing alcoholism and
can welfare recipients. Journal of Clinical drug abuse with the MMPI: Implication for
Psychology, 36, 147-151. employment screening. In C. D. Spielberger
McGrath, R. E., & O’Malley, W. B. (1986). The & J. N. Butcher (Eds.), Advances in person¬
assessment of denial and physical com¬ ality assessment (Vol. 5, pp. 1-39). Hillsdale,
plaints: The validity of the Hy scale and asso¬ NJ: Erlbaum.
ciated MMPI signs. Journal of Clinical Megargee, E. I., & Cook, P. E. (1975). Negative
Psychology, 42, 754-760. response bias and the MMPI overcontrolled-
McGrath, R. E., O’Malley, W. B., & Dura, J. R. hostility scale: A response to Deiker. Journal
(1986). Alternative scoring system of re¬ of Consulting and Clinical Psychology, 43,
peated items on the MMPI: Caveat emptor. 725-729.
Journal of Personality Assessment, 50, 182— Megargee, E. I., Cook, P. E., & Mendelsohn, G.
185. A. (1967). Development and validation of an
McKegney, F. P. (1965). An item analysis of the MMPI scale of assaultiveness in over¬
MMPI F scale in juvenile delinquents. Jour¬ controlled individuals. Journal of Abnormal
nal of Clinical Psychology, 21, 201-205. Psychology, 72, 519-528.
McKinley, J. C., & Hathaway, S. R. (1940). A Megargee, E. I., & Mendelsohn, G. A. (1962). A
multiphasic personality schedule (Minne¬ cross-validation of twelve MMPI indices of
sota): II. A differential study of hypochon¬ hostility and control. Journal of Abnormal
driasis. Journal of Psychology, 10, 255-268. and Social Psychology, 65, 431 -438.
McKinley, J. C., & Hathaway, S. R. (1942). A Mehlman, B., & Rand, M. E. (1960). Face validity
multiphasic personality schedule (Minne¬ of the MMPI. Journal of General Psychol¬
sota): IV. Psychasthenia. Journal of Applied ogy, 63, 171-178.
Psychology, 26, 614-624. Meikle, S., & Gerritse, R. (1970). MMPI “cook¬
McKinley, J. C., & Hathaway, S. R. (1944). The book” pattern frequencies in a psychiatric
MMPI: V. Hysteria, hypomania, and psy¬ unit. Journal of Clinical Psychology, 26, 82-84.
chopathic deviate. Journal of Applied Psy¬ Mezzich, J. E., Damarin, F. L., & Erickson, J. R.
chology, 28, 153-174. (1974). Comparative validity of strategies
McKinley, J. C., Hathaway, S. R., & Meehl, P. E. and indices for differential diagnosis of de¬
(1948). The MMPI: VI. The K scale. Journal pressive states from other psychiatric condi¬
of Consulting Psychology, 12, 20-31. tions using the MMPI. Journal of Consulting
McLaughlin, J. D., Helmes, E., & Howe, M. G. and Clinical Psychology, 42, 691-698.
(1983). Note on the reliability of three MMPI Miller, C., Knapp, S. C., & Daniels, C. W. (1968).
short forms. Journal of Personality Assess¬ MMPI study of Negro mental hygiene clinic
ment, 47, 357-358. patients. Journal of Abnormal Psychology,
Meehl, P. E. (1945). An investigation of a general 73, 168-173.
492 References

Miller, C., Wertz, C., & Counts, S. (1961). Racial itentiary on two scales of the MMPI. Journal
differences on the MMPI. Journal of Clinical of Clinical Psychology, 18, 137-139.
Psychology, 17, 159-161. Murray, J. B., Munley, M. J., & Gilbart, T. E.
Miller, H. R., & Streiner, D. L. (1985). The Har- (1965). The Pd scale of the MMPI for college
ris-Lingoes subscales: Fact or fiction? Jour¬ students. Journal of Clinical Psychology, 21,
nal of Clinical Psychology, 41, 45-51. 48-51.
Miller, H. R., & Streiner, D. L. (1986). Differ¬ Nakamura, C. Y. (1960). Validity of K scale
ences in MMPI profiles with the norms of (MMPI) in college counseling. Journal of
Colligan et al. Journal of Consulting and Counseling Psychology, 7, 108-115.
Clinical Psychology, 54, 843-845. Navran, L. (1954). A rationally derived MMPI
Monroe, J. J., Miller, J. S., & Lyle, Jr., W. H. scale to measure dependence. Journal of
(1964). The extension of psychopathic devi- Consulting Psychology, 18, 192.
ancy scales for the screening of addict pa¬ Nelson, L. D. (1987). Measuring depression in a
tients. Educational and Psychological clinical population using the MMPI. Journal
Measurement, 24, 47-56. of Consulting and Clinical Psychology, 55,
Montgomery, G. T., & Orozco, S. (1985). Mexi¬ 788-790.
can Americans’ performance on the MMPI Newmark, C. S., Gentry, L., Simpson, M., &
as a function of level of acculturation. Jour¬ Jones, T. (1978). MMPI criteria for diagnos¬
nal of Clinical Psychology, 41, 203-212. ing schizophrenia. Journal of Personality
Moore, C. D., & Handal, P. J. (1980). Ado¬ Assessment, 42, 366-373.
lescents’ MMPI performance, cynicism, es¬ Newmark, C. S., & Hutchins, T. C. (1980). Age
trangement, and personal adjustment as a and MMPI indices of schizophrenia. Journal
function of race and sex. Journal of Clinical of Clinical Psychology, 36, 768-769.
Psychology, 36, 932-936. Newmark, C. S., Ziff, D. R., Finch, Jr., A. J., &
Moreland, K. L. (1984). Comparative validity of Kendall, P. C. (1978). Comparing the empir¬
the MMPI and two short forms: Psychiatric ical validity of the standard form with two
ratings. Journal of Personality Assessment, abbreviated MMPIs. Journal of Consulting
48, 265-270. and Clinical Psychology, 46, 53-61.
Morey, L. C., Blashfield, R. K., Webb, W. W., & Newton, J. R. (1968). Clinical normative data for
Jewell, J. (1988). MMPI scales for DSM-III MMPI special scales: Critical items, manifest
personality disorders: A preliminary valida¬ anxiety, and repression-sensitiziation. Jour¬
tion study. Journal of Clinical Psychology, nal of Clinical Psychology, 24, 427-430.
44, 47-50. Nichols, D. S. (1987). Interpreting the Wiggins
Morey, L. C., & Smith, M. R. (1988). Personality MMPI content scales. Clinical notes on the
disorders. In R. L. Greene (Ed.), The MMPI: MMPI, No. 10. Minneapolis: National Com¬
Use with specific populations (pp. 110-158). puter Systems.
Philadelphia: Grune & Stratton. Nichols, D. S. (1988). Mood disorders. In R. L.
Morey, L. C., Waugh, M. H., & Blashfield, R. K. Greene (Ed.), The MMPI: Use in specific
(1985). MMPI scales for DSM-III personal¬ populations (pp. 74-109). San Antonio:
ity disorders: Their derivation and correlates. Grune & Stratton.
Journal of Personality Assessment, 49, 245- Nichols, D. S., Greene, R. L., & Schmolck, P.
251. (1989). Criteria for assessing inconsistent
Muller, B. P., & Bruno, L. N. (1988, March). The patterns of item endorsement on the MMPI:
effects of ethnicity on personality assessment Rationale, development, and empirical trials.
in police candidates. Paper presented at the Journal of Clinical Psychology, 45, 239-250.
23rd Annual Symposium on Recent Develop¬ O’Connor, J. P., & Stefic, E. C. (1959). Some pat¬
ments in the Use of the MMPI, St. Peters¬ terns of hypochondriasis. Educational and
burg, FL. Psychological Measurement, 19, 363-371.
Muller, B. P. & Bruno, L. N. (1990, August). The Olmsted, D. W., & Monachesi, E. D. (1956). A va¬
myth of ethnic effects on the MMPI/MMPI- lidity check on MMPI scales of responsibility
2. Paper presented at the annual meeting of and dominance. Journal of Abnormal and
the American Psychological Association, Social Psychology, 53, 140-141.
Boston. Osborne, D., & Colligan, R. C. (1986). Linear
Murphree, H. B., Karabelas, M. J., & Bryan, L. equations for the development of non-nor-
L. (1962). Scores of inmates of a federal pen¬ malized T score tables based on the contem-
 References 493

porary normative study of the MMPI. Jour¬ Personality and Social Psychology, 46, 598-
nal of Clinical Psychology, 42, 482-484. 609.
Osborne, D., Colligan, R. C., & Offord, K. P. Paulhus, D. L. (1986). Self-deception and impres¬
(1986). Normative tables for the F - K index sion management in test responses. In A. An-
of the MMPI based on a contemporary nor¬ gleitner & J. S. Wiggins (Eds.), Personality
mal sample. Journal of Clinical Psychology, assessment via questionnaires: Current issues
42, 593-595. in theory and measurement (pp. 143-165).
Otto, R. K., Lang, A. R., Megargee, E. I., & Berlin: Springer-Verlag.
Rosenblatt, A. I. (1988). Ability of alcoholics Paulson, M. J., Afifi, A. A., Thomason, M. L., &
to escape detection by the MMPI. Journal of Chaleff, A. (1974). The MMPI: A descriptive
Consulting and Clinical Psychology, 56, measure of psychopathology in abusive par¬
452-457. ents. Journal of Clinical Psychology, 30,
Overall, J. E., & Gomez-Mont, F. (1974). The 387-390.
MMPI-168 for psychiatric screening. Educa¬ Payne, F. D., & Wiggins, J. S. (1972). MMPI pro¬
tional and Psychological Measurement, 34, file types and the self-report of psychiatric
315-319. patients. Journal of Abnormal Psychology,
Padilla, E. R., Olmedo, E. L., & Loya, F. (1982). 79, 1-8.
Acculturation and the MMPI performance Penk, W. E., Keane, T., Robinowitz, R., Fowler,
of Chicano and Anglo college students. Jour¬ D. R., Bell, W. E., & Finkelstein, A. (1988).
nal of Behavioral Sciences, 4, 451-466. Post-traumatic stress disorder. In R. L.
Page, J., Landis, C., & Katz, S. E. (1934). Schizo¬ Greene (Ed.), The MMPI: Use in specific
phrenic traits in the functional psychoses and populations (pp. 198-213). San Antonio:
in normal individuals. American Journal of Grune & Stratton.
Psychiatry, 90, 1213-1225. Penk, W. E., Roberts, W. R., Robinowitz, R.,
Page, R. D., & Bozlee, S. (1982). A cross-cultural Dolan, M. P., Atkins, H. G., & Woodward,
MMPI comparison of alcoholics. Psycholog¬ W. A. (1982). MMPI differences of black
ical Reports, 50, 639-646. and white polydrug abusers seeking treat¬
Pancoast, D. L., & Archer, R. P. (1989). Original ment. Journal of Consulting and Clinical
adult MMPI norms in normal samples: A re¬ Psychology, 50, 463-465.
view with implications for future develop¬ Penk, W. E., Robinowitz, R., Roberts, W. R.,
ments. Journal of Personality Assessment, Dolan, M. P., & Atkins, H. G. (1981).
53, 376-395. MMPI differences of male Hispanic-Ameri-
Panton, J. H. (1959a). Inmate personality differ¬ can, black, and white heroin addicts. Journal
ences related to recidivism, age, and race as of Consulting and Clinical Psychology, 49,
measured by the MMPI. Journal of Correc¬ 488-490.
tional Psychology, 4, 28-35. Penk, W. E., Woodward, W. A., Robinowitz, R.,
Panton, J. H. (1959b). The response of prison in¬ & Hess, J. L. (1978). Differences in MMPI
mates to MMPI subscales. Journal of Social scores of black and white compulsive heroin
Therapy, 5, 233-237. users. Journal of Abnormal Psychology, 87,
Patalano, F. (1978). Personality dimensions of 505-513.
drug abusers who enter a drug-free therapeu¬ Pepper, L, J., & Strong, P. N. (1958). Judgmental
tic community. Psychological Reports, 42, subscales for the Mf scale of the MMPI. Un¬
1063-1069. published manuscript.
Patterson, E. T., Charles, H. L., Woodward, W. Persons, R. W., & Marks, P. A. (1971). The vio¬
A., Roberts, W. R., & Penk, W. E. (1981). lent 4-3 MMPI personality type. Journal of
Differences in measures of personality and Consulting and Clinical Psychology, 36,
family environment among black and white 189-196.
alcoholics. Journal of Consulting and Clini¬ Peteroy, E. T., & Pirrello, P. E. (1982). Compari¬
cal Psychology, 49, 1-9. son of MMPI scales for black and white hos¬
Pauker, J. D. (1966). Identification of MMPI pro¬ pitalized samples. Psychological Reports, 50,
file types in a female, inpatient, psychiatric 662.
setting using the Marks and Seeman rules. Peterson, C. D. (1989). Masculinity and feminin¬
Journal of Consulting Psychology, 30, 90. ity as independent dimensions on the MMPI-
Paulhus, D. L. (1984). Two-component models of 2. Unpublished doctoral dissertation,
social desirable responding. Journal of University of North Carolina, Chapel Hill.
494 References

Peterson, D. R. (1954). The diagnosis of subclini- pendency measures and welfare chronicity.
cal schizophrenia. Journal of Consulting Journal of Consulting Psychology, 26, 559—
Psychology, 18, 198-200. 560.
Pichot, P., Perse, J., Lebeaux, M. O., Dureau, J. Quay, H., & Rowell, J. T. (1955). The validity of a
L., Perez, C., & Rychewaert, A. (1972). La schizophrenic screening scale of the MMPI.
personalite des sujets presentant des douleurs Journal of Clinical Psychology, 11, 92-93.
dorsales fonctionnelles valeur de l’inventaire Quinsey, V. L., Maguire, A., & Varney, G. W.
Multiphasique de Personalite du Minnesota. (1983). Assertion and overcontrolled hostil¬
Revue de Psychoiogie Applique, 22, 145— ity among mentally disordered murderers.
172. Journal of Consulting and Clinical Psychol¬
Plemons, G. (1977). A comparison of MMPI ogy, 51, 550-556.
scores of Anglo- and Mexican-American psy¬ Rader, C. M. (1977). MMPI profile types of ex¬
chiatric patients. Journal of Consulting and posers, rapists, and assaulters in a court ser¬
Clinical Psychology, 45, 149-150. vices population. Journal of Consulting and
Pollack, D., & Shore, J. H. (1980). Validity of the Clinical Psychology, 45, 61-69.
MMPI with Native Americans. American Rand, S. W. (1979). Correspondence between psy¬
Journal of Psychiatry, 137, 946-950. chological reports based on the Mini-Mult
Posey, C. D., & Hess, A. K. (1984). The fakability and the MMPI. Journal of Personality As¬
of subtle and obvious measures of aggression sessment, 43, 160-163.
by male prisoners. Journal of Personality Reilley, R. R., & Knight, G. E. (1970). MMPI
Assessment, 48, 137-144. scores of Mexican-American college stu¬
Posey, C. D., & Hess, A. K. (1985). Aggressive re¬ dents. Journal of College Student Personnel,
sponse sets and subtle-obvious MMPI scale 11, 419-422.
distinctions in male offenders. Journal of Rich, C. C., & Davis, H. G. (1969). Concurrent
Personality Assessment, 49, 235-239. validity of MMPI alcoholism scales. Journal
Post, R. D., Clopton, J. R., Keefer, G., Rosen¬ of Clinical Psychology, 25, 425-426.
berg, D., Blyth, L. S., & Stein, M. (1986). Ries, H. A. (1966). The MMPI K scale as a predic¬
MMPI predictors of mania among psychiat¬ tor of prognosis. Journal of Clinical Psy¬
ric patients. Journal of Personality Assess- chology, 22, 212-213.
ment, 50, 248-256. Robins, L. N., Helzer, J. E., Weisman, M. M.,
Post, R. D., & Gasparikova-Krasnec, M. (1979). Orvaschel, H., Gruenberg, E., Burke, Jr., J.
MMPI validity scales and behavioral distur¬ D., & Regier, D. A. (1984). Lifetime preva¬
bance in psychiatric inpatients. Journal of lence rates of specific psychiatric disorders in
Personality Assessment, 43, 155-159. three sites. Archives of General Psychiatry,
Poythress, N. G., & Blaney, P. H. (1978). The va¬ 41, 949-958.
lidity of MMPI interpretations based on the Rogers, R. (1983). Malingering or random? A re¬
Mini-Mult and the FAM. Journal of Person¬ search note on obvious vs. subtle subscales of
ality Assessment, 42, 143-147. the MMPI. Journal of Clinical Psychology,
Pritchard, D. A., & Rosenblatt, A. (1980a). Racial 39, 257-258.
bias in the MMPI: A methodological review. Rogers, R., Dolmetsch, R., & Cavanaugh, Jr., J.
Journal of Consulting and Clinical Psychol¬ L. (1983). Identification of random respond¬
ogy, 48, 263-267. ers on MMPI protocols. Journal of Person¬
Pritchard, D. A., & Rosenblatt, A. (1980b). Reply ality Assessment, 47, 364-368.
to Gynther and Green. Journal of Consulting Rogers, R., Harris, M., & Thatcher, A. A. (1983).
and Clinical Psychology, 48, 273-274. Identification of random responders on the
Prokop, C. K. (1986). Hysteria scale elevations in MMPI: An actuarial approach. Psychologi¬
low back pain patients: A risk factor for mis¬ cal Reports, 53, 1171-1174.
diagnosis? Journal of Consulting and Clini¬ Rohan, W. P., Tatro, R. L., & Rotman, S. R.
cal Psychology, 54, 558-562. (1969). MMPI changes in alcoholics during
Prokop, C. K. (1988). Chronic pain. In R. L. hospitalization. Quarterly Journal of Studies
Greene (Ed.), The MMPI: Use with specific on Alcohol, 30, 389-400.
populations (pp. 22-49). San Antonio: Rosen, A. (1952). Reliability of MMPI scales.
Grune & Stratton. American Psychologist, 7, 341.
Pruitt, W. A., & Van de Castle, R. L. (1962). De¬ Rosen, A. (1954). Detection of suicidal patients:
 References 495

An example of some limitations in the predic¬ the 77th Annual Convention of the APA, 4,
tion of infrequent events. Journal of Con¬ 519-520.
sulting Psychology, 18, 397-403. Schwartz, M. S., & Krupp, N. E. (1971). The
Rosen, A. (1958). Differentiation of diagnostic MMPI “conversion V” among 50,000 medi¬
groups by individual MMPI scales. Journal cal patients: A study of incidence, criteria,
of Consulting Psychology, 22, 453-457. and profile elevation. Journal of Clinical
Rosen, A. C. (1974). Brief report of MMPI char¬ Psychology, 27, 89-95.
acteristics of sexual deviation. Psychological Schwartz, M. S., Osborne, D., & Krupp, N. E.
Reports, 35, 73-74. (1972). Moderating effects of age and sex on
Rosenblatt, A. I., & Pritchard, D. A. (1978). the association of medical diagnoses and
Moderators of racial differences on the 1-3/3-1 MMPI profiles. Journal of Clinical
MMPI. Journal of Consulting and Clinical Psychology, 28, 502-505.
Psychology, 46, 1572-1573. Serkownek, K. (1975). Subscales for Scales 5 andO
Rubin, H. (1954). Validity of a critical-item scale of the MMPI. Unpublished manuscript.
for schizophrenia on the MMPI. Journal of Sheppard, D., Smith, G. T., & Rosenbaum, G.
Consulting Psychology, 18, 219-220. (1988). Use of MMPI subtypes in predicting
Ruch, F. L., & Ruch, W. W. (1967). The K factor completion of a residential alcoholism treat¬
as a (validity) suppressor variable in predict¬ ment program. Journal of Consulting and
ing success in selling. Journal of Applied Clinical Psychology, 56, 590-596.
Psychology, 51, 201-204. Shore, R. E. (1976). A statistical note on “Differ¬
Saunders, T. R., & Gravitz, M. A. (1974). Sex dif¬ ential misdiagnosis of blacks and whites by
ferences in the endorsement of MMPI critical the MMPI.’’ Journal of Personality Assess¬
items. Journal of Clinical Psychology, 30, ment, 40, 21-23.
557-558. Shultz, T. D., Gibeau, P. J., & Barry, S. M.
Scagnelli, J. (1975). The significance of depen¬ (1968). Utility of MMPI “cookbooks.”
dency in the paranoid syndrome. Journal of Journal of Clinical Psychology, 24, 430-433.
Clinical Psychology, 31, 29-34. Shweder, R. A. (1977a). Illusory correlation and
Schenkenberg, T., Gottfredson, D. K., & the MMPI controversy. Journal of Consult¬
Christensen, P. (1984). Age differences in ing and Clinical Psychology, 45, 917-924.
MMPI scale scores from 1,189 psychiatric Shweder, R. A. (1977b). Illusory correlation and
patients. Journal of Clinical Psychology, 40, the MMPI controversy: Reply to some of the
1420-1426. allusions and elusions in Block’s and
Schlenger, W. E., & Kulka, R. A. (1987, August). Edwards’ commentaries. Journal of Consult¬
Performance of the Keane-Fairbank MMPI ing and Clinical Psychology, 45, 936-940.
scale and other self-report measures in identi¬ Silver, R. J., & Sines, L. K. (1962). Diagnostic ef¬
fying post-traumatic stress disorder. Paper ficiency of the MMPI with and without the K
presented at the annual meeting of the Amer¬ correction. Journal of Clinical Psychology,
ican Psychological Association, New York. 18, 312-314.
Schmidt, H. O. (1948). Notes on the MMPI: The Simon, W., & Gilberstadt, H. (1958). Analysis of
K factor. Journal of Consulting Psychology, the personality structure of 26 actual sui¬
12, 337-342. cides. Journal of Nervous and Mental Dis¬
Schneck, J. M. (1948). Clinical evaluation of theF ease, 127, 555-557.
scale on the MMPI. American Journal of Sines, J. O. (1966). Actuarial methods in personal¬
Psychiatry, 104, 440-442. ity assessment. In B. A. Maher (Ed.), Prog¬
Schretlen, D. (1990). A limitation of using the ress in experimental personality research (pp.
Wiener and Harmon obvious and subtle 133-193). New York: Academic Press.
scales to detect faking on the MMPI. Journal Sines, J. O. (1977). M-F: Bipolar and probably
of Clinical Psychology, 46, 782-786. multidimensional. Journal of Clinical Psy¬
Schwartz, M. F., & Graham, J. R. (1979). Con¬ chology, 33, 1038-1041.
struct validity of the MacAndrew alcoholism Sines, L. K., Baucom, D. H., & Gruba, G. H.
scale. Journal of Consulting and Clinical (1979). A validity scale sign calling for cau¬
Psychology, 4 7, 1090-1095. tion in the interpretation of MMPIs among
Schwartz, M. S. (1969). “Organicity” and the psychiatric inpatients. Journal of Personality
MMPI 1-3-9 and 2-9 codes. Proceedings of Assessment, 43, 604-607.
496 References

Singer, M. I. (1970). Comparison of indicators of between Asian-American and non-Asian stu¬

homosexuality on the MMPI. Journal of dents utilizing a student health psychiatric
Consulting and Clinical Psychology, 34, 15-18. clinic. Journal of Counseling Psychology,
Smith, C. P., & Graham, J. R. (1981). Behavioral 27,423-427.
correlates for the MMPI standard Fscale and Super, D. E. (1942). The Bernreuter Personality
for a modified F scale for black and white Inventory: A review of research. Psychologi¬
psychiatric patients. Journal of Consulting cal Bulletin, 39, 94-125.
and Clinical Psychology, 49, 455-459. Sutker, P. B., Allain, A. N., & Geyer, S. (1978).
Smith, E. E. (1959). Defensiveness, insight, and Female criminal violence and differential
the K scale. Journal of Consulting Psychol¬ MMPI characteristics. Journal of Consulting
ogy , 23,275-277. and Clinical Psychology, 46, 1141-1143.
Snyder, D. K. (1989). Assessing chronic pain with Sutker, P. B., Archer, R. P., & Allain, A. N.
the MMPI. In T. W. Miller (Ed.), Chronic (1978). Drug abuse patterns, personality
pain: Clinical issues in health care manage¬ characteristics, and relationships with sex,
ment. Madison, CT: International Universi¬ race, and sensation seeking. Journal of Con¬
ties Press. sulting and Clinical Psychology, 46, 1374-
Snyder, D. K., Kline, R. B., & Podany, E. C. 1378.
(1985). Comparison of external correlates of Sutker, P. B., Archer, R. P., & Allain, A. N.
MMPI substance abuse scales across sex and (1980). Psychopathology of drug abusers:
race. Journal of Consulting and Clinical Psy¬ Sex and ethnic considerations. The Interna¬
chology, 53, 520-525. tional Journal of the Addictions, 15, 605-
Snyter, C. M., & Graham, J. R. (1984). The utility 613.
of subtle and obvious MMPI subscales based Sutker, P. B., & Kilpatrick, D. G. (1973). Person¬
on scale-specific ratings. Journal of Clinical ality, biographical, and racial correlates of
Psychology, 40, 981-985. sexual attitudes and behavior. Proceedings
Spirito, A., Faust, D., Myers, B., & Bechtel, D. of the American Psychological Association,
(1988). Clinical utility of the MMPI in the 8, 261-262.
evaluation of adolescent suicide attempters. Sweetland, A., & Quay, H. (1953). A note on the
Journal of Personality Assessment, 52, 204- K scale of the MMPI. Journal of Consulting
211. Psychology, 17, 314-316.
Stanton, J. M. (1956). Group personality profile Swenson, W. M. (1961). Structured personality
related to aspects of antisocial behavior. testing in the aged: An MMPI study of the
Journal of Criminal Law, Criminology, and gerontic population. Journal of Clinical Psy¬
Police Science, 47, 340-349. chology, 17, 302-304.
Stein, K. B. (1968). The TSC scales: The outcome Swenson, W. M., Pearson, J. S., & Osborne, D.
of a cluster analysis of the 550 MMPI items. (1973). An MMPI source book: Basic item,
In P. McReynolds (Ed,), Advances in psy¬ scale, and pattern data on 50,000 medical pa¬
chological assessment. (Vol. I, pp. 80-104). tients. Minneapolis: University of Minnesota
Palo Alto CA: Science & Behavior Books. Press.
Sternbach, R. A., Wolf, S. R., Murphy, R. W., & Tamkin, A. S., & Scherer, I. W. (1957). What is
Akeson, W. H. (1973). Traits of pain pa¬ measured by the “Cannot Say” scale of the
tients: The low-back “loser.” Psychosomat- group MMPI? Journal of Consulting Psy¬
ics, 14, 226-229. chology, 21, 370.
Stevens, M. R., & Reilley, R. R. (1980). MMPI Tanner, B. A. (1990). Composite descriptions as¬
short forms: A literature review. Journal of sociated with rare MMPI two-point code
Personality Assessment, 44, 368-376. types: Codes that involve Scale 5. Journal of
Strauss, M. E., Gynther, M. D., & Wall- Clinical Psychology, 46, 425-431.
hermfechtel, J. (1974). Differential misdiag¬ Tarter, R. E., & Perley, R. N. (1975). Clinical and
nosis of blacks and whites by the MMPI. perceptual characteristics of paranoids and
Journal of Personality Assessment, 38, 55- paranoid schizophrenics. Journal of Clinical
60. Psychology, 31, 42-44.
Streiner, D. L., & Miller, H. R. (1986). Can a good Taulbee, E. S., & Sisson, B. D. (1957). Configura¬
short form of the MMPI ever be developed? tional analysis of MMPI profiles of psychiat¬
Journal of Clinical Psychology, 42, 109-113. ric groups. Journal of Consulting Psy¬
Sue, S., & Sue, D. W. (1974). MMPI comparisons chology, 27,413-417.
 References 497

Terman, L. M., & Miles, C. C. (1938). Manual of and zero items in a psychiatric outpatient
information and directions for use of Atti¬ group. Journal of Personality Assessment,
tude-Interest Analysis Test. New York: Mc¬ 38, 61-64.
Graw-Hill. Vestre, N. D., & Watson, C. G. (1972). Behavioral
Towne, W. S., & Tsushima, W. T. (1978). The use correlates of the MMPI Paranoia scale. Psy¬
of the low back and the dorsal scales in the chological Reports, 31, 851-854.
identification of functional low back pa¬ Vincent, K. R. (1984). MMPI-168codebook. Nor¬
tients. Journal of Clinical Psychology, 34, wood, NJ: Ablex.
88-91. Vincent, N. M. P., Linsz, N. L., & Greene, M. I.
Truscott, D. (1990). Assessment of Overcontrolled (1966). The L scale of the MMPI as an index
Hostility in adolescence. Psychological As¬ of falsification. Journal of Clinical Psychol¬
sessment: A Journal of Consulting and Clin¬ ogy, 22, 214-215.
ical Psychology, 2, 145-148. Wales, B., & Seeman, W. (1968). A new method
Tsushima, W. T., & Onorato, V. A. (1982). Com¬ for detecting the fake good response set on
parison of MMPI scores of white and Japan- the MMPI. Journal of Clinical Psychology,
ese-American medical patients. Journal of 24, 211-216.
Consulting and Clinical Psychology, 50, Wales, B., & Seeman, W. (1972). Instructional
150-151. sets and MMPI items. Journal of Personality
Tyler, F. T. (1951). A factorial analysis of fifteen Assessment, 36, 282-286.
MMPI scales. Journal of Consulting Psy¬ Walker, C. E., & Ward, J. (1969). Identification
chology, 15, 451-456. and elimination of offensive items from the
Tyler, F. T., & Michaelis, J. U. (1953). 76-scores MMPI. Journal of Projective Techniques
applied to MMPI scales for college women. and Personality Assessment, 33, 385-388.
Educational and Psychological Measure¬ Walters, G. D. (1984). Identifying schizophrenia
ment, 13, 459-466. by means of Scale 8 Sc of the MMPI. Journal
Uecker, A. E., Boutilier, L. R., & Richardson, E. of Personality Assessment, 48, 390-391.
H. (1980). “Indianism” and MMPI scores of Walters, G. D. (1985). Scale 4 {Pd) of the MMPI
men alcoholics. Journal of Studies on Alco¬ and the diagnosis Antisocial Personality. Jour¬
hol, 47,357-362. nal of Personality Assessment, 49, 474-476.
Velasquez, R. J. (1984). An atlas of MMPI group Walters, G. D. (1986). Screening for psychopa¬
profiles on Mexican Americans. Los Ange¬ thology in groups of black and white prison
les: Spanish Speaking Mental Health Re¬ inmates by means of the MMPI. Journal of
search Center. Personality Assessment, 50, 257-264.
Velasquez, R. J., & Callahan, W. J. (1987). Walters, G. D. (1988a). Assessing dissimulation
MMPI differences among three diagnostic and denial on the MMPI in a sample of max-
groups of Mexican-American state hospital ium security, male inmates. Journal of Per¬
patients. Psychological Reports, 60, 1071- sonality Assessment, 52, 465-474.
1074. Walters, G. D. (1988b). Schizophrenia. In R. L.
Velasquez, R. J., & Callahan, W. J. (1990a). Greene (Ed.), The MMPI: Use in specific
MMPI comparisons of Hispanic- and White- populations (pp. 50-73). San Antonio:
American veterans seeking treatment for al¬ Grune & Stratton.
coholism. Psychological Reports, 67, 95-98. Walters, G. D., & Greene, R. L. (1988). Differen¬
Velasquez, R. J., & Callahan, W. J. (1990b). tiating between schizophrenic and manic in¬
MMPIs of Hispanic, black, and white DSM- patients by means of the MMPI. Journal of
III schizophrenics. Psychological Reports, Personality Assessment, 52, 91-95.
66, 819-822. Walters, G. D., Greene, R. L., & Jeffrey, T. B.
Velasquez, R. J., Callahan, W. J., & Carrillo, R. (1984). Discriminating between alcoholic and
(1989). MMPI profiles of Hispanic-Ameri- nonalcoholic blacks and whites on the
can inpatient and outpatient sex offenders. MMPI. Journal of Personality Assessment,
Psychological Reports, 65, 1055-1058. 48, 486-488.
Velasquez, R. J., & Gimenez, L. (1987). MMPI Walters, G. D., Greene, R. L., Jeffrey, T. B.,
differences among three diagnostic groups of Kruzich, D. J., & Haskin, J. J. (1983). Racial
Mexican-American state hospital patients. variations on the MacAndrew Alcoholism
Psychological Reports, 60, 1071-1074. scale of the MMPI. Journal of Consulting
Vesprani, G. J., & Seeman, W. (1974). MMPI 26 and Clinical Psychology, 51, 947-948.
498 References

Walters, G. D., White, T. W., & Greene, R. L. Weiss, R. W., & Russakoff, S. (1977). Relation¬
(1988). Use of the MMPI to identify malin¬ ship of MMPI scores of drug-abusers to per¬
gering and exaggeration of psychiatric symp¬ sonal variables and type of treatment
tomatology in male prison inmates. Journal program. Journal of Psychology, 96, 25-29.
of Consulting and Clinical Psychology, 56, Welsh, G. S. (1956). Factor dimensions A and R.
111-117. In G. S. Welsh & W. G. Dahlstrom (Eds.),
Ward, L. C. (1986). MMPI item subtlety research: Basic readings on the MMPI in psychology
Current issues and directions. Journal of and medicine (pp. 264-281). Minneapolis:
Personality Assessment, 50, 73-79. University of Minnesota Press.
Ward, L. C., & Dillon, E. A. (1990). Psychiatric Welsh, G. S. (1965). MMPI profiles and factor
symptoms of the Minnesota Multiphasic Per¬ scales A and R. Journal of Clinical Psychol¬
sonality Inventory (MMPI) Masculinity- ogy, 27,43-47.
Femininity scale. Psychological Assessment: Werner, P. D., Becker, J. M. T., & Yesavage, J.
A Journal of Consulting and Clinical Psy¬ A. (1983). Concurrent validity of the Over¬
chology, 2, 286-288. controlled Hostility scale for psychotics. Psy¬
Ward, L. C., & Ward, J. W. (1980). MMPI read¬ chological Reports, 52, 93-94.
ability reconsidered. Journal of Personality Wiener, D. N. (1948). Subtle and obvious keys for
Assessment, 44, 387-389. the MMPI. Journal of Consulting Psychol¬
Warman, R. E. & Elannum, T. E. (1965). MMPI ogy, 12, 164-170.
pattern changes in female prisoners. Journal Wiggins, J. S. (1959). Interrelationships among
of Research in Crime and Delinquency, 2, MMPI measures of dissimulation under stan¬
72-76. dard and social desirability instructions.
Wasyliw, O. E., Grossman, L. S., Haywood, T. Journal of Consulting Psychology, 23, 419—
W., & Cavanaugh, Jr., J. L. (1988). The de¬ 427.
tection of malingering in criminal forensic Wiggins, J. S. (1966). Substantive dimensions of
groups: MMPI validity scales. Journal of self-report in the MMPI item pool. Psycho¬
Personality Assessment, 52, 321-333. logical Monographs, 80(22, Whole No. 630).
Watson, C. G., Klett, W. G., Walters, C., & Vas- Wiggins, J. S. (1973). Personality and prediction:
sar, P. (1984). Suicide and the MMPI: A Principles of personality assessment. Read¬
cross-validation of predictors. Journal of ing, MA: Addison-Wesley.
Clinical Psychology, 40, 115-119. Wiggins, J. S., Goldberg, L. R., & Appelbaum,
Wauck, L. A. (1950). Schizophrenia and the M. (1971). MMPI content scales: Interpreta¬
MMPI. Journal of Clinical Psychology, 6, tive norms and correlations with other scales.
279-282. Journal of Consulting and Clinical Psychol¬
Webb, J. T. (1971). Regional and sex differences ogy,37, 403-410.
in MMPI scale high-point frequencies of psy¬ Wiggins, N., & Hoffman, P. J. (1968). Three
chiatric patients. Journal of Clinical Psy¬ models of clinical judgment. Journal of Ab¬
chology, 27, 483-486. normal Psychology, 73, 70-77.
Wechsler, D. (1980). Wechsler Adult Intelligence Wilcox, P., & Dawson, J. G. (1977). Role-played
Scale-Revised manual. New York: Psycho¬ and hypnotically induced simulation of psy¬
logical Corporation. chopathology on the MMPI. Journal of
Weed, N. C., Ben-Porath, Y. S, & Butcher, J. N. Clinical Psychology, 33, 743-745.
(1990). Failure of the Wiener and Harmon Williams, C. L. (1983). Further investigation of
Minnesota Multiphasic Personality Inven¬ the Si scale of the MMPI: Reliabilities, corre¬
tory (MMPI) subtle scales as personality de¬ lates, and subscale utility. Journal of Clinical
scriptors and as validity indicators. Psy¬ Psychology, 39, 951-957.
chological Assessment: A Journal of Con¬ Williams, C. L. (1986). MMPI profiles from ado¬
sulting and Clinical Psychology, 2, 281-285. lescents: Interpretive strategies and treatment
Weinstein, J., Averill, J. R., Opton, Jr., E. M., & considerations. Journal of Child and Adoles¬
Lazarus, R. S. (1968). Defensive style and cent Psychotherapy, 3, 179-193.
discrepancy between self-report and physio¬ Williams, C. L., & Butcher, J. N. (1989a). An
logical indexes of stress. Journal of Personal¬ MMPI study of adolescents: I. Empirical va¬
ity and Social Psychology, 10, 406-413. lidity of the standard scales. Psychological
Weisenberg, M. (1977). Pain and pain control. Assessment: A Journal of Consulting and
Psychological Bulletin, 84, 1008-1044. Clinical Psychology, 7, 251-259.
 References 499

Williams, C. L., & Butcher, J. N. (1989b). An rection in the detection of psychopathology

MMPI study of adolescents: II. Verification and its impact on profile height and configu¬
and limitations of code type classifications. ration among young adult men. Journal of
Psychological Assessment: A Journal of Consulting and Clinical Psychology, 52,
Consulting and Clinical Psychology, 1, 260- 468-473.
265. Worthington, D. L., & Schlottmann, R. S. (1986).
Wilson, R. L. (1980). A comparison of the predic¬ The predictive validity of subtle and obvious
tive validities of subtle versus obvious MMPI empirically derived psychological test items
items: Predicting the elusive neurosis. Un¬ under faking conditions. Journal of Person¬
published Master’s thesis, Auburn Univer¬ ality Assessment, 50, 171-181.
sity. Wrobel, T. A., & Lachar, D. (1982). Validity of
Wimbish, L. G. (1984). The importance of appro¬ the Wiener subtle and obvious scales for the
priate norms for the computerized interpreta¬ MMPI: Another example of the importance
tions of adolescent MMPI profiles. of inventory-item content. Journal of Con¬
Unpublished doctoral dissertation, Ohio sulting and Clinical Psychology, 50, 469-
State University, Columbus. 470.
Winter, W. D., & Stortroen, M. (1963). A com¬ Yonge, G. D. (1966). Certain consequences of ap¬
parison of several MMPI indices to differen¬ plying the K factor to MMPI scores. Educa¬
tiate psychotics from normais. Journal of tional and Psychological Measurement, 26,
Clinical Psychology, 19, 220-223. 887-893.
Winters, K. D., Newmark, C. S., Lumry, A. E., Zager, L. D., & Megargee, E. I. (1981). Seven
Leach, K., & Weintraub, S. (1985). MMPI MMPI alcohol and drug abuse scales: An
codetypes characteristic of DSM-III schizo¬ empirical investigation of their interrelation¬
phrenics, depressives, and bipolars. Journal ships, convergent and discriminant validity,
of Clinical Psychology, 41, 382-386. and degree of racial bias. Journal of Person¬
Witt, P. H., & Gynther, M. D. (1975). Another ex¬ ality and Social Psychology, 40, 532-544.
planation for black-white MMPI differences. Zelin, M. L. (1971). Validity of the MMPI scales
Journal of Clinical Psychology, 31, 69-70 for measuring twenty psychiatric dimen¬
Wong, M. R. (1984). MMPI Scale Five: Its mean¬ sions. Journal of Consulting and Clinical
ing, or lack thereof. Journal of Personality Psychology, 37, 286-290.
Assessment, 48, 279-284. Zuckerman, M., Levitt, E. E., & Lubin, B. (1961).
Woodworth, R. S. (1920). Personal data sheet. Concurrent and construct validity of direct
Chicago: Stoelting. and indirect measures of dependency. Jour¬
Wooten, A. J. (1984). Effectiveness of the K cor¬ nal of Consulting Psychology, 25, 316-323.
.
Author Index

Aaronson, B. S., 45, 47, 157, 158 Becker, J. M. T., 208

Abramoff, E., 160 Bell, W. E., 212, 252
Adams, N., 342 Ben-Porath, Y. S., 4, 21, 132, 137, 177, 195
Afifi, A. A., 276 Benarick, S. J., 169
Akeson, W. H., 214 Berg, I. A., 182
American Psychiatric Association, 136, 215 Bernreuter, R. G., 2
Alexander, R. S., 4, 196, 199, 201 Bernstein, I. H., 141
Allain, A. N., 274, 343, 348 Bertelson, A. D., 343
Altman, H., 47, 110, 232, 290, 346 Betman, J. A., 20
Anastasi, A., 49 Bieliauskas, L. A., 143
Anderson, W. P., 147, 150, 163, 170, 185, 209, Bieliauskas, V. J., 157
210, 213, 290, 291, 292 Birtchnell, J., 213, 217
Anderten, P., 46, 143 Blanchard, J. S., 32
Anthony, N., 79, 87, 90, 123 Blaney, P. H., 228
Apfeldorf, M., 188 Blashfield, R. K., 215
Appelbaum, M., 196 Blau, B. I., 342
Archer, R. P., 12, 13, 14, 17, 111, 113, 128, 290, Block, J., 182, 183
334, 335, 336, 342, 343, 348 Blumberg, S., 110
Armentrout, J. A., 276 Blyth, L. S., 171
Arredondo, R., 194, 195 Boerger, A. R., 136, 158, 216
Arthur, G., 352, 353 Bohm, K., 46, 143
Astin, A. W., 152 Bond, J. A., 66
Atkins, H. G., 48, 342, 345, 348, 350, 351 Boutilier, L. R., 351
Auvenshine, C. D., 192 Bowman, E., 154
Averill, J. R., 146 Bozlee, S., 350, 351
Bradley, L. A., 215
Ball, B., 340, 348 Brantner, J. P., 138, 163, 290
Ball, H., 61 Braswell, L., 343, 352
Ball, J. C., 52, 54, 332, 340, 341, 344 Brennan, J., 353
Barley, W. D., 99, 122, 236, 259 Briggs, P. F., 30, 38, 335
Barron, F., 20, 29, 186, 187 Brilliant, P. J., 122
Barry, J., 146 Broughton, R., 253
Barry, S. M., 233 Brower, D., 137
Baucom, D. H., 119, 157, 217, 218 Brown, M. N., 52
Baughman, E. E., 333, 344, 349 Brown, R. C., 258
Baxter, J. C., 120 Brozek, J., 336, 337
Bechtel, D., 217 Bruno, L. N., 341
Beck, S. J., 343, 345 Bryan, J. H., 250

501
502 Author Index

Bryan, L. L., 342 85, 90, 96, 138, 143, 147, 154, 163, 166, 171,
Buck, J. A., 271 182, 190, 191, 194, 204, 237, 238, 242, 243,
Buechley, R., 61 253, 335, 336
Buffaloe, J. D., 34 Comrey, A. L., 137, 140, 145, 152, 160, 165,
Burchstead, G. N., 212 168, 169, 173, 176, 226, 227
Burish, T. G., 107 Constantinople, A., 157
Burke, J. D., 192 Cook, P. E., 29, 182, 208, 348, 351
Burkhart, B. R., 131, 132, 136 Cooke, G., 341
Buros, O. K., 1, 294 Costello, R. M., 339, 342, 343, 344, 345, 347
Burton, A., 158 Counts, S., 344, 348
Butcher, J. N., 1, 4, 12, 20, 21, 24, 25, 26, 27, Cowan, M. A., 48, 346
28, 29, 31, 32, 33, 34, 40, 47, 48, 49, 106, Coyle, F. A., 42, 107, 196, 233
108, 111, 113, 116, 132, 137, 138, 140, 143, Craig, R. J., 122
147, 154, 159, 160, 163, 166, 171, 172, 173, Cuellar, I., 339
175, 177, 178, 182, 195, 199, 204, 220, 221,
223, 224, 225, 226, 227, 228, 229, 246, 331, Dahlstrom, L., 1, 12, 31, 49, 54, 136, 138, 181,
334, 335, 336, 339, 341, 343, 348 185, 290, 339
Byrne, D., 146 Dahlstrom, W. G., 1, 6, 12, 30, 31, 32, 42, 49,
54, 61, 76, 106, 107, 108, 110, 111, 114, 115,
Calden, G., 343, 348 116, 128, 131, 133, 136, 138, 139, 140, 143,
Caldwell, A. B., 48, 116, 182, 220, 221, 223 146, 147, 154, 155, 157, 159, 161, 163, 164,
Caldwell, M. G., 341 166, 171, 175, 178, 181, 183, 185, 187, 204,
Callahan, W. J., 349, 350, 351 214, 217, 218, 219, 221, 229, 290, 293, 294,
Calsyn, D. A., 214 331, 333, 339, 344, 349, 352
Caraveo-Ramos, E., 342, 348 Damarin, F. L., 196
Carp, A., 115 Dana, R. H., 339
Carrillo, R., 349 Danahy, S., 192
Carroll, D., 52, 54 Daniels, C. W., 344, 347, 348
Carroll, J. L., 341, 348 Davies, J., 82, 83
Carson, R. C., 135, 139, 142, 161, 163, 174, 217 Davis, H. G., 104, 188, 194, 195
Cass, W. A., 115 Davis, K. R., 208, 271
Cassel, C. A., 153 Davis, L. J., 190, 191
Cavanaugh, J. L.s 64, 132 Davis, W. E., 48, 171, 343, 345, 346
Chaleff, A., 276 Dawson, J. G., 123
Chambers, E. D., 4, 196 Dean, R. B., 158
Chance, J., 49 Deiker, T. E., 207, 208
Charles, H. L., 342 Denny, N., 212
Chestnut, E., 223 Devries, A. G., 217
Christensen, P., 45 Diers, C. J., 104, 182
Christian, W. L., 131, 132, 133, 136, 177 Dillon, E. A., 158
Clark, C. G., 346 Dinning, W. D., 65
Clark, L. A., 339 Dolan, M. P., 48, 342, 345, 348, 350, 351
Clavelle, P. R., 225 Dolmetsch, R., 64
Clopton, J. R., 50, 52, 171, 181, 182, 188, 208, Drake, L. E., 9, 175, 176, 290
217, 218, 232, 261, 265 Dubinsky, S., 128
Coche, E., 64 Dubro, A. F., 215
Cofer, C. N., 49, 78, 87, 95, 96, 97, 98, 104 Duckworth, J. C., 92, 99, 122, 147, 150, 185,
Cole, K. M., 214 209, 210, 213, 236, 259, 290, 291, 292
Colligan, R. C., 1, 12, 17, 18, 19, 20, 23, 26, 29, Duker, J., 196, 232, 233, 252, 261, 262, 263, 268,
30, 38, 39, 40, 43, 44, 46, 52, 53, 63, 67, 69, 270, 271, 272, 274, 277, 280, 282, 284, 290
 Author Index 503

Dura, J. R., 65 Fry, F. D., 342

Dureau, J. L., 214 Fuller, G. B., 341, 348
Duvall, S. W., 65
Dye, C. J., 46, 143 Gallucci, N. T., 78, 87, 192
Gamble, D. J., 128
Eaddy, M. L., 54 Garbin, C. P., 141
Edinger, J. D., 229 Garvin, R. D., 47, 188, 189, 190, 252, 261, 265,
Edwards, A. L., 4, 104, 182, 183 276
Ehrenworth, N. V., 334, 335 Gasparikova-Krasnec, M., 118
Eichman, W. J., 137 Gauron, E., 110
Elion, V. H., 342, 346 Gayton, W. F., 212
Endicott, N. A., 160, 161, 163 Gearing, M. F., 153, 208
Englehart, R., 110 Geertsma, R. H., 188
Erdberg, P., 341, 346 Genthner, R. W., 343, 346
Erdberg, S. P., 340, 341 Gentry, F., 122, 170
Erickson, J. R., 196 Gentry, W. D., 215
Erickson, W. D., 274, 276 Gerritse, R., 233
Eron, L. D., 276 Gerton, M. I., 63
Evans, C., 9, 175 Geyer, S., 272
Evans, R. G., 65, 213, 223, 227, 229 Giannetti, R. A., 219, 334
Exner, J. E., 87, 95 Gibeau, P. J., 233
Gier, R. H., 339
Fairbank, J. A., 212 Gilbart, T. E., 154
Farberow, N. L., 217 Gilberstadt, H., 196, 217, 232, 233, 252, 261,
Farr, S. P., 174, 252 262, 263, 268, 270, 271, 272, 274, 277, 280,
Faschingbauer, T., 12, 226, 227, 229 282, 284, 290
Faust, D., 217 Gillum, B., 143, 337
Fekken, G. C., 65, 66 Gillum, R., 143, 337
Ferguson, G. A., 6 Gimenez, F., 349
Fillenbaum, G. G., 338 Glaser, R., 342
Finch, A. J., 228 Goldberg, F. R., 196, 219
Fine, H. J., 342 Golden, C. J., 262
Finkelstein, A., 212, 252 Gomez-Mont, F., 226, 227, 228, 229
Finney, J. C., 192 Gonen, J. Y., 157
Fisher, J., 145, 146 Gonzalez, J., 350
Fjordbak, T., 107 Good, P. K., 138, 163, 290
Flanagan, J., 342 Goodstein, F. D., 154, 274
Flint, G., 352 Gordon, N. G., 141
Forgac, G. E., 153 Gordon, R. A., 128, 224
Forsyth, D. R., 154 Gottesman, I. I., 188, 192, 335
Fowler, D. R., 212, 252 Gottfredson, D. K., 45
Fowler, R. D., 42, 166, 196, 233, 265, 276, 341, Gough, H. G., 11, 29, 49, 78, 86, 87, 88, 89, 91,
346 92, 98, 124, 157, 209, 210, 348, 349
Freeman, C. W., 214 Gouze, M., 143, 337
Friberg, R. R., 158 Graham, J. R., 1, 4, 31, 47, 49, 132, 136, 137,
Fricke, B. G., 148 138, 142, 154, 156, 157, 159, 176, 177, 188,
Friedman, A. F., 33, 290, 292 190, 192, 193, 195, 204, 208, 216, 217, 221,
Friedrich, W. N., 252 227, 228, 246, 249, 250, 265, 271, 290, 294,
Friesen, J., 65 331, 343, 344, 346
Fromuth, M. E., 132 Gravitz, M. A., 54, 55, 63, 106, 109, 222
504 Author Index

Gray, W. S., 32 156, 157, 159, 164, 167, 172, 195, 239, 252,
Grayson, H. M., 84, 95, 220, 221, 222 259, 331, 332, 333, 335
Green, S. B., 339 Hauer, A. L., 276
Greene, M. I., 107 Haven, G. A., 214
Greene, R. L., 12, 42, 47, 49, 52, 53, 55, 61, 62, Haywood, T. W., 132
63, 64, 65, 66, 67, 68, 69, 70, 79, 81, 82, 83, Heap, R. F., 107
85, 88, 90, 92, 96, 104, 119, 122, 132, 157, Hedlund, J. L., 45, 54, 56, 64, 68, 70, 82, 83,
171, 174, 188, 189, 190, 193, 194, 195, 222, 86, 88, 91, 92, 95, 97, 98, 100, 101, 136, 143,
226, 234, 252, 258, 261, 265, 276, 339, 341, 147, 154, 163, 166, 171, 193, 227, 229, 235,
343, 348, 349 236, 241, 242, 246, 248, 249, 253, 259
Griffith, A. V., 166 Heilbrun, A. B., 48, 114, 153
Grisell, J. L., 335 Heiman, E. M., 350
Gross, L. R., 118, 119 Helmes, E., 229
Grossman, L. S., 132 Helzer, J. E., 192
Gruba, G. H., 119 Henrichs, T. F., 219
Gruenberg, E., 192 Herreid, C. F., 352
Grumpelt, H. R., 158 Herreid, J. R., 352
Gulas, L., 147 Hess, A. K., 132
Guthrie, G. M., 169 Hess, J. L., 342, 348
Gynther, M. D., 43, 47, 110, 122, 128, 131, 132, Hibbs, B. J., 350
136, 232, 233, 261, 262, 264, 265, 268, 270, Hilf, F. D., 162
271, 276, 277, 280, 282, 284, 290, 292, 339, Hill, G., 342, 348
340, 341, 344, 345, 346, 347, 349 Hill, H. E., 65, 342
Hiner, D. L., 120
Haertzen, C. A., 65, 342
Hodo, G. L., 265, 276
Hale, G., 92
Hoffman, N., 228
Hall, G. C. N., 276, 282
Hoffman, P. J., 219
Haller, D. L., 92, 122, 232, 290, 333
Hoffmann, H., 192, 276
Hampton, P. J., 188
Hoffmann, N. G., 220, 227, 228, 229
Handal, P. J., 341
Hokanson, J. E., 343, 348
Hannum, T. E., 250
Holcomb, W. R., 342
Hanson, D. R., 335
Holden, R. R., 65, 66, 132
Hanvik, L. J., 213, 214
Holland, T. R., 153
Harding, C. F., 169
Hare, R. D., 153 Holmes, G. R., 223
Harkey, B., 46 Holmes, W. O., 188
Harrell, T. H., 34, 249 Holz, W. C., 169
Harris, L. C., 339 Honaker, L. M., 34, 246, 247
Harris, M., 64 Hoppe, C. M., 208
Harris, R. E., 140, 142, 145, 152, 157, 160, 164, Hopson, L., 215
165, 168, 169, 173, 176, 182 Horowitz, L. M., 253
Harris, R. J., 162 Hostetler, K., 177, 226
Harrison, R. H., 340, 341, 344, 347, 349 Houston, B. K., 107
Harrison, W. R., 46 Hovanitz, C., 132
Hartshorne, H., 11 Hovey, H. B,, 136
Harvey, M. A., 79 Howe, M. G., 229
Haskin, J. J., 190, 348 Hoyt, D. P., 188
Hathaway, S. R., 5, 8, 10, 17, 30, 37, 38, 45, 47, Hryckowian, M. J., 128
57, 65, 80, 106, 108, 109, 110, 111, 112, 113, Hsu, L. M., 20, 39
114, 132, 136, 138, 139, 144, 145, 151, 154, Huber, N. A., 192
 Author Index 505

Huesmann, L. R., 276 Kirchner, F. H., 128

Hunt, H. F., 87, 95, 115 Kirk, A. R., 341, 344
Hutchins, T. C., 45 Kirk, L., 87
Hyer, L., 46, 79 Kleban, M. H., 95
Klein, G. L., 182
Ingram, J. C., 342, 348 Kleinmuntz, B., 29, 211
Klett, W. G., 217
Jackson, D. N., 4, 132, 183, 195 Kline, J. A., 352
Jarnecke, R. W., 4, 196 Kline, R. B., 348
Jasso, R., 339 Klinefelter, D., 336
Jeffrey, T. B., 190, 341, 343, 348 Kling, J. S., 208
Jenkins, G., 114 Klinge, V., 335, 344
Jewell, J., 215 Klingler, D. E., 95, 219
Johnson, J. H., 95, 219 Knapp, R. R., 209, 210
Johnston, N. G., 341 Knapp, S. C., 344, 347, 348
Jones, E. E., 347, 349 Knight, G. E., 350
Jones, F. W., 64 Kobos, J. C., 350
Jones, M. H., 343 Koppell, B., 162
Jones, T., 122, 170 Koss, M. P., 220, 221, 222, 223, 224
Jones, W. C., 217 Kroeker, T. A., 335
Jortner, S., 160 Kroger, R. O., 123
Judson, A. J., 49 Krupp, N. E., 261
Kruzich, D. J., 190, 348
Kaemmer, B., 1, 31, 49, 138, 204, 221, 246, 331 Kulka, R. A., 212
Kahn, M. W., 215, 350 Kunce, J. T., 170
Kameoka, V., 353
Kamerown, D. B., 190 Lachar, D., 4, 78, 84, 85, 91, 92, 94, 98, 100,
Kammeier, M. L., 192, 276 101, 110, 196, 199, 201, 221, 222, 223, 225,
Kantor, R. E., 115 226, 233, 234, 258, 262, 268, 290, 335, 336,
Karabelas, M. J., 342 344
Karol, R. L., 59 Lacks, P. B., 171
Kass, E. H., 340, 341, 344, 347, 349 Lair, C. V., 148
Katz, S. E., 2, 4 Landis, C., 2, 4
Kawakami, D., 169 Lane, P. J., 208
Kay, D. C., 192 Lang, A. R., 96
Kazan, A. T., 110 Lansky, L. ML, 157
Keane, T. M., 212, 252 Lanyon, R. I., 121, 123, 290, 292, 293
Keefer, G., 171 Larde, J., 217
Kelley, C. K., 47, 154, 158, 234, 259, 264, 265, Larsen, R. M., 1
267, 268, 270, 272, 273, 274, 275, 277, 279, Lavin, P., 53, 83
280, 282, 283, 285, 290 Lawson, H. H., 350
Kendall, P. C., 228 Lawton, M. P., 95
Kennard, J., 213 Lazarus, R. S., 146
Khoury, L., 223 Leach, K., 122
Kilpatrick, D. G., 341 Lebeaux, M. O., 214
Kincannon, J. C., 226, 227, 228, 338 Lefkowitz, M. M., 276
King, G. D., 47, 154, 158, 234, 259, 264, 265, Leon, G. R., 143, 337
267, 268, 270, 272, 273, 274, 275, 277, 279, Leonard, C. V., 208, 217
280, 282, 283, 285, 290 Lester, D., 208
King, H. F., 341, 348 Levi, M., 153
506 Author Index

Levitt, E. E., 23, 181, 182, 213 McCreary, C., 276, 342, 348, 350, 351
Levy, L. H., 288 McDonald, R. L., 340, 342, 344
Lewak, R., 33, 290 McDowell, E., 87
Lewis, E. G., 136 McGill, J. C., 343, 350, 351
Lewis, G., 342 McGrath, R. E., 65, 147
Lichenstein, E., 250 McKegney, F. P., 110, 111
Lilly, R. S., 136, 156, 227 McKinley, J. C., 5, 8, 17, 38, 47, 57, 65, 106,
Lingoes, J. C., 140, 141, 142, 145, 152, 157, 160, 108, 109, 110, 111, 112, 113, 114, 136, 138,
164, 165, 168, 169, 173, 176, 182 139, 144, 145, 151, 164, 172, 195, 331
Linsz, N. L., 107 McLaughlin, J. D., 229
Liske, R., 344 McNeil, B., 342, 348
Listiak, R. L., 135 Meehl, P. E., 10, 17, 29, 37, 45, 80, 113, 114,
Little, K. B., 145, 146 209, 218, 219, 234, 239, 252, 259, 289, 293,
Loper, R. G., 192, 276 348
Louks, J., 214 Megargee, E. I., 29, 96, 182, 188, 207, 208, 209,
Lowenstein, E., 253 342, 346, 348, 351
Loya, F., 350 Mehlman, B., 89
Lubin, B., 1, 213 Meikle, S., 233
Lumry, A. E., 122 Mendelsohn, G. A., 29, 182, 207, 348, 351
Lutz, R. W., 121 Mezzich, J. E., 196
Luxenberg, M. G., 274 Michaelis, J. U., 114
Lyle, W. H., 152 Michaels, E. J., 153
Miles, C. C., 155
MacAndrew, C., 29, 188, 190, 191, 193, 258, Miller, C., 344, 347, 348
347, 348, 351, 353 Miller, H. L., 346
Macdonald, D. I., 190 Miller, H. R., 20, 140, 146, 228, 229
Maguire, A., 208 Miller, J. S., 152
Mallory, C. H., 208 Monachesi, E. D., 209, 210, 331, 332, 333, 335
Malloy, P. F., 212 Montgomery, G. T., 339, 350
Maloney, M. P., 65 Monroe, J. J., 152
Manosevitz, M., 158 Moore, C. D., 341
Marchioni, P., 342, 348 Moreland, K. L., 228
Marggraff, W. M., 168, 169 Morey, L. C., 215, 252
Margolis, R., 215 Morse, R. M., 190, 191
Marks, P. A., 92, 121, 122, 208, 232, 233, 252, Muller, B. P., 341
261, 262, 268, 270, 271, 273, 275, 276, 277, Munley, M. J., 154
280, 282, 284, 290, 333, 334, 336, 343, 346 Murphree, H. B., 342
Marsella, A. J., 353 Murphy, R. W., 214
Martin, P. W., 157, 174, 252 Murray, J. B., 154
Matarazzo, J. D., 1, 107 Myers, B., 217
Matthews, G. R., 212
Mauiro, R. D., 276 Nakamura, C. Y., 114
May, G. D., 343 Navran, L., 213, 348
May, M. A., 11 Nelson, D., 146
Mayo, M. A., 190 Nelson, L. D., 140
McAnulty, D. P., 95 Neuringer, C., 50, 52, 64, 181
McAnulty, R. D., 95 Newmark, C. S., 45, 122, 170, 226, 228
McClosky, H., 29, 209, 348 Newton, J. R., 222
McCormick, R., 344 Nicholas, D., 92
McConnell, T. R., 9, 175 Nichols, D. S., 64, 67, 68, 70, 82, 199
 Author Index 507

O’Connor, J. P., 137 Pincus, H. A., 190

Oetting, E. R., 290 Pirrello, P. E., 344
Offord, K. P., 1, 17, 20, 38, 39, 52, 138, 182, Plemons, G., 350
190, 191, 237, 238, 242, 243, 253, 335, 336, Podany, E. C., 348
337 Pogany, E., 341
Ogren, D. J., 120 Pollack, D., 352
Olinger, L. B., 95 Ponder, H. M., 342
Olmedo, E. L., 350 Posey, C. D., 132
Olmstead, D. W., 209, 210 Post, R. D., 118, 171, 174, 217
O’Malley, W. B., 65, 147 Powell, B. J., 227
Onorato, V. A., 353 Poythress, N. G., 228
Opton, E. M., 146 Prescott, C. A., 188, 192
Orozco, S., 339, 350 Prieto, E. J., 215
Orvaschel, H., 192 Pritchard, D. A., 48, 339, 342, 345, 346
Osborne, D., 1, 17, 20, 38, 39, 40, 43, 52, 138, Proctor, W. C., 276
182, 185, 261 Prokop, C. K., 146, 215, 252, 262, 271
Osmon, D. C., 262 Pruitt, W. A., 213
Otto, R. K., 96
Overall, J. E., 226, 227, 228, 229 Quay, EE, 114, 169
Quinsey, V. L., 208
Pabst, J., 87
Padilla, E. R., 342, 348, 350, 351 Rader, C. M., 276
Page, J., 4 Rand, M. E., 89
Page, R. D., 350, 351 Rand, S. W., 228
Pallis, D. J., 217 Raney, D., 343, 352
Pancheri, P., 339 Rappaport, N. B., 95
Pancoast, D. L., 12, 13, 14, 17, 336 Ray, E., 341, 348
Panton, J. EE, 141, 342 Regier, D. A., 192
Parad, H. W., 253 Reilley, R. R., 226, 350
Patalano, F., 342, 348 Rich, C. C., 188
Patterson, E. T., 342 Richardson, E. H., 351
Patterson, T. W., 64 Richardson, EE, 158
Pauker, J. D., 233, 250 Ries, H. A., 115
Paulhus, D. L., 104 Roberts, A. C., 352
Paulson, M. J., 276 Roberts, W. R., 48, 342, 345, 348, 350, 351
Payne, F. D., 196 Robinowitz, R., 48, 212, 252, 342, 345, 348, 350,
Pearson, J. S., 43, 138, 185, 337 351
Penk, W. E., 48, 212, 252, 342, 345, 348, 350, Robins, L. N., 192
351 Robinson, H. M., 32
Pepper, L. J., 156, 157 Rogers, M. L., 128
Perez, C., 214 Rogers, R., 64, 78
Perley, R. N., 162 Rohan, W. P., 192
Perse, J., 214 Rosen, A., 52, 108, 217
Persons, R. W., 208, 271 Rosen, A. C., 158
Peteroy, E. T., 344 Rosenbaum, G., 276
Peterson, C. D., 157, 211 Rosenberg, D., 171
Peterson, D. R., 218 Rosenblatt, A., 48, 339, 342, 345, 346
Petzel, T. P., 110 Rosenblatt, A. E, 96
Pfeiffer, E., 338 Rothenberg, P. J., 171
Pichot, P., 214 Rotman, S. R., 192
508 Author Index

Rowell, J. T., 169 Sipprelle, C. N., 79

Rozynko, V. V., 352 Sisson, B. D., 218, 219
Rubin, H., 169 Skeeters, D. E., 192
Ruch, F. L., 115 Sletten, I. W., 47, 232, 290
Ruch, W. W., 115 Smith, C. P., 344, 346
Russakoff, S., 343 Smith, D. F., 192
Ryan, T. A., 343, 345 Smith, E. E., 114
Rychewaert, A., 214 Smith, G. T., 276
Smith, M. R., 215, 252
Sabalis, R. F., 223 Smith, R. L., 250
Sabo, T. W., 122 Snyder, D. K., 215, 348
Sanborn, K. O., 353 Snyder, W. U., 169
Saunders, T. R., 222 Snyter, C. M., 132
Scagnelli, J., 162 Spirito, A., 217
Schenkenberg, T., 45, 46 Stackman, W., 87
Scherer, I. W., 52, 54 Stanton, J. M., 342
Schlenger, W. E., 212 Steer, R. A., 64
Schlottmann, R. S., 132 Stefic, E. C., 137
Schmidt, H. O., 115 Stein, K. B., 137, 140, 160, 165, 169, 173, 177,
Schmolck, P., 64, 67, 68, 70 182, 215, 216
Schneck, J. M., 110 Stein, M., 171
Schretlen, D., 81 Sternbach, R. A., 214
Schroeder, H. E., 156, 227 Stevens, M. R., 226
Schwartz, G. F., 250 Stone, L. A., 135
Schwartz, M. F., 188, 192, 193 Stortroen, M., 219
Schwartz, M. S., 261, 262 Strauss, M. E,, 344, 346, 347
Sedlacek, G. M., 188 Streiner, D. L., 20, 140, 146, 228, 229
Seeley, R. K., 274 Strenger, V. E., 265
Seeman, W., 92, 121, 122, 128, 131, 132, 133, Strong, P. M., 156, 157
232, 233, 290, 333 Sue, D. W., 353
Seever, M., 1 Sue, S., 353
Serkownek, K., 156, 176, 177 Super, D. E., 2
Severson, R., 110 Sutker, P. B., 274, 341, 342, 343, 348
Sheinberg, I. M., 110 Swart, E. C., 141
Shekelle, R. B., 143 Sweet, J. J., 262
Sheppard, D., 276 Sweetland, A., 114
Shimkunas, A. M., 110 Swenson, W. M., 1, 17, 20, 39, 43, 44, 45, 46,
Shizuru, L., 353 52, 138, 143, 147, 154, 163, 166, 171, 182,
Shore, J. H., 352 185, 209, 213, 337, 338
Shore, R. E., 346, 347
Shultz, T. D., 233 Tamkin, A. S., 52, 54
Shweder, R. A., 183 Tanner, B. A., 262, 265, 272, 274, 277, 279
Silver, R. J., 115 Tarter, R. E., 162
Simon, W., 217 Tatro, R. L., 192
Simpson, M., 122, 170 Taulbee, E. S., 218, 219
Sines, L. K., 115, 119 Tellegen, A. M., 1, 31, 33, 47, 49, 138, 182, 221,
Sines, J. O., 157, 208, 233, 271 246, 331
Singer, M. I., 158 Terman, L. M., 155
Singer, R. D., 208 Thatcher, A. A., 64
Singles, J., 334 Thomason, M. L., 276
 Author Index 509

Thornby, J. I., 192 Weiss, R. W., 343

Tiffany, D. W., 339 Welch, S. M., 104
Tisdale, M. J., 159 Welsh, G. S., 1, 12, 20, 29, 31, 49, 54, 136, 138,
Towne, W. S., 214 181, 182, 183, 185, 290, 339, 348
Trapp, E. P., 148 Werner, P. D., 208
Truscott, D., 208 Wertz, C., 344, 348
Tsushima, W. T., 214, 353 Wetzler, S., 215
Turnbull, W., 123 White, T. W., 79
Tyler, F. T., 114, 182 Wiener, D. N., 26, 49, 78, 79, 80, 91, 92, 93, 96,
97, 98, 100, 101, 102, 103, 104, 128, 131,
Uecker, A. E., 351 132, 140, 145, 152, 157, 160, 163, 166, 169,
Unger, B. L., 171 173, 177, 182, 258, 268
Upshaw, H. S., 166 Wiggins, J. S., 4, 20, 30, 104, 137, 140, 146, 152,
160, 165, 169, 172, 173, 177, 182, 191, 195,
Van de Castle, R., 213 196, 197, 198, 199, 200, 211, 215, 219, 268,
Varney, G. W., 208 348, 351
Vassar, P., 217 Wilcox, P., 123
Velasquez, R. J., 349, 350, 351 Williams, C. L., 4, 137, 176, 195, 334, 335, 336
Vesprani, G. J., 128, 131, 132 Williams, T. A., 95, 219
Vestre, N. D., 161, 163 Wilson, R. L., 132
Vincent, K. R., 228 Wimbish, L. G., 335, 336
Vincent, N. M. P., 107 Winder, C. L., 115
Vitaliano, P. P., 276 Winter, W. D., 219
Winters, K. D., 122
Witt, P. H., 347
Walbek, N. H., 274
Wittner, W., 162
Wales, B., 128, 133
Wolf, S. R., 214
Walker, C. E., 33, 208
Won Cho, D., 45, 46, 54, 56, 64, 68, 70, 82, 83,
Wallhermfechtel, J., 346
86, 88, 91, 92, 95, 97, 98, 100, 101, 143, 147,
Walters, C., 217
154, 163, 166, 171, 193, 227, 235, 236, 241,
Walters, G. D., 79, 153, 170, 171, 174, 190, 252,
242, 246, 248, 249, 253, 259
341, 342, 343, 348
Wong, M. R., 158
Warbin, R., 110, 346
Woodward, W. A., 342, 345, 348
Ward, J. W., 33
Woodworth, R. S., 1
Ward, L. C., 32, 128, 158
Wooten, A. J., 114, 115
Warman, R. E., 250
Worthington, D. L., 132
Wasyliw, O., 132
Wright, J. C., 253
Watkins, B. A., 48, 346 Wrobel, T. A., 78, 84, 85, 91, 92, 94, 98, 100,
Watson, C. G., 161, 163, 217 101, 221, 222, 223, 225, 226, 258, 268
Wauck, L. A., 171, 215
Waugh, M. H., 215 Yesavage, J. A., 208
Webb, J. T., 33, 47, 290 Yonge, G. D., 114
Webb, W. W., 215
Wechsler, D., 32 Zager, L. D., 348
Weed, N. C., 132 Zelin, M. L., 136
Weinstein, J., 146 Ziff, D. R., 228
Weintraub, S., 122 Zimostrad, S., 92
Weisenberg, M., 214 Zucker, R. A., 341, 344
Weisman, M. M., 192 Zuckerman, M., 213
Subject Index

Accuracy of item endorsement: Cannot Say (?) scale, 21, 35, 50-59
by codetype, 102, 103, 104 Carelessness (CLS) scale, 49, 61, 64, 65-68, 69,
cutting scores to assess, 100, 101 70, 78
general guidelines for, 76-78 Clinical scales. See specific scales
overreporting of psychopathology, 77, 78-93, Codetypes:
101, 111, 132 concordance, 20, 29, 245-250, 251
summary, 91-93, 97-99 defined, 43
underreporting of psychopathology, 77, 93- frequency on MMPI, 235-241
97, 102, 132 frequency on MMPI-2, 235, 241-245
Administration: general issues, 193, 194, 231-232, 258-259
interpretation of, 136
computer, 34
and prototypic scores, 253-258
overview, 31-34
and psychiatric diagnosis, 252
taped, 32, 76
stability, 12, 15-16, 250-252
Admission (Ad) scale, 145, 146
College Maladjustment (Mt) scale, 29, 185, 186,
Adolescents, 29-30, 32, 43, 54, 61, 80, 85, 106,
187, 211, 212
110, 111, 113, 175, 189, 190, 191,
Computer-based interpretations, 294
331-336
Caldwell Report, 307-311, 326-328
Age. See Demographic variables
Minnesota Report, 304-305, 321-323
Aged, 143, 336-338 MMPI-2 Adult Interpretive System, 305-307,
American Indians. See Demographic variables 323-326
Anger (ANG), 29, 202, 203, 205, 206 Configural analysis, 292, 293
Antisocial Practices (ASP), 29, 202, 203, 205, Consistency of item endorsement:
206 by codetype, 102, 103
Anxiety {ANX), 29, 165, 203, 205, 206 cutting scores to assess, 100, 101
Asian-Americans. See Demographic variables general guidelines, 61, 76, 128
Augmented profile, 58-60 summary, 76
Automated interpretation. See Computer-based Content areas:
interpretation dropped on MMPI-2, 23
on MMPI, 9-10
Content scales. See also specific scales
Back F (FB) scale, 29, 69-74, 78, 91, 108, 111- MMPI-2, 4, 29, 195, 199, 202-207
113, 186 Wiggins, 4, 20, 30, 137, 140, 146, 152, 160,
Bernreuter Personality Inventory, 2-3 165, 169, 172, 173, 177, 182, 195-199,
Bizarre Mentation (BIZ), 29, 160, 169, 172, 202, 200, 201, 202, 205, 206, 207
203, 205, 206 Cookbook interpretive systems:
Black F scale, 110 general overview, 232-235
Blacks. See Demographic variables Gilberstadt & Duker, 232

511
512 Subject Index

Cookbook interpretive systems (cont.) F - K index, 20, 86-89, 91, 92, 95-96, 98
Gynther, 233, 234 Factor scales:
King & Kelley, 234 first factor, 182, 183, 185, 188, 211, 212
Lachar, 233, 234 second factor, 182, 183
Marks, Seeman, & Haller, 232-233 Faking bad. See Overreporting of psychopath¬
Critical items, 220-226 ology
Caldwell, 220, 221, 223 Faking good. See Underreporting of psycho¬
defined, 220 pathology
Grayson, 84, 220, 221, 222, 223 Family problems (EAM), 29, 169, 204, 205, 206
Koss & Butcher, 220, 221, 222, 223, 224 Fears (ERS), 29, 202, 203, 205, 206
Lachar & Wrobel, 78, 84-86, 91, 92, 94-95, First factor. See Factor scales
98, 101, 221, 222, 223, 224, 225, 226 Forms. See Test forms
Cynicism (Ck7V), 29, 202, 203, 205, 206
Gender. See Demographic variables
Defensiveness. See Underreporting of psycho¬ Gender Role—Feminine (GF) scale, 29, 157,
pathology and K scale 186, 211-212
Demographic variables: Gender Role—Masculine (GM) scale, 29, 157,
age, 32, 43-47, 137, 143, 147, 154, 163, 166, 186, 211-212
171, 175, 178, 185, 189, 190, 191, 213 Goldberg index. See Psychotic-Neurotic indexes
education, 23, 26, 31-32, 47-48, 76, 107, 116, Gough dissimulation index. See F - K index
121, 143, 158, 159
ethnicity, 26, 43, 47, 110, 189, 190, 191, 192, Harris & Lingoes subscales, 140-142, 145-146,
338-354 152, 157, 160, 162, 165, 168-169, 170,
gender, 26, 47, 80, 85, 106, 108, 137, 147, 154, 173, 174, 176, 182
163, 166, 171, 175, 185, 189, 190, 191, Health Concerns (HEA), 29, 137, 169, 202, 203,
223, 224 205, 206
intelligence, 32, 47-48, 76, 107, 116, 124, 137, High-point pairs. See Codetypes
153 Hispanics. See Demographic variables
marital status, 26, 171, 178
social class, 23, 26, 47-48, 107, 116, 121, 137 Impression management, 104, 106
Demographics of MMPI-2 sample, 26 Intelligence. See Demographic variables
Denial (Dn) scale, 107, 145, 146 Interpretation of the profile. See Profile inter¬
Dependency (Dy) scale, 162, 210, 213 pretation
Depression (DFP), 29, 140, 165, 169, 202, 203, Item(s):
205, 206 content categories, 5
Dissimulation (Ds; Ds-r) scales, 11, 49, 86, 89- dropped on the MMPI-2, 21
91, 92 frequently omitted, 57
Dominance (Do) scale, 186, 194, 209-210, 211 interpretation of content, 9-10, 84
objections, 33
Education. See Demographic variables religious, 21, 54, 55
Edwards’ Personal Preference reworded on the MMPI-2, 20-21
Schedule, 4 selection procedure, 6-7
Edwards’ Social Desirability (ESD) sexual, 54
scale, 104
Ego Strength (Es) scale, 20, 182, 186-188, 212 ALcorrection, 8, 11-12, 26, 35, 37, 40, 59, 60, 79,
Errors in scoring. See Scoring errors 113-115, 124, 145, 167, 169
K scale, 11, 12, 44, 46, 47, 48, 67, 78, 113-116,
F scale, 11, 44, 45, 49, 61, 62, 63, 64, 65, 67, 69- 117, 118, 121, 145, 147, 152, 169, 182,
74, 78, 87, 91, 92, 108-111, 112, 113, 187
121, 124, 152, 160, 166, 169, 170, 194 K+ profile. See Normal K+ profile
 Subject Index 513

Lie (I) scale, 11-12, 23, 35, 37, 44, 46, 49, 78, Keane {PK), 29, 185, 186, 188, 211, 212
96, 104, 106-108, 109, 121, 124, 131 Schlenger {PS), 29, 185, 186, 188, 211, 212
Low back {Lb) scale, 213-215 Profile:
Low-points: elevation, 290-291, 293
frequency on MMPI, 239-240 interpretation:
frequency on MMPI-2, 243-245 examples, 294-329
general issues, 115, 135, 291, 292 general process, 43, 287-294
Low Self-esteem (LSE) scale, 29, 165, 202, 204, phasicity, 294
205, 206 slope, 293, 294
Prototypic scores. See Codetypes and prototypic
MacAndrew Alcoholism {MAC; MAC-R) scale, scores
29, 47, 182, 186, 188-194, 195, 210, 211 Psychotic-Neurotic indexes, 218-220
Malingering. See Accuracy of item endorsement Psychotic tetrad, 124, 136, 161, 174, 293
Meehl-Dahlstrom rules. See Psychotic-Neurotic
indexes Random sorts. See Random response sets
Minority groups. See Demographic variables Re scale. See Social Responsibility scale
Mp scale. See Positive Malingering scale Reading level, 31-32, 76, 128
Mt scale. See College Maladjustment scale Reliability, 108, 111, 113, 116, 137, 143, 147,
154, 159, 163, 166, 171, 175, 178
Negative Treatment Indicators (77?7), 29, 202, Repression-Sensitization scale, 146
204, 205, 206 Response sets, 124-128
Neurotic triad: “all deviant,” 130
defined, 145, 293 “all false,” 64, 66, 126, 136, 140, 145, 184
configurations, 124, 148-151 “all nondeviant,” 130
Non-K-corrected scores, 47, 115 “all true,” 64, 66, 125, 161, 169, 171, 173, 184
Normal K+ profile, 92, 99, 116, 121-122 random, 64, 70, 71, 127, 129, 171
Norms: Role playing, 122-124
changes in MMPI, 12-14
contemporary normals, 17-20, 30 SAP. See Substance Abuse Proclivity scale
MMPI-2, 20-25, 30 Scale 1 {Hs: Hypochondriasis), 5-8, 12, 17, 23,
original MMPI, 6, 12, 17, 30, 38 35, 37, 38, 41, 44, 45, 46, 47, 107, 114,
131, 136-138, 139, 144, 145, 147, 151,
Obsessiveness {OBS), 29, 165, 202, 205, 206 187, 196, 202, 211, 217
Obvious and subtle subscales, 26-28, 49, 78-84, Scale 2 {D: Depression), 8, 29, 38, 41, 44, 45, 46,
91, 92, 93-94, 98, 101, 128, 130-133, 55, 60, 79, 131, 137, 138-143, 147, 165,
136, 140, 145, 157, 160, 166, 169, 173, 171, 172, 174, 175, 177, 179, 183, 185,
177 187, 196, 202, 210, 211, 212, 217
Overcontrolled-Hostility (O-H) scale, 29, 182, Scale 3 {Hy: Hysteria), 8, 12, 44, 79, 102, 107,
186, 207-209 114, 135, 136, 139, 142, 143-148, 151,
Overreporting of psychopathology. See Accu¬ 163, 183, 187, 202, 211, 212
racy of item endorsement Scale 4 {Pd: Psychopathic Deviate), 8, 12, 17,
35, 37, 38, 44, 45, 46, 48, 79, 106, 131,
Pepper & Strong subscales, 156, 157 135, 142 151-154, 160, 174, 179, 194, 202,
Personal Data Sheet, 1 210, 212
Personality Disorder scales, 215 Scale 5 {Mf: Masculinity-Femininity), 8-9, 17,
Plotting the profile, 35-37 26, 35, 37, 38, 40, 43, 44, 48, 55, 121,
135, 136, 151, 154-159, 160, 161, 211, 227
Positive Malingering {Mp) scale, 49, 78, 96-97, Scale 6 {Pa: Paranoia), 8, 29, 45, 47, 61, 79, 108,
98, 104 111, 135, 136, 142, 159-163, 169, 183,
Post Traumatic Stress Disorder scales: 202
514 Subject Index

Scale 7 (Pt: Psychasthenia), 8, 17, 35, 37, 38, 44, T scores

46, 59, 60, 61, 79, 114, 142, 164-166, linear, 20, 29, 35, 37-38, 39, 40, 41, 57, 136,
167, 169, 170, 177, 179, 182, 185, 187, 158, 179, 235
210, 211, 212, 217 normalized, 20, 38-39, 40, 41, 46
Scale 8 (Sc: Schizophrenia), 8, 17, 35, 37, 38, 44, uniform, 26, 29, 35, 38, 40-41, 135, 136, 158,
45, 61, 65, 79, 102, 106, 107, 108, 111, 179, 207, 235
114, 115, 136, 142, 160, 165, 167-171, Test forms, 34
172, 173, 174, 179, 182, 185, 187, 202, Test-Retest (TR) index, 10-11, 49, 61-65, 66, 67,
211, 212 69, 70, 78
Scale 9 (Ma: Hypomania), 8, 12, 17, 35, 37, 38, Test-taking attitudes, 10-12
44, 45, 46, 47, 79, 107, 131, 135, 142, Total T score difference:
153, 166, 169, 172-175, 176, 179, 183, by codetype, 105
194, 210 computation, 79, 80
Scale 0 (Si: Social Introversion), 9, 23, 29, 37, in normal persons, 81
40, 44, 47, 55, 60, 61, 113, 136, 156, 158, percentile equivalents, 83, 92
171, 172, 174, 175-179, 185, 202, 210, in psychiatric clients, 82, 93
211, 212, 227 True Response Inconsistency (TRIN)
Scale 0 subscales, 177, 178 scale, 29, 61, 74-76, 140, 184, 207
Scale 4-5-6 configuration, 163-164, 165 TSC Cluster scales, 137, 140, 160, 165, 169, 173,
177, 182, 215-217
Scale 6-7-8 configuration, 171-172, 173
Type A Behavior (TPA), 29, 204, 205, 206
Scarlet O’Hara V. See 4-5-6 configuration
Scoring:
Underreporting of psychopathology. See Accu¬
computer, 41-42 racy of item endorsement
errors in, 42-43
hand, 34-35 Validity:
Second factor. See Factor scales configurations, 116-121
Self-deception, 104, 106 scales. See individual scales
Serkownek subscales, 156, 157, 176, 177 use of the term, 49-50
Sex offenders, 121 Variable Response Inconsistency (VRIN) scale,
Short forms, 226-229 29, 49, 61, 68-74, 78
FAM, 226, 227, 228, 229
Mini-mult, 226, 227, 228 Welsh Anxiety (A) scale, 20, 29, 182-186, 187,
210, 211, 212
MMPI-168, 226, 227, 228, 229
Welsh Repression (R) scale, 20, 29, 182-186
Social class. See Demographic variables
Wiggins’ Content scales. See Content scales
Social Discomfort (SOD) scale, 29, 202, 204,
Wiggins’ Social Desirability (Wsd) scale, 104
205, 206
Within-Normal-Limit (WNL) profiles, 235, 236,
Social Responsibility (Re) scale, 29, 194, 210-
259, 293
211, 212
Work Interference (WRK), 29, 202, 204, 205,
Specific codetypes and diagnosis, 252 206
Substance Abuse Proclivity (SAP) scale, 188
Suicide, 142, 217, 218 X items, 128, 131, 133
Suicide scales, 217-218
Supplementary scales. See specific scales Zero items, 128, 131, 133
''...outstanding...a very significant and lasting contribution to the field. ”

“I should begin by stating that I think that this book is outstanding, and
should make a very significant and lasting contribution to the field. It is
certainly a suitable revision of Roger's very successful 1980 text.

...Roger tends to provide the researcher and clinician with a very complete
discussion of topics based upon a very solid empirical foundation. Roger
provides a scholarly and empirical approach, however, in a manner that
maintains the interest of the average test user and clinician. ”

Dr. Robert P. Archer, Professor

Department of Psychiatry and Behavioral Sciences
Eastern Virginia Medical School

"...a very important text for both the MMPI student and researcher... ”

“...The discussion on demographic patterns affecting the MMPI is...very

good...The presentation of the neurotic triads and other three point
configurations are well done.

In my opinion Dr. Greene's revision of his MMPI book is, to date, the most
comprehensive and objective coverage on the MMPI and the
re-standardization attempts of the MMPI-2. I have also used [other books]
and prefer Dr. Greene's book. I appreciate the level of detail and research
information that this text presents. Yet, it still remains relatively user-friendly,
particular by providing many tables that assist in the interpretation process.
Thus, I consider this a very important text for both the MMPI student and
researcher.”
Dr. Eric A. Zillmer, Director
Graduate Programs in Psychology
Drexel University

Allyn and Bacon

Longwood Division
160 Gould Street
Needham Heights, MA
02194-2310