THE REVISED NEO

PERSONALITY
INVENTORY
CLINICAL AND RESEARCH
APPLICATIONS
THE PLENUM SERIES IN
SOCIAL/CLINICAL PSYCHOLOGY
Series Editor: C. R. Snyder
University of Kansas
Lawrence, Kansas

Current Volumes in the Series:

ADVANCED PERSONALITY
Edited by David F. Barone, Michel Hersen, and
Vincent B. Van Hasselt

AGGRESSION
Biological, Developmental, and Social Perspectives
Edited by Seymour Feshbach and Jolanta Zagrodzka

AVERSIVE INTERPERSONAL BEHAVIORS

Edited by Robin M. Kowalski

COERCION AND AGGRESSIVE COMMUNITY TREATMENT

A New Frontier in Mental Health Law
Edited by Deborah L. Dennis and John Monahan

THE IMPORTANCE OF PSYCHOLOGICAL TRAITS

A Cross-Cultural Study
John E. Williams, Robert C. Satterwhite, and Jose L. Saiz

PERSONAL CONTROL IN ACTION

Cognitive and Motivational Mechanisms
Edited by Miroslaw Kofta, Gifford Weary, and Grzegorz Sedek

THE PSYCHOLOGY OF VANDALISM

Arnold P. Goldstein

THE REVISED NEO PERSONALITY INVENTORY

Clinical and Research Applications
Ralph L. Piedmont

SOCIAL COGNITIVE PSYCHOLOGY

History and Current Domains
David F. Barone, James E. Maddux, and C. R. Snyder

SOURCEBOOK OF SOCIAL SUPPORT AND PERSONALITY

Edited by Gregory R. Pierce, Brian Lakey, Irwin G. Sarason, and
Barbara R. Sarason

A Continuation Order Plan is available for this series. A continuation order will bring
delivery of each new volume immediately upon publication. Volumes are billed only
upon actual shipment. For further information please contact the publisher.
 THE REVISED NEO
PERSONALITY
INVENTORY
CLINICAL AND RESEARCH
APPLICATIONS

RALPH L. PIEDMONT
Loyola College in Maryland
Baltimore, Maryland

SPRINGER SCIENCE+BUSINESS MEDIA, LLC

 Library of Congress Cataloging-in-Publication Data

Library of Congress Cataloging in Publication Data

Piedmont, Ralph, L., [DATE]
The revised NEO Personality Inventory: clinical and research applications / Ralph L.
Piedmont.
p. cm.—(The Plenum series in social/clinical psychology)
Includes bibliographical references and index.

1. NEO Personality Inventory. 2. NEO Five-Factor Inventory. I. Title. II. Series.

BF698.8.N46P54 1998 98-42281
155.2'83—dc21 CIP

Figures 3.1-3.4, 4.1-4.7, 5.1-5.7, and the appendices on pages 111 and 157 are
reproduced by special permission of the Publisher, Psychological Assessment
Resources, Inc., 16204 North Florida Avenue, Lutz, Florida 33549, from the
NEO Personality Inventory-Revised, by Paul Costa, and Robert McCrae,
Copyright 1978, 1985, 1989, 1992 by PAR, Inc. Further reproduction is
prohibited without permission of PAR, Inc.

ISBN 978-1-4899-3590-8 ISBN 978-1-4899-3588-5 (eBook)

DOI 10.1007/978-1-4899-3588-5

© Springer Science+Business Media New York 1998

Originally published by Plenum Press, New York in 1998
Softcover reprint of the hardcover 1st edition 1988
http: / / www.plenum.com
All rights reserved
1098765432
No part of this book may be reproduced, stored in a retrieval system, or
transmitted in any form or by any means, electronic, mechanical,
photocopying, microfilming, recording, or otherwise, without written
permission from the Publisher
 To Rose P., Joanna P., and Dominic P.,
Three of the most endearing NEO PI-R profiles
I have ever come across
 FOREWORD

The assessment of individual differences has a long history. As early as

2200 B.C. the Chinese were employing methods to select candidates for civil
service positions. Over the ensuing centuries philosophers, theologians,
and the nobility all noticed and debated the role of "character" in shaping
the destiny and quality of individual lives. This interest spawned widely
different methods of evaluating the timbre of temperament-bumps on
the head, lines on the hand, shape of the body-all of which were em-
ployed in attempts to gain insight into basic human motives. The emer-
gence of the scientific method and its application to this endeavor
reinvigorated society's efforts in this direction, and an abundant variety of
assessment instruments consequently became available.
The outbreak of World War I created a need for the efficient assess-
ment of individual differences in large groups. Such instruments as the
Woodworth Personal Data Sheet and the Army Alpha Test resulted in gen-
uine breakthroughs in assessment technology. These tests provided stan-
dardized sets of items that permitted quantitative comparisons among
people. Over the years, numerous scales have been developed which have
been based on widely differing levels of psychometric sophistication.
Today, personality scales, clinical assessment devices, and batteries of cog-
nitive tests are established and accepted in society. It is surprising how
many people know their "type" or have taken an MMPI or CPI at some
point in their lives. Just about everyone in America today has taken at least
some form of aptitude test (e.g., SAT, GRE). Assessment appears to be here
to stay, at least for the foreseeable future.
Unfortunately, the sheer number of such instruments has created new
problems for the field of assessment, not the least of which is the problem
of establishing a meaningful taxonomy for classifying the different

vii
 viii FOREWORD

measures. The diverse collection of currently available instruments and

scales reflects the variety of theoretical perspectives on personality and
temperament held by professionals. New scales and instruments are being
developed to fill identifiable gaps between assessment theory and assess-
ment practice. Other scales are being developed in an attempt to keep pace
with the new adaptive styles that have emerged in response to the rapid
changes created by postindustrial society. In addition, the very number of
these scales, together with their potpourri of underlying constructs, is
likely to baffle and bewilder even the most sophisticated assessment pro-
fessional.
Fortunately, an important paradigm shift in the field has occurred
during the past ten years with the emergence of the five-factor model of
personality (FFM). The FFM is a robust taxonomic model that has orga-
nized the myriad of existing scales just described with respect to five
superordinate dimensions: Neuroticism, Extraversion, Openness, Agree-
ableness, and Conscientiousness. These broadband dimensions of person-
ality provide a useful and succinct summary of the various qualities that
personologists have found to be important, relevant, and predictive of life
events. The FFM also provides an organizing framework for understand-
ing and interpreting ongoing discussions about personality. This volume
focuses on a single instrument, the Revised NEO Personality Inventory
(NEO PI-R), which is currently the only commercially available instru-
ment explicitly designed to measure the dimensions of the FFM. However,
the model underlying the NEO PI-R is part of the much broader history of
personality assessment.
The first chapter details a review of the FFM and outlines what this
model can and cannot do. This treatment provides an empirical and theo-
retical context for understanding the constructs embodied in this instru-
ment. It is important to realize that this book is not merely a cookbook" of
II

NEO PI-R interpretations. Instead, the author makes every effort to pro-
vide the reader with a context and a foundation for understanding how
and why the instrument was developed, and how and when it should be
applied. The wide-ranging coverage of the research literature in ensuing
chapters illuminates the versatility of the NEO PI-R in differing contexts
that emphasize clinical, applied, cross-cultural, and psychometric consid-
erations. I believe that this book really hits its mark and that it will prove
to be an effective resource for anyone who wishes to develop a deeper un-
derstanding of the area of personality assessment. The book is delightfully
reader-oriented and user-friendly. Dr. Piedmont has a flair for presenting
technical issues in clear, readable prose that should be both accessible and
engaging to a wide audience of both students and professionals.
FOREWORD ix

The chapter on profile analysis provides insight into addressing the

multidimensionality of personality represented in the NEO PI-R. The
challenge in this case is to learn how to integrate the wide personological
spectrum covered by 35 scales into cohesive, interpretively useful, and
predictively salient "chunks." Again, Dr. Piedmont provides a practical
and understandable framework for making these kinds of interpretations
that are so imperative for sound assessment practice.
The chapter on using observer ratings opens still another door to as-
sessment. Although the availability of a validated rater form is a rarity
among current instruments, there are many advantages to this form of as-
sessment. This is most clearly evident in the discussion of Cross-Observer
Agreement analysis with married couples. The logic and empirical utility
of this approach is nicely outlined and the case histories demonstrate the
value of the interpretive process.
Finally, the chapter on research applications provides an excellent ar-
ticulation of sound empirical strategies for using the NEO PI-R, or any
other assessment protocol. Useful materials are provided to help both clin-
icians and researchers formulate relevant theoretical and applied ques-
tions. Regardless of one's view on the utility of the FFM, this book will be
useful for learning about both profile interpretive approaches to personal-
ity assessment and about the manner in which test validation should pro-
ceed. There is no question that this book will assist the reader in
developing proficiency in the use of the NEO PI-R. More important, it will
provide instruction in conducting the general enterprise of personality as-
sessment and research.
Paul T. Costa, Jr.
Baltimore, Maryland
 PREFACE

The mental health field is going through a number of important transfor-

mations. One such development is the field's increasing theoretical diver-
sity. There are many different types of practitioners who are applying a
broad range of therapeutic paradigms, from the more traditional behav-
ioral and cognitive frameworks to the emerging areas of spirituality and
holistic medicine. The philosophies and treatment approaches are as nu-
merous as the professionals themselves. Another reality is the growing
concern over health care and how it will be managed. The rise of managed
care companies is an outgrowth (or impetus) for this new trend. These or-
ganizations have emphasized the need for clinicians to provide effective
treatments in a timely and cost-efficient manner. Finally, with the increas-
ing awareness of the ethical dimensions of practice, there is the recognition
that practitioners need to be made accountable for their services, both to
the consumers of their services and to themselves. Therapists must be able
to provide evidence of the efficacy of their services to consumers. Cer-
tainly, many factors influence how much success a client will experience;
some of these are beyond the reach of the therapist, but consumers have
the right to know just how much improvement they are likely to experi-
ence in therapy. Clients, like consumers in other areas, are entitled to valid
assurances that their therapy will have a potential benefit. The responsi-
bility of therapists to themselves is to maintain a high level of clinical ex-
pertise. This requires that the therapist receive useful feedback on the
effects of his or her interventions. Therapists must be aware of such factors
as the overall extent of improvement experienced by clients as a result of
interventions, whether interventions are more effective with certain types
of problems than with others, and whether certain intervention strategies
are working better than others.
xi
 xii PREFACE

All of the forces just mentioned are moving the mental health field to-
ward a recognition of the need for better documentation of clinical effi-
cacy. The mental health field needs to provide support for what it does
[quantitatively] in an objective, substantive manner. The technology best
suited for providing this kind of information is psychological assessment.
Traditionally, measures of clinical practice have been the purview of psy-
chology. Psychologists usually receive extensive training in the application
and interpretation of psychological measures. In fact, many clinical as-
sessment instruments are usually restricted to psychologists who have had
detailed graduate training in testing. Therefore, many nonpsychologists
believe that they do not have access to any types of psychological measures
and thus do not employ testing as part of their practice, nor do they see it
as part of their clinical identity. This is unfortunate, because there is a wide
range of nonclinical psychological measures that are appropriate for
nonpsychologists and that would also be relevant to, and helpful for, their
practice. Being unaware of these measures deprives such therapists of im-
portant sources of information that could be useful for client assessment,
for documenting clinical efficacy, and for obtaining valuable clinical feed-
back.
The measures to which I refer are instruments designed to assess nor-
mal personality qualities. Such scales are less restricted by test publishers.
Usually, a graduate course in psychological assessment and/ or some su-
pervision in their use is all that is required to gain access to these materials.
Personality questionnaires can be very useful in telling a therapist about a
wide range of client dispositions, needs, and motivations. Although they
may not be diagnostically revealing in their own right, these individual-
difference variables set important parameters for any therapeutic interac-
tion. For example, a person who is closed and rigid, who does not
experience a wide range of emotions, and who requires much structure
may not benefit from insight therapy. Conversely, an individual who is
open and has very permeable inner boundaries may keenly experience
whatever negative affect is dominant. By assessing a client's personality, a
therapist can gain insight on how the client sees the world and copes with
it. Depending on the constructs being measured, a therapist may also be
able to anticipate therapeutically relevant outcomes. For example, a client
high on Machiavellianism may have trouble in establishing trust, while a
client high on Anxiety may need a great deal of reassurance and support.
There is a wide range of personality questionnaires that span numer-
ous constructs. These measures vary from assessing specific constructs
(e.g., Fear of Fat Scale) to more global, multidimensional inventories (e.g.,
the California Personality Inventory). Personality measures are usually
easy-to-use scales that, with some reading and practice, can be used by
PREFACE xiii

most mental health professionals. The purpose of this book is to introduce

clinicians to a general personality measure that is most relevant for the
clinical context-the Revised NEO Personality Inventory (NEO PI-R). This
well-developed instrument is designed to measure the five major dimen-
sions of personality: Neuroticism, Extraversion, Openness to Experience,
Agreeableness, and Conscientiousness. These constructs define what is
known as the five-factor model of personality (FFM)-a trait-based taxon-
omy of personality dispositions. These five factors represent independent
constructs that have been shown to provide a comprehensive description
of normal personality. There is an extensive and constantly expanding re-
search literature that continues to document the utility of this model for
predicting real-life outcomes in a number of applied contexts. This book
shows how the FFM can provide a very useful framework for conceptual-
izing people and for anticipating the directions in which they will move.
This book therefore introduces clinicians to the FFM and the instru-
ment designed explicitly to measure it-the NEO PI-R. My goal is to chal-
lenge the ways readers think about personality and to familiarize them
with an empirically sound method for concretizing those conceptualiza-
tions. Strategies are presented for using the instrument in clinical contexts.
The research applications of the NEO PI-R are also discussed. Paradigms
for approaching many of the most salient issues in research are provided.
Every attempt has been made to provide information that is as current as
possible. I hope that this book will prove the clinical value of personality
assessment and interest readers in employing this technology in their own
clinical work. The eventual benefit to our field will be the accumulation of
a body of knowledge that demonstrates the efficacy of our interventions
and the value of our services.
Ralph L. Piedmont, Ph.D.
Baltimore, Maryland
 ACKNOWLEDGMENTS

In the writing of any book, there are numerous people who contribute to
the project even though there may be only one author. In this case, I am in-
debted to a number of colleagues and friends who contributed information
and material, as well as moral support. First and foremost, I would like to
thank my wife, Rose, who read and reread the manuscript several times to
check structure and grammar. I would also like to thank Gail Worrall, who
edited the entire manuscript. The manuscript was certainly made more
"user friendly" because of the efforts of Rose and Gail. I would also like to
express my gratitude to Drs. Joseph Ciarrocchi and Thomas Rpdgerson,
who shared with me some of their clinical experiences using the NEO PI-R.
Special thanks go to Reg Watson, who created the NEO PI-R profile forms
used in this book. I would also like to thank those individuals who con-
sented to have their NEO PI-R profiles included as examples throughout
the book. I would like to thank Dr. Robert Wicks, my colleague and friend,
who provided the initial inspiration for doing this book. Finally, I would
like to thank Paul T. Costa, Jr. and Robert R. McCrae for all they taught me
about the NEO PI-R in particular, and about good science in general.

xv
 CONTENTS

Chapter 1
Personality and Its Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

What Is Personality? ............................... , . . . . . . . 2

Genotype and Phenotype. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Change and Stability in Personality. . . . . . . . . . . . . . . . . . . . . . 5
The Value of Personality Assessment in a Clinical Context . . . . . . 8
Overview............................................. 8
The Clinical Yield from Personality Assessment. . . . . . . . . . . . 10
Qualities of Personality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
What Is a Taxonomy?... .. . ...... ..... . .......... .... .... .. 17
Taxonomy versus Typology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
The Value of a Taxonomy for Personality Assessment. . . . . . . 19
The Lexigraphic Hypothesis.. . . . . . .. . . . . . . . . . . . . . . . . . . . . .. . . 20
Searching for a Linguistic Structure . . . . . . . . . . . . . . . . . . . . . . 24
From Adjectives to Sentences: The NEO Model. . . . . . . . . . . . 27
Recommendations for Approaching This Book . . . . . . . . . . . . 32

Chapter 2
Psychometric Overview of the NEO PI-R . . . . . . . . . . . . . . . . . . . . . . . . 35

Outline of Scales. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Reliability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Factor Structure. . . . . . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . 40
 xviii CONTENTS

The Question of Comprehensiveness . . . . . . . . . . . . . . . . . . . . . . . . . 42

Correspondence between the NEO PI-R and Other
Measurement Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
What the NEO PI-R Can Tell Us about Other Scales.. . . .. . . 46
The Question of Self-Distortion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Self-Peer Congruence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
The Reverse Acquaintanceship Effect. . . . . . . . . . . . . . . . . . . . . 54
The Logic of Assessment Using the NEO PI-R. . . . . . . . . . . . . 56
Normal versus Abnormal Personality. . . . . . . . . . . . . . . . . . . . . . . . 57
Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
The Five-Factor Model and Its Relations to
Clinical Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Robustness of the Five-Factor Model . . . . . . . . . . . . . . . . . . . . . . . . . 71
Heritability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Cross-Cultural Generalizability. . . . . . . . . . . . . . . . . . . . . . . . . . 73
Theoretical Foundation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

Chapter 3
Interpreting the NEG PI-R . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

The Use of Validity Scales in Assessment. . . . . . . . . . . . . . . . . . . . . . 79

Social Desirability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Content-Free Validity Scales. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
The Value and Limits of Self-Report Data. . . . . . . . . . . . . . . . . 83
NEO PI-R Facet Scales and Their Interpretations. . . . . . . . . . . . . . . 84
Neuroticism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Extraversion .......................................... 86
Openness to Experience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Agreeableness. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Conscientiousness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
NEO PI-R Interpretations and Select Case Profiles. . . . . . . . . . . . . 92
Case History: Debbie K. ........................... . . . . . 93
Case History: Joe W.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Case History: Frank and Judy V. . . . . . . . . . . . . . . . . . . . . . . . . . 100
Providing Feedback. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
What If My NEO PI-R Profile Does Not Match
the Self-Concept? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Error in Test Completion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Recontextualizing the Client's Motivation. . . . . . . . . . . . . . . . . 108
 CONTENTS xix

Chapter 4
Profile Analysis Using the NEG PI-R. . . .. . . . . . . . . . .. . . . . . . .. . . . 113

Organizing NEO PI-R Information. . . . . . . . . . . . . . . . . . . . . . . . . . . 113

General. . . . . . . . . . . . ... . . . ... . .. . . . . . . . .. . . . .. . .. .. . . . . .. . . 113
Profiling NEO PI-R Scores . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Selected Case Histories. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Case History: Robert R.. . . . . . . . . . . . . . .. . . .. . . . . .. .. . .. . . 127
Case History: Angela W. . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . 131
Case History: Brian M.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
Case History: Barbara W. ............................... 139
Case History: Beverly N.. . . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . 144
Case History: Torn S. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
Case History: Erica J. . . . . . . .. . . . . . .. . . . . . .. . . . .. . . . ... . . 149
Case History: Sally Ines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

Chapter 5
Applications of the Rater Version of the NEG PI-R . . . . . . . . . . . . . . . . 165

Using Observer Ratings. . .. . . . .. . . . . . . . .. .. . . . . . . .. . . . . . .. . . 166

Observer Ratings with Married Couples. . . . . . . .. . . . . . . . . . . .. . 171
Self-Other Congruence in Couples. . . . . . . . . . . . . . . . . . . . . . . 173
Cross-Observer Agreement Analyses. . . . . . . . . . . . . . . . . . . . . 176
Cross-Observer Agreement Analysis: Selected Cases . . . . . . . 186
Clinical Yield from Cross-Observer Agreement Analysis. . . . 204
Observer Ratings as Predictors of Outcome .. . . . . . . . . . . . . . . . . . 205

Chapter 6
Research Applications with the NEG PI-R . . . . . . . . . . . . . . . . . . . . . . . 209

The Need and Role of Research. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210

Overview.. .... ....................................... 210
How Clinicians Can Provide Important Insights. . . . . . . . . . . 212
Issues Relating to the Factor Analysis of NEO PI-R Information. . . 213
Factor Orthogonality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
Number of Factors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
Recovering the NEO PI-R Structure in New Samples. . . . . . . 222
xx CONTENTS

Application of NEO PI-R Data to Validity Research . . . . . . . . . . . . 232

Construct Validity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
The Incremental Validity Paradigm . . . . . . . . . . . . . . . . . . . . . . 237
Outcome Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
Understanding Change over Time . . . . . . . . . . . . . . . . . . . . . . . 244
Selecting Treatments for Persons. . . . . . . . . . . . . . . . . . . . . . . . . 247
Understanding Our Clients. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . 252

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255

Author Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277

Subject Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281

 CHAPTER 1

PERSONALITY AND
ITS ASSESSMENT

Every time I teach a course in personality theory, I begin the class by ask-
ing the students, "What is personality?" Given the newness of the class sit-
uation and the perennial need of college students to remain anonymous to
teachers, this question is always followed by a long silence. In order to
jump-start this social interaction, I follow up with a request. I ask students
to raise their hands if they believe they have a personality. Invariably, there
are muffled coughs, eyes begin to dart around the room, and perhaps two
or three people will, grudgingly, raise their hands. "Good," I say. "It seems
that some people are certain that they have a personality but others are not
as sure as to whether they have one or not. Perhaps some people may have
had a personality at some time in the past, but do not have one currently,
while still others may be engaged in an ongoing search to find one."
On a good day, these comments will elicit not only several good chuck-
les, but some hands will be raised in an attempt to define the term "person-
ality." The variety of definitions I get for the term clearly tells me that we all
know personality when we see it, but we cannot put our fingers on what ex-
actly we mean by "it." Personality is something that defines who we are as
people, yet some believe that it is always changing. Some people are said to
have "personality"; others seem to lack it. Some believe that personality is the
aggregate of our behaviors and attitudes, yet others see personality as some-
thing more fundamental: It is the "why" to our behavior. No doubt a course
on personality theory is exactly what these students need! Personality is a
central concept in the social sciences because it speaks about people: Who
they are, how they come to be, and where they are heading in their lives. It is
the foundation for building theories of psychopathology and treatment.
1
2 CHAPTER 1

The purpose of this chapter is to outline some of the important issues

surrounding current notions of personality and to describe a model of per-
sonality that is becoming quite prominent in the field: the Five-Factor Tax-
onomy. The development and application of this model in the social
sciences is discussed. The remaining parts of this book are concerned with
the psychometric and clinical utility of the NED Personality Inventory-Re-
vised (NED PI-R), a measure designed explicitly to capture the dimen-
sions underlying this personality model. It is hoped that this volume will
assist users of the NED PI-R in developing their interpretive skills as well
as providing strategies for using the instrument in a research context.

WHAT IS PERSONALITY?

Before beginning, it seems appropriate to offer a working definition of

personality that can be amended as we go along. For our purposes, per-
sonality can be defined as the intrinsic organization of an individual's
mental world that is stable over time and consistent over situations. There
are three important points to this definition. First, personality represents
some structured system by which individuals organize themselves and
orient to the world around them. This system is clearly located within the
person and is not imposed by the environment. Second, personality is sta-
ble over time. This means that there is something about who we are and
what we are like that remains consistent through our lives. There is more
to say about personality change, but for now we can modestly say that de-
spite all the "change" we may have experienced over our lifetimes, there is
some thread that seems to be consistent over all the years. There is some-
thing about who I was as a child or adolescent that lingers with me today.
Finally, personality is consistent from one situation to another. As the old
saying goes, "Wherever you go, there you are." Although specific behav-
iors may change from one context to the next, who we are inside and how
we perceive the world remains the same. Although our behaviors may
change, the personal goals we may be pursuing remain essentially the
same.
This person-centered definition should not be construed as denying
the role of the environment in shaping our personalities. Culture, context,
and situation all have an influence on our development. But there lies
within us some kind of psychological "stuff" which provides the basis for
the needs we have, the ways in which we perceive and interpret the outer
world, and the goals we ultimately pursue in our lives. Nor should this de-
finition be thought of as reflecting personality as a static quality. Quite the
contrary; personality is a dynamic structure. It is always responding to
 PERSONALITY AND ITS ASSESSMENT 3

needs that arise from within and without the individual. Further, there is
a developmental aspect to personality. Just as we are not born physically
mature, so, too, personality is something that begins in a simple form and
becomes progressively more sophisticated over time. These changes also
follow a lawful, orderly path.
This definition is saying there is something inside us that emerges
over our lifetimes that comes to define us as unique individuals. This inner
system helps to orient us toward the world and enables us to adapt to the
demands of our environment. This inner structure has an inherent lawful-
ness to it, in that different personalities will lead individuals to follow dif-
ferent life paths. Further, the process by which personality unfolds will
also follow certain regularities that can be known and understood. Thus
the process of change is complex and there is a need to differentiate be-
tween changes that reflect superficial, adaptive modifications and changes
that reflect more fundamental, structural shifts.

GENOTYPE AND PHENOTYPE

To understand personality, one needs to have an appreciation of the

various levels on which personality comes to exist and be expressed. This
is particularly true when one wishes to evaluate personality change. We
need to know what it is that is changing and what that represents about
one's personality. To begin this analysis we need to understand the con-
cepts of genotype and phenotype and their relevance for personality.
A genotype refers to the fundamental organization of the individual;
the basic "stuff" that a person has been endowed with through their ge-
netics. Genotypes are latent dimensions that establish for an individual a
trajectory through time. Phenotype, on the other hand, refers to the ex-
pression of the genotype at any given moment in time. Put all the pheno-
types together, and you have the genotype. These two concepts tell us that
there is a lawfulness to change. Certain quantities or qualities may vary
over time, but such variability may not necessarily reflect any fundamen-
tal change. Rather, change may be indicative of a gradual unfolding
process for a given characteristic.
A good example of this is physical characteristics. Some readers may
remember a fellow named Jack La Lane. Back in the 1950s he was one of
the first to bring exercise and fitness to television. I remember watching his
show when I was very little. Jack is an enthusiastic person (one could say
he had a great personality) who was very muscular. He would always
wear a tight jumpsuit that would dearly show off his well-developed bi-
ceps, muscular chest, and angular waist. No doubt he was a physically
well-developed individual. Recently I saw a photo of him (at the time of
4 CHAPTERl

this writing he is still alive and in his 80s), and he was still wearing that
tight jumpsuit! He is still energetic, still works out and stays fit. If you were
to see him today, you would most certainly say that he is a muscular man.
But there is no doubt that Jack's body today is quite different from what it
was 40 years ago. His body has experienced the inevitable changes of
aging: There is more body fat, muscles are not as "tight" as they once were,
and the skin appears looser around the body. But despite these very obvi-
ous changes, Jack is still described as a muscular man. The reason for this
consistency in labeling is that we recognize that what is meant by "mus-
cularity" is different at different stages of development. Phenotypically he
is different, but genotypically he has not changed at all.
The same can be said for personality. We can think of many different
qualities, but these qualities may mean different types of behaviors at dif-
ferent ages. For example, being an extravert in college may mean going to
many different parties and staying out late. It may mean being "rowdy"
and loud. However, at age 40 this quality may take on a quite different
form. Rather than the raucous college parties, it may be hosting formal sit-
down dinner parties for friends and coworkers. It may mean getting in-
volved in community activities. Clearly extraversion may take a more
conventional form later in life. But there is no doubt that the person re-
mains an outgoing, socially active person.
It is important when we apply measurement instruments to assess in-
dividuals that we have some understanding of the level of analysis being
assessed. Measures of phenotypic behavior may be very prone to changes,
either over short periods of time or even across different situations. For ex-
ample, measures of state anxiety (e.g., State-Trait Anxiety Inventory, Speil-
berger, Gorsuch, & Lushene, 1970) or of depressive symptomatology (e.g.,
Hamilton Depression Inventory, Reynolds & Kobak,1995) are designed to
detect fluctuations in very specific aspects of functioning over relatively
short time intervals. They are useful for understanding treatment impact or
individual responses to specific situations. Changes on this level, although
adaptively important for specific situations, may be less diagnostic of more
fundamental, broad-based change. Genotypic measures, on the other hand,
aim at uncovering the basic psychological strata of the person. A measure
such as the NEO PI-R is designed to access a person's more fundamental
temperaments, those qualities that drive, direct, and select behaviors.
Scores on this type of scale are more likely to remain constant over time and
across situations. However, changes on this level may indicate more pro-
found shifts in personality. Thus, change needs to be evaluated carefully.
Another important realization related to this is the diversity of human
behavior through which genotypes can be expressed. People have devel-
oped extensive behavior repertoires for fulfilling their needs. They adapt
 PERSONALITY AND ITS ASSESSMENT 5

by means of behaviors that are successful in specific environments. There-

fore, changes in behavior may not necessarily signal a corresponding
change in the need being expressed. For example, consider the trait of
dominance. What types of behaviors characterize a dominant person? Is a
dominant person one who acts in a controlling, directive, forceful manner?
Or is a dominant person one who receives acts of submission from others?
The answer is, of course, "it depends." It depends on the situation. In ei-
ther situation the apparently different phenotypic behaviors reflect the
same personality quality. Perhaps Freud had it right when he said that any
need can be expressed through any number of behaviors, and any specific
behavior can gratify any number of needs.
The concepts of genotype and phenotype have important implica-
tions for how we need to conceptualize change in the individual. Certainly
we should not be fooled into thinking that behavior changes reflect deeper
shifts within the person. But we also need to rethink how our clinical in-
terventions are aimed at influencing our clients. This raises important is-
sues that we discuss later in this book about the goals of therapy. Does
therapy aim to make genotypic or phenotypic changes in the person? Are
genotypic changes possible? Before we proceed to those questions, we first
look at how change and stability are assessed.

CHANGE AND STABILITY IN PERSONALITY

Change, like most things, seems to be a very straightforward matter.

But experience teaches us otherwise. Things that look simple on the sur-
face usually contain many nuances and subtleties. Such is the case for un-
derstanding change. There are two issues that need to be considered: rank
order stability and mean level change (see Caspi & Bern, 1990, for a more
extensive evaluation of personality change).

Rank Order Stability

Perhaps one of the most frequently used procedures for determining
whether someone has changed on some dimensions is to obtain scores
from two time periods and correlate them. A high positive correlation
would show that those who scored high at Time 1 continued to score high
at Time 2. This would be evidence of temporal stability or test-retest relia-
bility for the scores. Could it be concluded from such a finding that scores
did not change? Have scores remained stable? The answer to the first ques-
tion is "no" and to the second, "yes."
Let's take a look at this from another perspective. Imagine that you are
a therapist interested in documenting the impact of your therapy on
 6 CHAPTER 1

improving the mental health of your clients. To this end, each client is as-
sessed on a mental health inventory when entering therapy. Ahigh score in-
dicates mental distress. After completing therapy each client completes the
mental health inventory again. After treating about 100 clients, you "crunch"
the numbers; you correlate Time 1 scores on the inventory with Time 2
scores. If you were successful in treating your clients, what would you ex-
pect to find? Should there be (a) a high positive correlation (i.e., high scores
at Time 1 continued to be high at Time 2), (b) a high negative correlation (Le.,
high scores at Time 1 became low scores at Time 2), or (c) should there be no
correlation (Le., no relationship between scores at the two time periods)?
Finding a zero correlation would not be helpful for drawing any types
of inferences. Such a result could be the product of many influences, in-
cluding the lack of reliability of your assessment instrument. A negative
correlation seems to be a desirable outcome, with those scoring high at first
becoming much lower at the end of treatment. However, the reverse
equally applies: Those who were low at Time 1 (indicating a degree of men-
tal health) scored higher at Time 2 (indicating that the treatment made them
less healthy). The most desirable result would be to find a positive correla-
tion. This would indicate some reliability in your assessment instrument.
But if you found a high positive correlation between the two sets of
scores would that mean you were not as successful as you thought in help-
ing your clients? That scores are basically the same at the end of treatment
as they were at the start? Not necessarily. In evaluating a correlation, it is
important to appreciate what that correlation tells you about your data.
Correlation is a measure of rank order stability. That means that if you obtain
a high positive correlation, the ranking of scores at Time 1 are similar to the
ranking of scores at Time 2. In terms of our example, those who were more
mentally distressed upon entering therapy were also among the more men-
tally distressed when leaving therapy. The term "more" is a relative term.
More distressed in comparison to whom? Well, the others in the study. How
distressed were they? That question is not answered by correlation.
Remember that rank order is an ordinal level of measurement. It tells
you position but not magnitude. Thus, the person who ranked number
one in mental distress at Time 1 could have improved greatly over treat-
ment, resulting in a much lower score at Time 2. However, if everyone else
in the sample improved at a similar rate over treatment, then this person
may again be ranked number one at Time 2. If everyone else also keeps the
same rank order, then the correlation between these two sets of scores will
be perfect (Le., +1.0), even though there were significant decreases in the
amount of psychological distress. Therefore, it is possible to obtain a high
positive correlation between Time 1 and Time 2 mental health scores even
when there are large changes in scores.
 PERSONALITY AND ITS ASSESSMENT 7

Correlation tells us about the relative standing of cases in a distribu-

tion. Maintaining relative standing will create higher correlations. Corre-
lations around zero indicate that the relative standing of cases is changing
in a nonpredictable way. So, if you are interested in determining whether
scores have changed, then correlation would not be the best way to analyze
your data. But if you were interested in knowing whether change over
time proceeded in an orderly manner, such that all individuals are chang-
ing at a constant rate, then correlating scores would be helpful.
In terms of understanding personality stability, one needs to have a
critical eye on test-retest coefficients that are presented by test manuals.
High test reliability does not mean, on the face of it, that individuals' per-
sonalities are not changing. Personality may indeed be changing, but at an
equal rate throughout the group. To evaluate whether there have been
shifts in the magnitude of various qualities, you need to consider the no-
tion of mean level change.

Mean Level Change

Mean level change refers to actual changes in the magnitude of a quan-
tity over time. As noted in our example, it is possible to have a high level of
stability while also evidencing a large amount of change. Therapy may in-
deed reduce the actual amount of "distress" experienced by clients. This
would be reflected in a comparison of means for the two time periods. Any
type of repeated measures analysis would certainly accomplish this task.
In terms of personality assessment the question of mean level change
seeks to understand whether various aspects of th~ individual increase or
decrease. Do individuals become more conventional as they age? Do lev-
els of negative affect decrease as one gets older? The question of change in
magnitude has important implications for both genotype and phenotype.
Our need is to identify not only increasing adaptive fit with the environ-
ment, but also to determine whether the underlying qualities of our per-
sonality also vary in degree over time. If genotype changes, then the issues
are how much and over what time frame?
What the data seem to indicate at this time is that there is a large
amount of genotypic plasticity in personality from birth to about age 30
(e.g., J. Block, 1971; Haan, Millsap, & Hartka, 1986; Helson & Moane, 1987;
Jessor, 1983; Siegler et al., 1990). That means that the underlying qualities of
our personality continues to evolve over the first third of the life span. Be-
tween the periods of 17 to 30 years of age there is a tremendous amount of
"personality" change in terms of both rank order stability and mean level.
Levels of negative affect and extraversion decrease while levels of conven-
tionality increase. After age 30, though, there is little or no genotypic
8 CHAPTERl

change, although phenotypic variability will continue to be noticed over

the entire life span. However, even this phenotypic variability can be mean-
ingfully related to genotypic standing (e.g., Kaiser & Ozer, 1997; Ormel &
Wohlfarth, 1991). This means that much phenotypic variation can be linked
to the operation of underlying, stable qualities of the individual.
To summarize the discussion, personality is an important concept for
guiding our understanding of individuals. We recognize that there is some
internal structure that organizes who we are and the goals we pursue. This
internal structure develops and matures over time, creating a real sense of
change within us. The things we value and the situations we seek evolve
over time. However, we reach a point at which this internal growth stops.
We reach a point where any change reflects not so much an internal reor-
ganization as a modification of more superficial aspects of our character
that maximize our adaptiveness to the particular environment we occupy.
Perhaps an equally important understanding is the realization that
this entire process follows a lawful plan. Our temperament, or personality,
creates for us a trajectory through time. Our needs, desires, wants, and
goals evolve in a specific direction. We can see this when we look back on
our own lives, and recognize the pattern in our behaviors, decisions, and
preferences. Even though we have changed genotypically from childhood
to adulthood, there is still a sense of connectedness. Life appears to
progress as a series of building blocks, one laid on top of the other, creating
a structure that we call our identity.
Knowing that there is this orderly process can provide us with im-
portant information about ourselves and others. Understanding how
personality is organized and the major qualities that constitute that orga-
nization can provide insights into human functioning and, ultimately, en-
able us to make predictions about meaningful life outcomes. It is here that
we find the great value of personality assessment for clinical work: en-
hanCing client understanding and clinical prediction.

THE VALUE OF PERSONALITY ASSESSMENT

IN A CLINICAL CONTEXT

OVERVIEW

Today's mental health care environment is undergoing rapid change.

Therapy used to be an unhurried and time-consuming endeavor. It was
not uncommon 15 years ago for a client to be in therapy for years. With the
rise of managed health care systems, this way of operating is quickly com-
ing to an end. Although this new system is making counseling and psy-
PERSONALITY AND ITS ASSESSMENT 9

chotherapy available to more individuals than ever before, the price for
this new availability is a level of cost consciousness heretofore unexperi-
enced in the social sciences. Third-party payers are expecting therapists to
do more, in less time, with greater impact. In common parlance, these ser-
vice providers are looking for greater "bang for the buck."
The result of this process has been a greater emphasis on short-term
therapies. Under the traditional system it would not be uncommon for the
first 6 months of therapy to focus on getting to know client and problem.
Today, in 6 months the therapy should be coming to an end. Therefore,
clinicians do not have the time for understanding the client that they once
enjoyed. New technologies are needed to expedite the evaluation and di-
agnostic process, and measures of personality can help fill this need.
There are numerous assessment devices that have recently appeared
aimed at uncovering important aspects of clients. These new-generation
instruments represent a high level of psychometric sophistication and can
be very useful for prOViding clinically relevant insights. But this you al-
ready recognize; after all, that is why you are reading this book. The NEO
PI-R can be very useful in a clinical context because it can make you aware
of the types of needs clients seek to satisfy, their interpersonal styles, value
systems, and levels of ongoing emotional distress. The NEO PI-R can also
be helpful in identifying therapeutic interventions that may be particularly
effective. For example, you would not want to put a social introvert in
group therapy, or try insight therapy with an individual who has little or
no personal insight or empathy. This book will provide you with a first
step in learning to use the NEO PI-R in this way. But there is another value
to personality assessment that is frequently overlooked.
As professionals we have an ethical responsibility to both ourselves
and our clients to provide them with high-quality, effective interventions.
As therapists, we need to have feedback on our interventions, information
about what therapies work and with what type of client. To determine the
amount of improvement we are able to bring about. This information is
also useful to our clients, who need to be informed about the efficacy of
our services in clear, concrete ways. Psychological assessment provides a
straightforward method for documenting treatment efficacy. Such mea-
sures provide a standardized, quantifiable means for determining the na-
ture and degree of change. It is important that people have access to
therapists, but it is even more important that people have access to thera-
pists who will actually improve their mental status.
You can ask any therapist whether the clients who come to see them
benefit from their services, and in almost every instance the reply will be
"Of course!" What other answer is there? After all, even therapists have
to maintain some sense of personal efficacy and professional value. If a
10 CHAPTER 1

therapist thought that he or she was not helping clients, the therapist
would certainly not stay in that line of work. But the real question is, "Do
your clients get better?" Just taking a therapist's word on the matter is in-
sufficient. Some kind of empirical documentation of benefit is needed. By
how much do clients improve? In what ways do clients improve? These
are important questions that health care providers, clients, and therapists
themselves are increasingly asking. The answers will need to be in a lan-
guage that is objective and universally understood: a quantitative one.
In this cost conscious, highly competitive environment it becomes cost
effective not only to identify effective therapists (individuals who can de-
liver some therapeutic clout in a short period), but also to be able to match
clients with therapists who are particularly adept with the kind of problem
the client is facing. Therefore, health care networks are now beginning to
conduct their own clinical outcome studies. Eventually, they will begin to
demand from their therapists some evidence of their own effectiveness.
Obtaining such evidence will require the use of standardized tests.
The paradigm for this type of study is quite straightforward. Simply give
a test(s) to clients when they enter therapy and again when they exit. In
this way it can be shown how much change there is over the identified di-
mensions. Combining these data with clinician ratings of effectiveness
(simple Likert scale ratings have been developed and are frequently used),
one can identify those personality characteristics of the individual noted at
the beginning of therapy that are linked with ratings of treatment efficacy
(these issues and methods to address them are discussed in more detail in
Chapter 6).
The implications of this type of research are legion. The value they serve
for improving therapy efficacy and developing new models of intervention
cannot be overstated. Although many bemoan the new health care behe-
moth, there may be a silver lining to this dark cloud. In developing a facil-
ity with assessment technologies now, you will give yourself an edge in this
developing health care landscape. When you begin to use the NED PI-R
along with other clinically relevant indices, you will not only be gathering
important information about your client that can be directly applied to your
therapeutic work, but you will also be creating a clinical database. This in-
formation can provide you with important feedback on the quality of your
work and important documentation of your psychotherapeutic effect.

THE CLINICAL YIELD FROM PERSONALITY ASSESSMENT

The value of using structured personality instruments in a clinical

context is that they provide very clear, tangible evidence of where a client
is psychologically. This information is in a form that is readily under-
PERSONALITY AND ITS ASSESSMENT 11

standable by a wide range of professionals and represents an objective,

broad perspective on the client's functioning. Costa and McCrae (1992a)
have identified six ways that this type of information can be fruitfully em-
ployed in your clinical work. These uses are outlined in Table 1-1.
The first value is for understanding the client. Rather than fOCUSing ex-
clusively on symptomatology, the NEO PI-R provides a broad-based
assessment of the individual's personality. This includes a thorough un-
derstanding of the patient's strengths and weaknesses. The NEO PI-R pro-
vides information relevant to interpersonal style, character, levels of
emotional well-being, aspiration levels, and a wide range of other psycho-
logically relevant information. The NEO PI-R can also provide a larger
context for understanding the presenting problems. Are the current dis-
tressing symptoms a reaction to recent events or are they symptoms of an
enduring and pervasive maladjustment? The answer to this question can
have important implications for determining the extent of improvement
likely to occur over treatment.
For example, consider a client who presents with the complaint of fre-
quent arguing with his wife. The level of marital distress is becoming so
high that the marriage may be in danger of failing. What kind of man is the
client? Is he someone who is habitually very argumentative and naturally
encounters many interpersonal conflicts in his life? Does this person have
stormy relationships with other family members, neighbors, and cowork-
ers? Is he just naturally antagonistic? Or is he usually an easy-going guy
who is mostly very pleasant and engaging? Could he just be going through
a very stressful period at work? The answers to these questions have very
direct implications for the type of treatment to be employed and the kind
of outcome expected.
In the first scenario, the client is temperamentally antagonistic and
much work will have to be done to develop a level of insight necessary to
effect real change in his behavior. Even then, his wife may not be able to
tolerate his curmudgeonlike persona. Much work will most likely need to

TABLE 1-1 Clinical Uses for Personality Measures

Understanding the Client
Differential DiagnOSiS
Empathy and Rapport
Feedback and Insight
Anticipating the Course of Therapy
Matching Treatments to Clients
Note. From "Normal Personality Assessment in Clinical Practice: The
NEO Personality Inventory," by P. T. Costa. fr. and R. R. McCrae. 1992,
Psychological Assessment, 4, p. 5-13.
12 CHAPTER 1

be done with both husband and wife. However, in the second scenario it
is clear that the problems are situationally induced. There is a transient
stressor that is turning a usually goodnatured person into an ogre. Cer-
tainly when the stressor subsides many of these problems may disappear
as well. Treatment may wish to focus on developing short-term coping
skills and a complete remission of symptoms is likely.
Personality instruments like the NEO PI-R can be helpful in outlining
the underlying dispositions of clients and alerting you to the presence of
internal motivations that may be fueling current problems.
The second clinical value concerns differential diagnosis. Although the
NEO PI-R is not designed to assess pathology, it can be useful for provid-
ing information relevant to making a diagnosis or for ruling one out. For
example, a high score on extraversion, one of the major domains assessed
by the NEO, would rule out schizoid personality disorder. A high level of
emotional dysphoria is a feature of all clinical patients, but the presence of
only high Self-Consciousness (a facet of the NEO) would suggest a social
phobia rather than depression.
Personality inventories can serve as a basis for thinking meaningfully
about the kinds of diagnoses that may be relevant in a particular case. As-
sessing the underlying dispositions of the client leads one to see the moti-
vational sources of behavior, which in tum can help clarify the etiology of
the presenting problem. Later chapters evaluate the relations between the
dimensions measured by the NEO PI-R and both DSM-IV (American Psy-
chiatric Association, 1994) Axis I and Axis II clinical phenomena. These
empirical relations can be very useful diagnostically.
The third value of personality assessment is empathy and rapport. The
insights into clients that are provided can help promote understanding of
and empathy with the client. A better understanding of the person's con-
flict, struggles, and aspirations helps to present a more nuanced portrait of
the issues that surround his or her seeking of treatment. Unlike a more
clinically focused scale and its emphasis on pathology, the NEO PI-R pro-
vides a more holistic, person-centered orientation that captures clients'
growth potential as well as their growing edges.
Feedback and insight represents another important value of personality
assessment. Providing clients with feedback on their NEOs can help pro-
vide a voice for the client to articulate his or her issues: the patient can feel
understood. Many times clients feel distress or conflict but do not have the
language to fully articulate their difficulties. Oftentimes, clients may feel
that their problem is something unique, not understood, and therefore not
treatable. Once clients receive feedback on the NEO PI-R and begin to see
how their own personalities are accurately captured, they have a sense of
 PERSONALITY AND ITS ASSESSMENT 13

hope. New words are learned that carry some of the nuances of their own
special psychological experiences.
This new language can then be shared with the therapist and a com-
mon language for working in therapy is then established. Presenting prob-
lems, clinical issues surrounding treatment goals, problems and issues,
and resistance (to name only a few) can all be framed in the dimensions of
the NEO PI-R. This language can also be used by clients to monitor their
own sense of change and growth throughout the therapeutic process.
These insights can readily be shared with the therapist, because both part-
ners have a common framework.
The NEO PI-R is one of a few instruments that provide a computer-
generated feedback report that can be given to clients. This report is pre-
sented in easily interpreted, jargon-free terms that give the client a broad
understanding of their personality. These reports are readily accepted by
clients and contain little objectionable material. One of the advantages of
giving feedback is that it can provide a springboard to meaningful dia-
logue between therapist and client.
Anticipating the course of therapy is the next value. Successful psychother-
apy depends not only on a therapist'S skill, but also on the client's coopera-
tion, motivation to work, and capacity for therapeutic change. No matter
what clinical school one works from, each client brings unique characteristics
to the session that need to be accommodated in treatment. The NEO PI-R can
provide important insights into these aspects of the individual.
For example, very low Agreeableness scores indicate a client who may
be skeptical about the entire therapeutic process and expect the clinician to
prove him- or herself at every tum. Such individuals may appear resistant
and mistrustful. Conversely, very high Agreeableness scores can indicate
an overly compliant client who may easily become dependent on the ther-
apist. The client may smile and nod in agreement but inside be completely
rejecting everything the therapist is saying. Or such a client may uncriti-
cally accept whatever the therapist says without critically judging its mer-
its for his or her particular situation. In such situations, the treatment fails
because the client does not struggle enough to tailor the intervention to his
or her own needs.
Because personalty scales provide information about adaptive styles,
the therapist can anticipate potential issues that may emerge in treatment
and take preemptive action. Scores on the various personality scales can
also be useful for identifying issues that were not part of the presenting
problem but may need to be explored.
Finally, matching treatments to clients is perhaps the most important
contribution personalty assessment can provide to practitioners. It has
14 CHAPTER 1

long been known that certain clients benefit more from certain types of
therapies than others. Personality measures can facilitate this matching
process. For example, individuals high on Extraversion, who are sociable
and talkative, will find therapies that require interpersonal interactions
more helpful than do introverts, who may prefer and benefit from behav-
ior therapy or Gestalt approaches. Individuals high on introspectiveness,
who are more willing to consider novel ideas and tryout unusual prob-
lem-solving approaches, may prefer Gestalt, psychoanalysis, or Jungian
analysis. Individuals who are closed to internal experiences will prefer di-
rective psychotherapies that offer sensible advice, behavioral techniques
that teach concrete skills, or client-centered therapies that provide emo-
tional support.
By obtaining high-quality information about the basic dispositions of
the client, therapists can think more precisely about the kinds of treat-
ments that would complement the client's style. Personality assessment
can help therapists better appreciate the uniqueness of clients and to think
flexibly about how to intervene. As clinical outcome research progresses,
we hope gains will be made in our ability to link specific treatments to par-
ticular types of individuals.
There is no doubt that measures of normal personality characteristics
provide a lot of "bang for the buck." Obtaining a comprehensive sampling
of personality traits can provide a pool of information that is as flexible in
its application as it is informative in its descriptions. The next section out-
lines some of the major models of personality constructs that dominate the
way personality psychologists conceptualize temperament. It also outlines
the problems these approaches create. The NEO PI-R is an outgrowth of
one of the solutions to these problems.

QUALITIES OF PERSONALITY

The value of personality assessment cannot be overstated. Capturing

the fundamental aspects of the client provides the therapist with a broad
understanding of the client and his or her psychosocial context. Symptoms
can then be seen as part of a larger ongoing life process. For example, an un-
derstanding of personality can inform clinicians about the etiology of the
presenting problem and about other potential weaknesses. The course of
therapy can be anticipated and client resources identified. The question that
arises is, "What are the personality variables that need to be assessed?"
Over the past century, numerous theorists have appeared and each
has presented his or her own model of personality. Each model not only in-
cludes insights into how personality is organized, but also identifies char-
 PERSONALITY AND ITS ASSESSMENT 15

acteristics that are the most salient for understanding underlying motiva-
tions. There is a rich literature in personality theory.
Modem scientific thought about personality finds a strong beginning
with the work of Sigmund Freud, who saw two major personality dimen-
sions: Eros (the life instinct) and Thanatos (the death instinct). For Freud
these two qualities were the basic sources of all personal motivations.
Every behavior or characteristic of an individual could be psychologically
parsed into some combination of these two forces.
Later theorists noted many limitations to Freud's analyses of tem-
perament and began to elaborate on other qualities of the person. Areview
of these theorists and their philosophical foundations is beyond the scope
of this book, but Table 1-2 outlines a few of the more modem major per-
sonality theorists whose ideas continue to influence current personality
theory and measurement. These theorists are merely a small sampling of
personality psychologists. They were selected because they represent a
spectrum of approaches to conceptualizing personality and also because
the instruments that are based on their approaches are multidimensional
inventories claiming to represent a comprehensive sampling of personal-
ity constructs. These theorists were selected largely because they see many
more qualities of the individual as important than Freud did.
The individuals listed in Table 1-2 represent various perspectives on
normal personality. Henry Murray provided a motivational approach,
while Cattell, a trait theorist, based his work heavily on quantitative meth-
ods. However, the reader should be aware that there are other personality
models that exist for explaining abnormal personality functioning. There
are the biosocial model of Millon (e.g., Millon, 1990), the psychobiological
model of Cloninger (e.g., Cloninger, 1987; Cloninger, Svrakic, & Przybeck,
1993), and a model semantically derived from the Diagnostic and Statistical
Manual of Mental Disorders itself (e.g., Harkness & McNulty, 1994). There
are numerous variables here that also need to be understood not only in re-
lation to each other but also with respect to those constructs that define
normal functioning.
Two issues emerge from Table 1-2. First, it is quite obvious that these
few theorists have identified a number of personality characteristics; there
are 80 constructs identified in the table alone. Yet, there are many other
psychologists who have identified more circumscribed, putatively impor-
tant constructs that define important aspects of personality and predict
salient life outcomes. Such constructs as cynicism, hostility, hopelessness, shy-
ness,fear of success, fear offailure, origenence, intellectence, critical parent, adult,
free child, nurturing parent, altruism, well-being, coping ability, burnout, empa-
thy, self-esteem, stress experience, and authoritarianism are but a few of liter-
ally thousands of such scales that occupy the psychological literature. The
16 CHAPTER 1

TABLE 1-2. Measurement Models of Personality and Their Constructs

Theorist Measurement device Constructs
Henry Murray T.A.T./ Adjective Achievement, dominance, abasement, suc-
Check List/Edwards corance, heterosexuality, play, exhibition,
Personal Preference nurturance, order, infavoidance, autonomy,
Schedule /Personality aggression, change, defendence, affiliation,
Research Form harmavoidance, counteraction, sentience,
understanding, rejection
Harrison California Personality Self-confidence, self-control, personal adjust-
Gough Inventory / Adjective ment, ideal self, creative personality, military
Check List leadership, masculinity, femininity, domi-
nance, capacity for status, SOciability, social
presence, self-acceptance, independence,
empathy, responsibility, socialization, self-
control, good impression, communality,
sense of well-being, tolerance, achievement
via conformance, achievement via indepen-
dence, intellectual efficiency, psychological
mindedness, flexibility
J. P. Guilford Guilford-Zimmerman General activity, restraint, ascendance, socia-
Temperament Survey bility, objectivity, friendliness, thoughtful-
ness, personal relations, masculinity
Raymond 16PF Warmth, reasoning, emotional stability, domi-
Cattell nance, liveliness, rule-consciousness, social
boldness, sensitivity, vigilance, abstracted-
ness, privateness, apprehension, openness
to change, self-reliance, perfectionism,
tension
Carl Jung Myers-Briggs Type Introversion-€xtraversion, thinking-feeling,
Indicator sensing-intuiting, judging-perceiving
Hans Eysenck Eysenck Personality Neuroticism, extraversion, psychoticism
Questionnaire

plethora of measures testifies to the richness of personality research and

underscores the interpretive and predictive value that personality holds
for the social sciences. However, the nearly limitless number of these scales
very easily boggles the mind.
This is the second issue that emerges from Table 1-2. Given all these
scales, how does one select instruments for use? Are all these constructs
needed to provide a comprehensive sampling of personality? If not, then
which ones should be used? As we scan the list of scales, it becomes clear
that there is much redundancy among the measures. Clearly Socialization,
Affiliation, Extraversion-Introversion, Personal Relations, and Privateness
 PERSONALITY AND ITS ASSESSMENT 17

all seem to have something in common. But one should never rely on scale
names to identify the personological content of the scale. It is entirely possi-
ble that two scales having the same name measure very different constructs.
Conversely, two scales having different names may in fact be measuring the
same quality. I give an example of this issue in the next chapter.
But here lies the real crux of the issue. There really is no established
framework for evaluating all these diverse scales or for linking them to-
gether conceptually. As things stand now, assessors need to select scales
that appear to them to be appropriate for their particular needs. However,
many of these scales and inventories proceed from very different theoreti-
cal traditions. For example, if one is interested in motivational needs, then
measures of Murray's constructs are appropriate. If one does not prefer
such a motivational perspective, then the more empirically based con-
structs offered by Cattell may be preferred. Then again, one may wish to
take the "folk" approach offered by Gough and use constructs he deems as
being important concepts for describing important interpersonal behaviors.
In each instance, though, the selection is one of theoretical preference.
Even though one's identification of a theoretical preference is based on a
serious consideration of both the heuristic value of the theory and the rel-
evant research data, unfortunately there is no larger, objective paradigm
for critically comparing and evaluating the merits of these many ap-
proaches. Therefore, the field is left in a chaotic state, in that there are many
different measurement models each with its own jargon and terminology
but with no conceptual or empirical linkage among them. When one sur-
veys the field of assessment instruments, it appears almost as a Tower of
Babel: so many constructs, so little understanding.
What is needed is a taxonomy of personality characteristics. Such a
framework would allow one to compare and contrast personality mea-
sures in terms of important, defining qualities. Ideally, this framework
would not work out of any particular theoretical tradition. Rather, it would
represent a more objective, preferably empirically based, system of classi-
fication. A taxonomic model would be useful in organizing the many per-
sonality constructs that exist and help in identifying areas of redundancy
and uniqueness. This would make the task of developing a comprehensive
assessment battery much easier.

WHAT IS A TAXONOMY?

A taxonomy is simply a framework for classifying things. As Bailey

(1994) noted, "classification is merely defined as the ordering of entities
into groups or classes on the basis of their similarity. Statistically speaking,
18 CHAPTER 1

we generally seek to minimize within-group variance, while maximizing

between-group variance" (p. 1). In order to accomplish this, one needs to
identify all the necessary qualities that distinguish the entities that are to
be classified and these distinguishing characteristics must be mutually
exclusive.
Taxonomies provide a better understanding of the things they clas-
sify by revealing similarities and differences among them. Taxonomies are
able to do this because, as already stated, they enumerate the most im-
portant defining qualities of the things to be classified. Further, these
qualities are not overlapping. Each dimension of the taxonomy provides
nonredundant information. As a result, things that are located closer to-
gether in the taxonomy have more things in common than elements lo-
cated farther apart.
Perhaps the best-known taxonomy is the one developed by Carl von
Linne (also known as Linnaeus) in the mid-eighteenth century. This model
is used to classify all living things. We are all aware, to some degree, of the
major categories of kingdom, phylum, class, order, family, genus, and
species. Every animal and plant receives a taxonomic classification based
on these eight dimensions. Creatures located close to one another in this
framework (e.g., in the same family), have more in common than creatures
that are not in the same family. The introduction of this framework was a
major step forward in our understanding of animal life, much of which we
take for granted today.
For example, think of whales. For centuries people thought that
whales were fish. After all, they swam in the ocean like fish, they looked
like fish, acted like fish, smelled like fish-fish! However, when Lin-
naeus's system of organizing physical characteristics of animals was in-
troduced we received quite an awakening. From his perspective whales
were much more related to human beings than to fish. Why? Because both
have similar characteristics, such as possessing mammary glands, lungs,
and bearing live young, whales and human beings were located very
close together in the taxonomic scheme (they share the class, mammalia).
Fish, on the other hand, do not have any of these physical characteristics
and therefore were placed very far away from both whales and human
beings (fish occupy several different classes: osteichthyes, chondrichthyes,
and agnatha).
From this discussion we can see that there are two important values of
a taxonomy. First, as already noted, a classification framework provides an
organization for understanding the elements that we wish to classify. Such
systematic description can provide new insights into already "familiar"
objects, as we saw with the case of whales. The second value of a taxo-
nomic structure is that it gives us a frame of reference for evaluating new
 PERSONALITY AND ITS ASSESSMENT 19

objects that may be discovered. Measuring new items in terms of the ex-
isting dimensions of the taxon allows one to determine whether the new
element represents something distinct from existing elements, or overlaps
them in specific ways. It is this "parsing" feature of taxonomies that can be
the most value for psychological assessment.

TAXONOMY VERSUS TYPOLOGY

Perhaps a more familiar term to social scientists is typology. A typology

is the flip side of a taxonomy. In a taxonomy one is interested in under-
standing some entity in terms of basic qualities or dimensions. With a ty-
pology, on the other hand, interest is in identifying the constellations of
basic qualities that co-occur. Thus a typology is usually multidimensional
in nature and represents a conceptual grouping of taxonomic elements.
For example, Jungian theory provides a good example of both con-
cepts. As noted in Table 1-2, Jung sees four major qualities that describe
people, Extraversion-Introversion, Thinking-Feeling, Sensing-Intuiting,
and Judging-Perceiving. These qualities would constitute a taxonomy. For
Jung, any behavior or motivation can be seen as some combination of these
qualities. However, the way one stands on all of these domains simultane-
ously constitutes one's type. The label INTJ, a type, represents a combina-
tion of qualities into a larger whole that has its own personological
implications. Using chemistry as an example, one can think of a taxonomy
as representing the elements, while a typology represents compounds.
Thus taxons and types represent important ways of organizing and
describing individuals. Their value is founded on the identification of im-
portant, discriminating constructs. The usefulness of taxons for personal-
ity theory and assessment has not been overlooked.

THE VALUE OF A TAXONOMY FOR PERSONALITY ASSESSMENT

Social scientists have also been very interested in developing tax-

onomies of personality functioning. Having such a model would help or-
ganize our understandings of how individuals come to be and the
qualities they display. It would enable us to distinguish a characteristic
representing a superficial quality, one that may be situationally deter-
mined, from a feature that may represent a more fundamental attribute,
stable over time and situations. To borrow a metaphor from Ozer and
Riese (1994), a personality taxonomy would be like the latitude and longi-
tude markings on a map. Just as explorers would locate their new landfalls
by reference to these calibrations on a globe, so, too, would personality
psychologists need to identify their new individual-difference variables
20 CHAPTER!

using the calibrations of the taxonomy. In this way, psychological phe-

nomena could be reliably located and competing claims identified.
Numerous social scientists have tried to establish taxonomies in order
to accomplish this important goal. As we noted earlier, Freud was perhaps
the first to specify some type of model, postulating the existence of Eros
and Thanatos. For him, these dimensions constituted the entire taxonomy.
But not all agreed with Freud. Jung proposed a separate set of dimensions
which, as already discussed, formed his typological model of personality.
As Table 1-2 reviewed, there are numerous personality models, each
with their own language and constructs of interest. However, there is no
objective manner here for selecting and interpreting psychological con-
structs. The theoretical basis of all these models makes any option one of
personal bias. The natural sciences always seemed to have a way to rise
above this subjectivity and to find a common reference point that all could
find acceptable and reasonable. For example, when early researchers were
in need of a temperature scale, the interval-level indices were in need of a
"zero" point. Where to put zero? It could be put anywhere, because of the
inherent subjectivity of an interval-level scale. Why not fix zero to where
water freezes? Why there? Why not? Water is a common element that all
are familiar with and, of course, like. In astronomy there is a need to talk
about distances among objects in the solar system. What metric would be
useful? What standard should be applied? The result, the astronomical unit,
which is the distance between the sun and earth. Why this unit? Again,
why not? The earth is a familiar place, one that all know about and feel
comfortable with. It is, after all, a reference to ourselves and what could be
more user friendly than that?
For our purposes, then, is there such a familiar standard that can be
applied in personality assessment? Is there a reference point or points that
would be meaningful to everyone? Given the plethora of personality vari-
ables and models that exist one may be t~mpted to think not. But where
would one begin in such a quest? Where is the common element? For Gor-
don Allport, that common basis was found in the English language.

THE LEXIGRAPHIC HYPOTHESIS

For the most part in personality psychology, the major resources for
identifying important dimensions of personality were the experiences and
insights of individual personality theorists. Although these theories have
important theoretical and clinical value, the fact remains that they are in-
herently subjective. Ellenberger (1970) has advanced the idea that many of
these influential theories find their basis in the "creative illnesses" of their
 PERSONALITY AND ITS ASSESSMENT 21

creators. In an attempt to make sense of their own social reality, theorists

become absorbed in finding answers to these very personal questions. The
theories that emerge, then, reflect the answers that seem to make the most
sense for the theorists.
If one does not find value or meaning in these "answers," then the
theory has little relevance. There is nothing compelling one to accept a the-
ory other than available data (which may be lacking) or a personal orien-
tation shared with the theorist. In either case, these theories are locked into
a particular set of life circumstances. Ultimately, something larger and
more compelling is needed for developing a catalogue of individual dif-
ference constructs.
In order to bypass this problem Gordon Allport took a different path.
Instead of necessarily evaluating his own personal experiences, he decided
to access the accumulated knowledge of our larger culture. The repository
of this wisdom, experience, and insight was the English language itself. To
understand the value of using language in this way, we need to evaluate
its function.
Language is an important tool that human beings have. From an evo-
lutionary point of view, there is no doubt that language serves an adaptive
function: Having language helps us to survive and reproduce. If we could
go back to primeval times and look at our evolutionary forebears, it would
be clear that those who could talk and communicate survived while those
who could not, died. Therefore speaking was selected "in" to our genetic
stock. But what do we talk about? For language to be effective, we would
need to talk about important aspects of our adaptive landscape. We would
need to create words that would help us to survive. Words that would di-
rect us away from danger and closer to food. Any object, event, or place
that would relate to our survival would certainly, over time, acquire a
word to represent it.
Take, for example, Eskimos. They have many different words for
snow. Why? Because snow is an important part of their adaptive world.
There is the type of snow to build one's igloo out of; a type of snow that is
good for traveling over, and so on. How many words do we have for
snow? One-snow. Why so few? Look around; how much snow do you
see? Other than a possible inconvenience at times, snow is a transitory
phenomenon for most of us in the United States. If you were to ask people
from, say, equatorial Africa, you would find that they may not have a word
for snow at all! Why? No snow.
As human beings we are also very much social animals. We live in
groups and depend on others for our survival. Because of our strong in-
terpersonal needs, it is clear that other people make up an important di-
mension of our own adaptive landscape. We need to distinguish the ones
22 CHAPTER 1

who will help us from the ones who will hurt us. We need to discover
whom can we depend on and whom we should select as a mate. In short,
to survive in the human community, one needs to be able to effectively an-
ticipate the motivations and behaviors of others. Because the personal
characteristics of others play a significant role in our own survival, it
stands to reason that over time our language has accrued terms that reflect
or describe these important aspects of people. Thus, the lexigraphic hy-
pothesis asserts: "Those individual differences that are most salient and
socially relevant in people's lives will eventually become encoded into
their language; the more important such a difference, the more likely is it
to become expressed as a single word. The analysis of the personality vo-
cabulary represented in a natural language should thus yield a finite set of
attributes that the people in the language community have generally
found to be the most important" (John, Angleitner, & Ostendorf, 1988,
p. 174). The view of language as a means for our species to codify impor-
tant dimensions of our environments is the central element of the lexi-
graphic hypothesis.
The rationale for the lexical hypothesis is relatively straightforward,
although there are three important limitations that need to be appreciated.
First, how words come to be developed and included in a lexicon is
not well understood. Further, words that do appear may not necessarily be
of importance scientifically. It is possible that some qualities of personal-
ity may not be apparent to lay observers and therefore overlooked in the
language. Also, some words that are developed may be more evaluative
and expressive than descriptive. Thus, language serves more than just a
descriptive function.
Second, the personality attributes that are defined in a lexicon are spe-
cific to the culture that generated them. Different cultures may emphasize
certain qualities over others, or may not exhibit various characteristics that
are apparent in other cultural contexts. Thus, word-based models may lack
generalizability.
Finally, although words do have a shared meaning within a culture,
there is a broad range of descriptive and explanatory convenience for any
given term. According to Allport (1961), words represent "ratbag cate-
gories," or broad ranges of qualities. For example, we can refer to two peo-
ple as being dominant, but that does not mean that their levels of
dominance are the same. Rather, the same term implies some level of gen-
eral agreement in temperament; how much is not known. Thus, the same
label may be applied to describe a wide array of behaviors. This lack of
precision makes it difficult to isolate the phenomena described.
These three issues raise important questions that need to be addressed
empirically if any lexically based taxonomy is to be found useful. For now,
 PERSONALITY AND ITS ASSESSMENT 23

these issues generate questions surrounding the robustness, inclusiveness,

and generalizability of any model based on language. These issues are con-
sidered again in Chapter 2. Despite these limitations, language does offer
the opportunity for identifying potentially important individual-differ-
ence variables. Although these "ratbag categories" do not possess the level
of precision demanded by science, they do provide a useful starting point.
Science can provide more rigorous definitions as the classification pro-
gresses.
Proceeding from this perspective, Allport and Odbert (1936) went to
Webster's New International Dictionary (1925) and extracted all terms that
were able to "distinguish the behavior of one human being from that of an-
other" (Allport & Odbert, 1936, p. 24). The resulting list included mostly
adjectives and participles, although some nouns and slang words were in-
cluded. This investigation netted some 17,953 different terms! This repre-
sented about 4.5% of the total number of words in the dictionary. The sheer
number of terms indicates just how important others are in our adaptive
environments. Our need to anticipate real interpersonal events resulted in
the development of a wide assortment of descriptive labels that certainly
capture the many nuances and subtleties of human functioning.
This list of words provides the basic fodder for evaluating personality
dispositions. Allport and Odbert (1936) organized these terms into four gen-
eral categories in order to facilitate classification. The first category included
"names that seemed to symbolize most clearly 'real' traits of personality.
They designate generalized and personalized determining tendencies-con-
sistent and stable modes of an individual's adjustment to his environment"
(Allport & Odbert, 1936, pp. 25-26). This category included terms such as
aggressive, introverted, and sociable. The second category contained terms
"descriptive of present activity, temporary states of mind, and mood" (p. 26)
and included terms such as abashed, gibbering, rejoicing, and frantic. The
third category was the longest of the four and contained characterological
evaluations, such as insignificant, acceptable, pretentious, and worthy. The
final category was a miscellaneous grouping of adjectives that had "possible
value in characterizing personality, even though they have no certain place
in the first three [categories]" (p. 27). This category included words such as
experienced, infantile, and undisciplined.
This fourfold breakdown of terms was a useful first step in classifica-
tion. However, the sheer volume of words is quite intimidating, especially
from sorting and interpretive perspectives. Nonetheless, Allport and
Odbert (1936) have given us an important first step in defining a universe
of qualities that are relevant for describing individuals. These terms do not
reflect any theoretical point of view, nor are they the working out of some
context-dependent appraisal. Rather, these words represent hundreds of
24 CHAPTER 1

years of interpersonal activity, the culmination of centuries of human ob-

servation and wisdom. These words represent our lexical heritage. Over-
all, not a bad place to start for learning about ourselves.
But are all these adjectives necessary? Certainly within these lists
there are many synonyms and antonyms. Not all these words need to be
considered. There are also many infrequently used terms, such as chrema-
tistic, hagiolatrous, stultiloquent, and thersitical. They may represent qual-
ities that are not as relevant today as they once were, or they may have
more common variants (such as greedy, saintly, foolish, and abusive). In
any case, the need is to somehow reduce this list into a more manageable
number of items. Ultimately, we need to distill these items down to their
fundamental dimensions, those basic, larger qualities that seem to capture
the essence of all these words. Identifying and naming such factors would
provide the basis for a taxonomic model of personality.

SEARCHING FOR A LINGUISTIC STRUCTURE

The process of finding an underlying structure for the adjective pool

identified by Allport and Odbert (1936) is the history of the development
of the five-factor model. Numerous individuals have been involved in this
process, and good historical accounts can be found in Digman (1996) and
John and colleagues (1988). What follows here is a basic, and necessarily
limited, overview of this ongoing story.
The first person really to take advantage of the Allport and Odbert
(1936) word list was Raymond Cattell. Although he had already been active
in analyzing personality ratings (e.g., Cattell, 1933), he was not the first do
so (see Guilford & Guilford, 1936; Webb, 1915), Cattell saw this word list as
an interesting place to identify salient personality descriptions systemati-
cally. Using a subset of 35 bipolar scales, he obtained ratings on 373 male
university students. Subjecting these ratings to the then emerging statistical
tool, factor analysis, he obtained 12 interpretable factors (Cattell, 1944). In
two additional studies using new samples of both men and women, Cattell
was able to replicate these factors and he incorporated them into his 16 PF
(Sixteen Personality Factors) Questionnaire (Cattell, 1947, 1948).
The dimensions of the 16PF (see Table 1-2) represented empirically de-
rived, cohesive patterns of intertrait relationships derived from peer rat-
ings. The value of this work is manifold. First, it was the first effort to
identify empirically how various personality traits are related in cornmon
usage. Rather than relying on rational interpretation, which may miss
some important dimensions or nuances in usage, the factor analytic ap-
proach provided a relatively unbiased effort at identifying an underlying
structure already in place. A second benefit is that it allowed for a more
 PERSONALITY AND ITS ASSESSMENT 25

precise evaluation of the replicability of these factor structures in new sam-

ples. Finally, it pioneered a new technology, factor analysis, that would
eventually prove exceedingly useful for condensing large amounts of in-
formation into more manageable clusters or factors.
Briefly stated, factor analysis is a data reduction tool (see Gorsuch,
1997, for a brief review). It examines the patterns of correlations among
items to determine whether there are "clumps" of variables that seem to
correlate more highly with each other than with other variables. These
"clumps" then come to be identified as factors. Factors are latent dimen-
sions (quantities that are not directly assessed) that are hypothesized to ex-
plain the interrelatedness of the variables. For example, descriptors such as
outgoing, talkative, gregarious, friendly, and happy may correlate more
strongly with each other than with other items, such as conservative, rigid,
anxious, and disciplined. These five adjectives would then form a factor that
we may wish to label "Extraversion." Extraverted people will be hypoth-
esized to be outgoing, talkative, and friendly. Thus, we may choose to talk
about a person's level of Extraversion rather than the five different de-
scriptors that define the factor. We have reduced five separate things into
a single dimension.
In Cattell's later work, much of his effort was focused on these 16 fac-
tors. It is interesting to note that although later research would show that
these 16 factors clearly fit into the more parsimonious five-factor model
(e.g., Boyle, 1989; Goldberg & Digman, 1994; Hofer, Horn, & Eber, 1997),
Cattell would continue to oppose this type of integration, referring to the
five-factor model as the "five big factors heresy" (Cattell, 1994, p. 8). Al-
though Cattell did not pursue any further reduction in the number of fac-
tors, others did.
Building on the work of Cattell was Donald Fiske, who used several
of the rating scales developed by Cattell. Factor analyzing results from
self-reports, peer ratings, and supervisor ratings, Fiske obtained five large
factors, labeled Social Adaptability, Conformity, Emotional Control, In-
quiring Intellect, and Confident Self-Expression. These five factors emerged
independently over the three sources of information (Fiske, 1949). Despite
the fact that this study was executed well and the results were quite robust,
these findings were largely ignored.
During the 1950s and early 1960s, several researchers continued to use
the adjective lists developed by Cattell in their studies on personality
structure (e.g., Thurstone, 1951; Tupes & Christal, 1961). In all of these
studies, a consistent five-factor solution emerged. In other words, when
the individual ratings were examined, the pattern of intercorrelations sug-
gested that the items created five different "clumps" or factors. Regardless
of where the information was coming from (e.g., peer rating versus self-
26 CHAPTER 1

report), five different factors would emerge, and the content of these fac-
tors remained very similar. Norman (1963) concluded from both his own
work and a review of the literature that these five factors seemed to con-
stitute an "adequate taxonomy of personality" (p. 582).
This was a pretty powerful conclusion from these data. Norman was
arguing that underlying the ratings on these personality variables were
only five dimensions, and no more. Many were skeptical that such a small
number of dimensions could adequately describe the entire realm of per-
sonality (e.g., Digman, 1996, p. 10). But, aside from a philosophical distrust
of only five factors, there was a very real limitation to the studies done by
Norman and..others. Simply put, they relied almost exclusively on the
scales identified and employed by Cattell in his early research: 35 bipolar
rating scales. Given that Allport and Odbert (1936) identified close to
18,000 adjectives, 35 bipolar ratings (70 adjectives) seemed hardly repre-
sentative of the original pool of items. Certainly there could be other fac-
tors residing among those remaining 17,883 adjectives.
The reason why such a small number of adjectives were selected in
the first place was that factor analysis is a very complicated and intricate
statistical tool. In the years before computers, individuals had to rely on
hand calculations. It was not uncommon for someone to apply for a re-
search grant to fund such an analysis and it was equally not uncommon
for such a process to take the efforts of dozens of individuals several
months to complete. Therefore, having only 35 to 50 items to factor analyze
was a major project. Today, with the easy availability of powerful, high-
speed computers and user-friendly statistical software, it may take a single
person about several moments to "do" a factor analysis. It is ironic that it
takes longer to print out the results of a factor analysis than it does for the
computer to actually do all the calculations. Nonetheless, the issue of the
representativeness of the original selection of adjectives was a valid criti-
cism and conceptual limitation.
In response to this, Norman (1963) went back to the original pool and
identified an additional 2,800 adjectives which he was able to rationally sort
into the five obtained factor dimensions. Picking up where Norman left off
was Lewis Goldberg who, starting in the early 1980s, began an impressive
program of research aimed at empirically evaluating the linguistic structure
of these adjectives. Over the next decade using large sets of ratings based on
an extensive pool of adjectives, Goldberg discovered that the five-factor
structure obtained in the early research continued to be observed in these
larger samplings of adjective ratings (Goldberg, 1981, 1982, 1990). By the be-
ginning of the 1990s, after nearly four decades and uncounted computer
time, it was becoming clear to most researchers that underlying the adjec-
tives identified originally by Allport and Odbert really were five large fac-
tors (Digman, 1990; Goldberg, 1993; McCrae & John, 1992).
 PERSONALITY AND ITS ASSESSMENT 27

These five factors have come to be labeled Neuroticism (or Emotional

Stability), representing the tendency to experience negative affect, such as
anxiety, depression, and hostility; Extraversion (or Surgency), which re-
flects the quantity and intensity of interpersonal interactions; Openness to
Experience (or Intellect, Culture), indicating the proactive seeking and ap-
preciation of new experiences; Agreeableness, reflecting the quality of one's
interpersonal interactions along a continuum from compassion to antago-
nism; and, finally, Conscientiousness (or Will to Achieve, Control, Con-
straint), which reflects the amount of persistence, organization, and
motivation to succeed in goal-directed endeavors. These five factors have
become known as the Five-Factor Model afPersonality and have been shown
to represent the underlying structure of the 17,953 adjectives originally de-
fined by Allport and Odbert. Whether one is working with self-report data
or observer ratings, whether one is assessing adolescents or adults, these
five factors seem to explain the patterns of correlations found among ad-
jectives in personality ratings.

FROM ADJECTIVES TO SENTENCES: THE NEO MODEL

The emergence of the five-factor model of personality represents a
true effort in the field of psychology to engage in programmatic, cumula-
tive research. In 9ther words, data are being obtained that build on previ-
ous findings and contribute to an ongoing informational structure. The
result of these efforts is a paradigm for understanding and evaluating per-
sonality constructs. However, the obtained five-factor structure derived
from the original Allport and Odbert (1936) adjective pool is but a first step
in understanding how individuals describe themselves and others. There
are some limitations to using single adjectives.
First, the number of adjectives found in the language for describing
each personality domain is not equal. Goldberg (1992) has shown that
there are more adjectives associated with Agreeableness than with any of
the other five domains. Extraversion is the next most frequently repre-
sented domain. As will be discussed later in this book, these two domains
together represent interpersonal styles: those characteristic ways that in-
dividuals approach others and the qualities of the relationships they form.
Given the highly social nature of humans, that these two domains would
gamer such a large number of adjectives makes sense. Conscientiousness,
Neuroticism, and Openness, in that order, are represented by fewer and
fewer adjectives. Goldberg did note that there are roughly equivalent
numbers of adjectives that describe each of the two poles for four of the
factors; there are extremely few adjectives marking low Neuroticism.
What are we to make of this uneven distribution of adjective descrip-
tors? Does this mean that some factors are psychologically more important
28 CHAPTER 1

than others? Or that some factors have been around human psychological
functioning longer than others, therefore having more of an opportunity to
emerge into more diverse synonym clusters? The answer to the second
question is certainly "No." All five factors have been shown to be psycho-
logically meaningful for understanding personality. Each domain is rela-
tively equal in size in terms of its capacity to explain a diverse range of
psychological phenomena. The answer to the third question is less clear
and more open to debate and discussion. As noted earlier when discussing
the lexigraphic approach, one drawback is that we do not understand how
words come to enter and leave the lexicon, or the factors that influence the
development nf new terms.
It would be interesting to perform lexigraphic analyses of the English
language as it developed over time. There are distinct periods of language
development (e.g., Old English, Middle English, Modern English); one
could determine the relative prevalence of adjectives representing the five
dimension at each stage. It may be that interpersonal behavior, a founda-
tion of human psychological and physical development and survival, is
most preeminent. As cultures develop, both socially and technologically,
new pressures are placed on humans to adapt and with this evolutionary
impetus come new terms to describe the emerging personality qualities
being expressed. One could formulate a number of hypotheses along this
line and the results of such investigations would be interesting testimoni-
als to human psychological development. But this takes us away from our
current theme. Suffice it to say that in working with adjectives, one needs
to recognize that the sheer volume of adjectives does not connote the sig-
nificance of a factor.
But the fourth question that emerges from the reality of an uneven dis-
tribution of adjectives over the five factors highlights one limitation to the
use of adjectives: "How many adjectives are necessary before one considers
a factor to exist?" Certainly, if there is a personality quality that has adap-
tive significance to humans, then many adjectives would be developed to
capture this quality and as a result, a factor would emerge representing
these numerous related terms. But there are replicable, smaller factors that
have been found beyond the "Big Five" that have not received much atten-
tion. Are these qualities as personologically trivial as their low numbers
suggest? The lexical approach does not establish any quantitative index for
determining whether a factor is large enough to warrant attention.
A second limitation to the use of simple adjectives is that there may be
qualities of people that are too complex to be reduced to a single word; they
may require several words, or a phrase or sentence in order to nuance
the quality. For example, Goldberg (1990) obtained smaller personality di-
mensions in his data when more than five factors were rotated, but they
 PERSONALITY AND ITS ASSESSMENT 29

represented very circumscribed characteristics, such as Thrift versus In-

temperance, and Sensuality versus Passionlessness. An additional repli-
cated factor obtained by Goldberg was Religiosity. It included items such as
Religious versus Nonreligious and Reverent versus Irreverent. What is in-
teresting about this small factor is that it represents a quality of people that
is very difficult to define using sentences, much less single adjectives only.
Imagine a very spiritual or religious person. What adjectives come to
mind? I can think of only a few, such as spiritual, religious, holy, transcendent,
mystical, and numinous. I would need to consult a thesaurus to find others.
Such a paucity of adjectives really underrepresents the importance of reli-
gion or spirituality in the lives of people. I have argued that spirituality
may represent a separate domain of personality worthy of study in its own
right (Piedmont, 1997a). However, the adjective record does not seem to
support such a claim. Therefore, adjectives themselves may be inadequate
to capture the full spectrum of psychologically significant personality
qualities. One may need to seek recourse in phrases and sentences. Also,
the use of phrases and sentences allows for greater subtlety and nuance in
defining personality attributes. Such added finesse may provide greater
predictive power.
A third limitation of adjectives is that they may have too much inter-
pretive variability, making assessment difficult and imprecise, not to men-
tion the technical psychometric problems such slippage introduces. For
example, take the word steady. What does it represent about someone? In
just evaluating this term, does it mean that someone is reliable, or emo-
tionally stable, or closed to experience? As it stands, there really is no way
to tell. If this term were to appear on a questionnaire, interpreting re-
sponses to the item would be subject to much error. Usually, to compensate
for this type of interpretive latitude bipolar terms are employed. This helps
to fix the interpretation of the term relative to some opposite. But again,
the term selected as the reference point can greatly influence the interpre-
tation. For example steady versus moody would represent qualities likely to
fall on the Neuroticism domain. Steady versus distractible would locate itself
on the Conscientiousness domain, while steady versus changeable may rep-
resent a quality relevant to Openness to Experience. Thus the same term
can have very different personological implications depending on the in-
terpretive context.
These three limitations to the lexical approach, (a) the lack of a clear
empirical criterion for determining the number of adjectives necessary to
define a significant personality factor, (b) the inadequacy of single terms to
capture important qualities of individuals, and (c) the wide interpretive
latitude that single terms possess, can all be circumvented by relying on
sentences and phrases for describing personality qualities. Qualities of
30 CHAPTER 1

interest can be defined theoretically and operationalized through multiple

sentences that provide a clear portrait of the dispositions. The personolog-
ical space containing this construct can be sharply defined and nuanced by
using phrases and sentences. Individuals can respond to very specific dis-
positional statements so that their responses can be unambiguously inter-
preted. Then the personality dimension can be evaluated for significance
by the extent to which it predicts important outcomes.
It was this approach that Costa and McCrae (1992c) used in develop-
ing their measure of the five-factor model-the NED Personality Inven-
tory (NED PI). Rather than relying on adjectives to broadly describe the
five factors, they constructed sentences that captured the subtleties of each
of the five domains. Further, the use of sentences enabled them to con-
struct facet scales for each of the domains. These facet scales are more pre-
cise articulations of the qualities subsumed by these five broad domains.
Each facet scale captures a more circumscribed psychological quality that
has been shown to have theoretical Significance in the field. In this way,
users can enjoy more interpretive precision as well as enhanced predictive
accuracy (Costa & McCrae, 1995a).
The NED model contains six facet scales for each of the five broad per-
sonality domains. For example, Neuroticism contains the facets of Anxiety,
Hostility, Depression, Self-Consciousness, Impulsiveness, and Vulnerabil-
ity to Stress. These facets were selected for two reasons: first, on the basis
of their historical and theoretical presence in the field; second, because
they have been shown empirically to represent a broad spectrum of quali-
ties within their given domain. These scales were created to be as nonover-
lapping as possible while still remaining on the same domain (Costa &
McCrae, 1992c; McCrae & Costa, 1992).
Although the presence of six facets per domain appears somewhat ar-
tificial, and it is, the NED PI-R authors recognize that there could be other
facets for a domain in addition to those already created. Dne example of
this concerns Neuroticism where a measure of somatic complaints was ex-
plicitly not created, although such a scale does appropriately belong on
that domain. The reason for this omission was simple: Because the NED
PI-R authors worked in a research context that would evaluate personal-
ity's role in predicting physical health, including such a measure would
contaminate their research. A measure of somatization would by necessity
include items relating to physical health and well-being. Such a measure
would certainly correlate with any physical functioning index or health
measure, not because personality was an important predictor, but because
the predictor and outcome would share similar items dealing with health
status. This would artificially inflate any associations between the two
variables, irrespective of any substantive overlap (see D. H. Schroeder &
 PERSONALITY AND ITS ASSESSMENT 31

Costa, 1984). Therefore, this type of scale was avoided. Certainly, future re-
search may be able to discern new specific personality constructs that are
relatively independent of the existing facet scales for a particular domain,
but yet belong in that personological space. This would argue for the in-
clusion of more facet scales.
Although the ultimate structure of the NEO model may not have been
settled on, the current instrument-the NEO PI-R-contains a large
amount of personological information: 30 separate facet scales combining
to assess the five major dimensions of personality. The NEO PI-R is
rapidly becoming one of the most popular measures of normal personality
in the research literature. Psychometric reviews of the instrument are uni-
formly favorable (e.g., Botwin, 1995; Juni, 1995; Piedmont, 1997b), citing
the robust empirical nature of the instrument as well as its impressive pre-
dictive validity in a number of applications (e.g., physical and psycholog-
ical health, job success, well-being, coping ability).
But given the empirical strength of the instrument, there is still a basic
question that needs to be asked: "With so many different personality
inventories available, why should one use the NEO PI-R?" There are sev-
eral responses to this query. First, as noted earlier, the five-factor model
of personality represents a comprehensive, empirically based taxonomy of
personality traits. The NEO PI-R is the only commercially available in-
strument designed explicitly to measure those dimensions. Although there
are a number of marker scales available in the general literature for as-
sessing these domains <e.g., Goldberg, 1992; John, 1990; Saucier, 1994), the
NEO PI-R is the only instrument that also provides more specific facet
scales within each domain. This allows for a more precise and differential
evaluation of a person's place within each personality dimension. For
example, the Extraversion domain has the facet scales of Warmth, Gregar-
iousness, Assertiveness, Activity, Excitement Seeking, and Positive Emo-
tions. It is possible for someone to score high on Extraversion, but possess
only the interpersonal component of the domain <e.g., Warmth, Gregari-
ousness, and Assertiveness) and none of the qualities relating to personal
energy or tempo <e.g., Activity, Excitement Seeking, and Positive Emo-
tions). Global scores miss these important distinctions.
Another reason for preferring the NEO PI-R is that the five-factor
model is rapidly becoming the preeminent measurement paradigm for
personality in a number of contexts, both clinical and applied (Barrick &
Mount, 1991; Costa, 1996; Goldberg, 1993). The empirical basis and con-
ceptual clarity afforded by the model makes it an ideal measure for talking
about individual differences that crosses subdisciplines in the social sci-
ences. Further, these major dimensions of personality also have been
shown useful for describing both normal and clinical populations.
 32 CHAPTER 1

Finally, given the taxonomic nature of the five-factor model, it can be

useful for linking information from diverse sources. Scores on the major
domains can speak about an individual's interpersonal style, coping abil-
ity, needs and motives, and response to psychotherapy. The five-factor
model can be useful for organizing all of this information into a cohesive
psychological portrait of the individual. Especially from a clinical per-
spective, the NEO PI-R can help streamline the assessment process and
promote a better understanding of the larger context within which the
client's presenting problems develop. As Miller (1991) noted, "the five-fac-
tor model can relate patient personality, presenting complaint, treatment
plan, and treatment outcome to each other in a reasonable, systematic way,
without loss of empathy or compassion for the patient" (p. 432).

RECOMMENDATIONS FOR ApPROACHING THIS BOOK

The purpose of this book is to provide you with a document that pre-
sents the technical and interpretive aspects of the NEO PI-R. The follow-
ing chapters provide basic psychometric information about the instrument
and interpretive guidelines for evaluating NEO PI-R profiles. Also cov-
ered are topics relating to the application of the instrument in a research
context. After reading this book I hope that you will gain an appreciation
for the potential value of the NEO PI-R for your own work. By becoming
familiar with the concepts of the five-factor model you will develop a lan-
guage for discussing individual differences that will enable you to dia-
logue with professionals from other assessment areas. The five-factor
model is the direction that personality assessment is moving toward and
in the future will serve as the basis for all conversations about personal dis-
positions.
To help you gain this expertise, many interpretive examples are pre-
sented for both individuals and couples. It is hoped that these materials
will illustrate the interpretive nuances represented by each of the scales.
Most important, efforts have been made fo perform profile interpretations
of the NEO PI-R. Profile interpretations rely on an understanding of how
several different scales combine to describe one's personality. Although
work examining specific NEO PI-R configural patterns is just beginning,
what is available demonstrates the exciting interpretive potential the in-
strument possesses.
As you read through the case histories you will begin to see how in-
terpretations are made. As you begin to develop a fluency with the termi-
nology and its application, try to interpret the case history profiles before
reading the evaluation. In this way you can compare your own insights
with those presented. It is certainly possible for you to see things that were
 PERSONALITY AND ITS ASSESSMENT 33

not included in the evaluation, so do not become concerned if some of

your insights are not reflected in the interpretation. Rather, see if you can
capture the most salient qualities in the profile.
Once you have completed this book, remember that there is no sub-
stitution for practice. It is recommended that you obtain the necessary
NEO materials from the publisher (Psychological Assessment Resources in
Lutz, FL) and take the NEO PI-R yourself. Compare your profile with
your self-concept with an eye toward discerning how your character be-
came expressed over the 35 scales. Then give the instrument to several oth-
ers that you know well and see how their personalities emerge. The more
you use the instrument, the better your interpretive skills will become. If
you plan to use the NEO PI-R in a clinical context, at first remember not
to let your interpretations of your client supersede good clinical judgment.
Complementing your use of the NEO PI-R should be a reading pro-
gram of the relevant research literature. The NEO PI-R is enjoying a wide-
spread use in research and there are literally hundreds of articles that have
already been published on the instrument. An extended bibliography of
these studies is available from the publisher when you purchase a manual.
Review these studies, which are sorted into different areas, and select
those that seem most germane to your own application. Staying familiar
with the current research will add to your interpretive skills. These studies
do much to extend the construct validity of the instrument, and provide
interpretive nuances to the scales. An examination of the bibliography at
the end of this book can point to the kinds of journals that contain these
types of articles.
For applying the NEO PI-R in a research context, the final chapter
highlights some of the techniques and procedures that can be used to max-
imize the instrument's interpretive and empirical utility. Despite a very
strong quantitative record, there is still much controversy surrounding
both the five-factor model in general and the NEO PI-R in particular.
Much of the issue concerns the number of factors necessary to describe
personality. Although this book tries to provide a conceptual framework
for approaching this question, it does not provide any definitive answers.
Rather, it outlines the questions that need to be asked and the kinds of re-
sults that are necessary to provide useful answers.
Another controversial area surrounds the factor structure of the NEO
PI-R in different populations (e.g., its cross-cultural generalizability). A
number of procedures have been used to evaluate factor comparability
over samples, each with differing levels of success. Although one can de-
bate the relative merits of each procedure (e.g., confirmatory versus ex-
ploratory factor analysis), the authors of the NEO PI-R have developed
their own technique for addressing this issue. That method is presented
34 CHAPTERl

here and is recommended for use. In this way, individuals wishing to join
the ongoing dialogue on the psychometric integrity of the NEO PI-R can
do so in a language that is the most relevant and persuasive for those who
already use the instrument.
Overall, this book should provide a useful first step to professionals
interested in developing a basic competence in the interpretation and ap-
plication of the NEO PI-R. Many of the materials presented here are pro-
vided in an attempt to facilitate your efforts at finding new ways of
applying both the instrument and the underlying measurement model.
Many of the interpretive strategies reflect preliminary results from a few
research studies. As such, you should be aware that most of the knowledge
about the NEO PI-R still needs to be revealed-this book should be con-
sidered a work in progress. Ultimately, it will be up to you, the reader, op-
erating in diverse contexts, to provide the additional information to help
fill out our understanding of this instrument and to document its utility to
our field.
 CHAPTER 2

PSYCHOMETRIC OVERVIEW
OF THE NEO PI-R

OUTLINE OF SCALES

The NEO PI-R consists of 240 items that clients answer on a (1) strongly dis-
agree to (5) strongly agree Likert-type scale and they are balanced to control
for the effects of acquiescence. The items are simple sentences describing
specific behaviors or attitudes. The NEO PI-R measures the five major do-
mains of personality and within each domain there are six facet scales that
are designed to assess more specific aspects of each domain. Alisting of the
facets is presented in Table 2-1.
In designing these 'facet scales every attempt was made to create
scales that were as nonredundant as possible while still assessing the
same overall dimension. This was done using a technique called validi-
max factoring (McCrae & Costa, 1989b). This procedure establishes the
structure of a scale by its pattern of correlations with external criteria. In
this manner, one can have confidence that information presented in one
facet is minimally redundant with information contained in other facets.
The facets for each domain were selected on the basis of their being psy-
chologically relevant and descriptively diverse (see Costa & McCrae,
1995).
Each facet scale is composed of eight items and domain scores are cal-
culated by summing scores over the six facet scales. As will be shown later,
the brevity of the scales does not compromise their reliability.
The NEO also has a rater version, Form R, which contains the same
items phrased in the third person. There are different versions for rating
men and women. Not many personality assessment instruments have this
35
36 CHAPTER 2

TABLE 2-1. Domains and Facets of the NED PI-R

Domain NEO-PIR facet Domain NEO-PIR facet
NEUROTICISM Anxiety AGREEABLENESS Trust
Hostility Straightforwardness
Depression Altruism
Self-consciousness Compliance
Impulsiveness Modesty
Vulnerability Tender-mindedness
EXTRAVERSION Warmth CONSCIENTIOUSNESS Competence
Gregariousness Order
Assertiveness Dutifulness
Activity Achievement
Excitement seeking Self-discipline
Positive emotions Deliberation
OPENNESS TO Fantasy
EXPERIENCE Aesthetics
Feelings
Actions
Ideas
Values

capability; the NEO PI-R's provision of a rater form has several real ad-
vantages. One advantage is the ability to gather five-factor model infor-
mation on an individual when a self-report cannot be trusted. Second,
rater data can be used to provide another perspective to the assessment of
a client. Finally, rater forms enable one to evaluate groups and their inter-
personal dynamics. These issues are discussed later in this book.

RELIABILITY

Table 2-2 presents internal consistency and test-retest information for

the domain and facet scales (see Costa & McCrae, 1988b, 1992c). As can be
seen, the alpha reliabilities for the facets range from .56 for Tender-mind-
edness to .81 for Depression, and for the Domains from .86 for Agreeable-
ness to .92 for Neuroticism. Considering that there are only eight items for
each facet, these alphas are all acceptable. Further, these values tell us that
the facets and the domains are quite unidimensional. From an interpretive
perspective, unidimensionality is an important quality. It tells us that high
or low scores represent a single, clear construct.
Column 2 in Table 2-2 provides test-retest reliabilities. It is important
to note that the large majority of test-retest reliability studies use a time in-
terval from 1 week to 6 months between tests. Such relatively small time
intervals are frequent because of both the logistics involved in following
people over time and the fact that the longer the interval is between test-
 PSYCHOMETRIC OVERVIEW 37

TABLE 2-2 Reliability Coefficients for the NEe-PI in an Adult, Mixed-sex Sample
NEG-PI Scale Alpha reliability" Retest reliability"
NEUROTICISM .92 .87
Anxiety .78 .75
Hostility .75 .74
Depression .81 .70
Self-consciousness .68 .79
Impulsiveness .70 .70
Vulnerability .77 .73
EXTRAVERSION .89 .82
Warmth .73 .72
Gregariousness .72 .73
Assertiveness .77 .79
Activity .63 .75
Excitement seeking .65 .73
Positive emotions .73 .73
OPENNESS TO EXPERIENCE .87 .83
Fantasy .76 .73
Aesthetics .76 .79
Feelings .66 .68
Actions .58 .70
Ideas .80 .79
Values .67 .71
AGREEABLENESS .86 .63'
Trust .79
Straightforwardness .71
Altruism .75
Compliance .59
Modesty .67
Tender-mindedness .56
CONSCIENTIOUSNESS .90 79'
Competence .67
Order .66
Dutifulness .62
Achievement .67
Self-discipline .75
Deliberation .71
'N= 1,539.
°
'Six-year stability coefficients for N, E, and domains and facets, 3-year coefficients for A and C domains,
all data from NEO-PI. When adjusted for attenuation, all coefficients are greater than .90. Ns range from
63 to 127 for various subsampJes of N, E, 0, A, and C.
'Retest coefficients for A and C are from NEO-PI original domain scales.

ings, the lower the retest correlations. The passage of time attenuates the
ability of a scale to correlate over two administrations. All things being
equal, the minimum acceptable retest reliability values are between .5 and
.6. Values between .6 and .7 are considered to be very good, between .7 to
.8 are excellent, and above .9 are impreSSive.
38 CHAPTER 2

The values presented in Table 2-2 are not 6-week coefficients nor are
they 6-month coefficients rather, they are 6-year values. As can be seen,
most are in the .7 to .8 range, suggesting very good stability over time. It
should also be pointed out that these values were obtained from data per-
taining to the NEO PI, the previous version of the NEO PI-R. The facet
scales for Neuroticism, Extraversion, and Openness are essentially the
same as in the current NEO PI-R. The NEO PI did not have facet scales for
the Agreeableness and Conscientiousness domains; scores for these two
domains are based on global I8-item scales. Data for these facets are still
forthcoming. In any event, adjusting these correlations for attenuation re-
sults in reliability estimates over .90!
What these data say is that there is impressive stability in these per-
sonality domains over time for adults (i.e., individuals over age 30). Costa
and McCrae (1994) provide longitudinal data on these dimensions that in-
dicate impressive stability, both in terms of rank order and mean level: 25-
year retest coefficients for these five major personality dimensions show
about 80% of the variance as stable, while 50-year estimates indicate about
60% of the variance remaining stable. Based on this great stability, Costa
and McCrae believe that after 30 years of age personality is "set like plas-
ter." From this point forward, an individual's personality has crystallized
and, all things being equal, will not change. A person begins now to select
his or her own environments and creates a direction for his or her life to
follow. Looking into this finding certainly raises a number of important
issues.
First, it seems almost counterintuitive to believe that personality is un-
changing in adulthood, suggesting that people are "locked" into their
lives. Any adult can review the past decade of life and see many areas of
personal development that have reconfigured his or her "personality."
How, then, can personality be destiny?
To answer this question, we need to revisit the concepts of genotype
and phenotype. Genotype refers to the basic underlying composition of
our nature. It is a trajectory through time, a pathway that our development
will follow. Phenotype, on the other hand, refers to the expression of the
genotype at any given moment in time. Phenotype refers to a dynamic and
lawful progression of change over time. One's underlying genotype re-
mains constant, but its expression over time will change. For example, Ex-
traversion represents a genotypic quality, but what Extraversion means, or
how it is expressed, will change over the course of one's lifetime. Being an
extravert in college may mean going to parties and staying out late, dating
many people, and being loud and boisterous. However, extraversion in
middle adulthood is something very different. Instead of going to many
parties and being loud, it could be hosting formal sit-down events, vaca-
 PSYCHOMETRIC OVERVIEW 39

tioning with friends, spending time with family, selecting an occupation

that brings one into contact with others, such as sales. Certainly the out-
ward features of our lives change from college to middle age, but the fun-
damental needs that we seek to have gratified remain constant.
The dimensions of the NEO PI-R capture genotypic qualities (McCrae
& Costa, 1996). Because of this, they should not vary over time and the
data in Table 2-2 support this contention. However, there is a large body of
evidence that documents the robustness of personality in adulthood.
Using subjects who were part of an ongoing longitudinal study, Costa and
McCrae (1989a) asked participants to rate the degree to which their whole
personality had changed over the previous 6 years. Three groups emerged,
one of those who believed that their personality had stayed pretty much
the same, another of those who believed that they had changed "a little,"
and a final group who believed that they had changed a great deal. Costa
and McCrae then compared the retest correlations between the NEO
scores of these individuals taken 6 years prior and currently for the three
groups. There were no significant differences among the groups in terms
of stability scores. The median uncorrected retest correlation was .80 for
the group that believed they had changed a great deal. Even for those who
claimed to have changed, and there is no reason to believe that these peo-
ples' outward lives did not change, their underlying dispositions re-
mained constant.
McCrae (1993) more formally evaluated the role of several factors
(personal agency, self-monitoring, private self-consciousness, and Open-
ness to Experience) as moderators for personality change. No consistent ef-
fects were found; any noted changes were due to measurement error.
Costa, McCrae, and Zonderman (1987) also failed to find any moderating
effects on personality by environmental variables, such as marital status,
employment status, or state of residence. Perhaps the most compelling ev-
idence for treating the dimensions of the NEO PI-R as genotypes comes
from the behavioral genetics literature which has documented the heri-
tability of these dimensions (Bergeman et al., 1993; Heath, Neale, Kessler,
Eaves, & Kendler, 1992; Viken, Rose, Kaprio, & Koskenvuo, 1994). Thus,
the dimensions of the five-factor model should not be considered mere
summary descriptions of behavior, but temperamental dispositions of in-
dividuals to think, act, and feel in consistent ways over time.
It should be pointed out that until adulthood, much of personality re-
mains fluid. Evidence suggests that perhaps up to 50% of the variance in
personality may be shifting throughout adolescence and early adulthood
(Block, 1971; Carmichael & McGue, 1994; Costa & McCrae, 1992d; Haan,
Millsap, & Hartka, 1986; Siegler et ai., 1990). However, even during adult-
hood there may be possibilities for change in personality. The onset of a
 40 CHAPTER 2

psychiatric disorder can have an impact on personality scores, and re-

search has shown that scores do change between periods of depression
and remission (Hirschfeld et al., 1983). However, whether this represents a
true change in personality or a distortion of the assessment process needs
to be determined. Another possibility for change concerns the experience
of catastrophic stressors, which may alter the way that one experiences the
world. Similarly, religious conversion may be another event that has the
potential for reshaping temperament. However, these are open questions
that are in need of empirical research.
A final possibility for finding personality change concerns psy-
chotherapy. The stability coefficients noted previously really raise ques-
tions over the intent of therapy and its effect on the individual. Should
clinicians intend to affect clients by helping them make shifts in their fun-
damental psychological structures or through improving their psychoso-
cial instrumentality (i.e., coping ability)? There are data that address this
issue. Several studies have been done evaluating the retest stability of
NEO PI and NED PI-R among clinical samples. All the studies found high
retest correlations pre- and posttreatment; however, there were significant
changes in mean level noted for the dimensions (Bagby, Joffe, Parker,
Kalemba, & Harkness, 1995; Piedmont & Ciarrocchi, in press; Trull, Useda,
Costa, & McCrae, 1995). Although the extent and magnitude of the ob-
served changes varied over the studies, the presence of shifts in scores
opens the possibility that those seeking change may be able to find it.
With these caveats noted, the reliability coefficients presented in Table
2-2 offer some important advantages to test users. First, because the NED
PI-R scales are relatively unidimensional, obtained scores represent clear,
unambiguous qualities of the individual. The 35 scales of the NED provide
a rich interpretive portrait of a person. The temporal stability of these
scales shows that this picture, in adulthood, will not change much over
time. Thus, the test interpreter can anticipate the kinds of life directions a
respondent will move toward and the types of outcomes likely to be expe-
rienced.

FACTOR STRUCTURE

The 30 facet scales of the NEO PI-R constitute five independent per-
sonality dimensions. The six facets for each domain were selected so that
there would be a sufficient number of defining variables to reliably evi-
dence each of the domains in later factor analyses of the instrument (Costa
& McCrae, 1995a). Although having six factors for each domain appears
very "neat" and artificial, there really is nothing magical about the num-
ber. There are certainly other aspects of personality that arguably could be
 PSYCHOMETRIC OVERVIEW 41

considered adequate markers or facets. Nonetheless, the chosen facets

were selected by the NEC PI-R's authors because of their psychological
significance in the literature.
Perhaps one of the most recurring issues about the NEO PI-R is its
factor structure and the orthogonality of the domains (e.g., J. Block, 1995;
Borkenau & Ostendorf, 1990). The value of independent dimensions is that
this structure ensures that information collected from one domain is not
redundant with information obtained from another. It would serve no pur-
pose to create 30 scales that assess the same quality. Many scales measur-
ing the same thing, even if from different vantage points, do not provide
an adequate sampling of personological content. However, a set of non-
overlapping dimensions helps to guarantee that a more varied and com-
prehensive evaluation is made.
The dimensions of the NEC PI-R are hypothesized to represent latent,
uncorrelated factors. However, factor analyses of the scale have consis-
tently demonstrated that some of the facet scales have significant (i.e.,
> .30) secondary loadings (Costa & McCrae, 1992c; Piedmont, 1994), which
results in some correlational overlap. This came about as a result of how
the NEO PI-R was developed.
Rather than relying on a conventional varimax rotation to identify fac-
tors (this identifies factors by obtaining simple structure among the items),
Costa and McCrae developed what they referred to as validimax rotation
(McCrae & Costa, 1989b). This technique instead identifies factors by their
pattern of correlations with external criteria of convergent and discrimi-
nant validity (see Costa & McCrae, in press; McCrae & Costa, 1987, for an
overview of this process). It was in this way that facets for each dimension
were created that were as different from one another as possible while still
remaining on the same factor. While external validity was maximized,
some of the internal simple structure was compromised.
This was not a problem for Costa and McCrae, who believed that the
developed facets represented important psychological qualities that
should be included in any assessment protocol, simple structure or not
(Costa & McCrae, in press). To delete some facet scales in order to preserve
simple structure would have compromised the substantive utility of the
instrument.
Several of the scales on the NEO PI-R have significant secondary
loadings. For example, the facet scales of Impulsiveness and Vulnerability
to Stress load negatively on Conscientiousness; Openness to Feelings loads
positively on Extraversion; Warmth, low Assertiveness, and low Excite-
ment Seeking load on Agreeableness. All of these facets represent impor-
tant personological characteristics of individuals, and to delete them
because of their secondary loading would be counterproductive. Yet, in
42 CHAPTER 2

other ways, these secondary patterns do make some intuitive sense. Indi-
viduals who are in control of their impulses and cope well with stress are
also able to compete and succeed. Individuals who experience a high de-
gree of positive affect are open to their inner, emotional worlds. People
who are warm and nonthreatening also appear very agreeable and com-
passionate.
Despite this conceptual overlap, it should be noted that the magni-
tude of these secondary loadings creates correlations that are relatively
small (averaging about .20). Thus, measures obtained from each domain
can be considered, for the most part, quite independent. However, truly
orthogonal domain scores can be calculated from factor score weights pre-
sented in the manual. These weights are applied to each of the 30 facets
and result in domain scores that are mutually independent.
For all practical purposes, the NED PI-R provides a factor structure
that it is quite rob~st. Minor overlaps notwithstanding, these five dimen-
sions have consistently emerged over instruments, raters, and even cul-
tures (McCrae & Costa, 1987; Piedmont, 1994; Piedmont & Chae, 1997).
Users of the NED PI-R can be confident that the information obtained
from the five dimensions represents a relatively comprehensive selection
of personological material.

THE QUESTION OF COMPREHENSIVENESS

As noted in Chapter 1, the five factors emerged from an analysis of the

English language. It could be argued that the five-factor model represents
only the ways in which everyday people think about and discuss others.
Thus, there is no reason to believe that this lay conceptualization would
have anything in common with the ways that psychologists construe
human characteristics. For example, psychodynamic models of personal-
ity find the source of human behavior in unconscious drives centered
around sexual and aggressive content. Eros and Thanatos represent such
constructs, as do various ego needs that reflect generalized affective forces
that drive, direct, and select behavior. Individuals are hypothesized to be
unaware of personal qualities such as penis envy, oedipal drives, or the
need for infavoidance; it seems unlikely that such constructs would be-
come encoded in the common language. Thus, it is possible that the types
of constructs that social scientists have developed to describe and explain
behavior have nothing in common with those dimensions found in the
common lexicon. If this is indeed the case, then any comprehensive model
of personality would require more than five factors; the five found in the
 PSYCHOMETRIC OVERVIEW 43

language plus any additional factors that have been discovered by the so-
cial sciences.

CORRESPONDENCE BETWEEN THE NED PI-R

AND OTHER MEASUREMENT MODELS

In order to evaluate the true comprehensiveness of the five-factor

model, a number of studies were conducted that evaluated the five-factor
model in relation to measures of psychological constructs representing di-
verse theoretical orientations. Using most often a joint factor analytic par-
adigm, these studies sought to determine the number of factors necessary
to explain the majority of common variance among these measures. Table
2-3 presents an overview of such studies that have used major personality
instruments. What has emerged from this very large and extensive litera-
ture is that only five factors are necessary for explaining the lion's share of
common variance in standard personality assessment inventories; those
factors are Neuroticism, Extraversion, Openness, Agreeableness, and Con-
scientiousness.
As can be seen in Table 2-3, the list of instruments that have been
evaluated includes Murray's needs, Gough's Folk Concepts, Jungian Ty-
pologies, Interpersonal Behavior, vocational interests, and even psy-
chopathological dynamics. These five dimensions are present, to varying
degrees, among these theoretically diverse instruments. This large and
growing literature documents the comprehensiveness of the five-factor
model. Despite its seemingly humble origins in natural language, these
five personality dimensions also permeate scientific conceptualizations of
personality. As can also be noted in Table 2-3, some of these findings have
been cross-culturally replicated using the NEO PI-R (e.g., Chae, Pied-
mont, Estedt, & Wicks, 1995; Heaven, Connors, & Stones, 1994; Levin &
Montag, 1991), and other measures of the model (e.g., Paunonen, Jackson,
Trzebinski, & Forsterling, 1992). There are also numerous articles that
have evaluated the relations between the five-factor model and scales de-
signed to capture more circumscribed personality constructs, such as pos-
itive and negative affect, performance motivation, and health-related
variables such as anger control and locus of control. Research here also
demonstrates the comprehensiveness of the five-factor model: All these
smaller constructs fit very nicely into the model (e.g., Costa & McCrae,
1980a, 1984; Marshall, Wortman, Vickers, Kusulas, & Hervig, 1994; Pied-
mont, 1995).
Table 2-3 also shows that some of these major personality instruments
do not capture all aspects of the five-factor model. Their theoretical and sci-
44 CHAPTER 2

TABLE 2-3. Bibliography of Joint Analyses Using the NEO PI-R

Instrument Study Findings
MMPI Factor Scales Costa, Busch, Zonderman, & Conscientiousness not
McCrae (1986) present
MMPIItems Costa, Zonderman, McCrae, Conscientiousness not
Williams (1985) present
MMPI-PD scales Trull (1992) Openness not well
represented
Personality Disorder Trull (1992) All five factors recovered
Scale-R
Personality Research Costa & McCrae (1988b) All five factors recovered
Form
Adjective Check List Piedmont, McCrae, & Costa All five factors recovered
(1991)
Edwards Personal Piedmont, McCrae, & Costa All five factors recovered
Preference Schedule (1992)
California Psychological McCrae, Costa, & Piedmont Agreeableness
Inventory (1993); Deniston & underrepresented
Ramanaiah (1993)
Wiggins's Interpersonal McCrae & Costa (1989c) Extraversion and
Adjective Scales agreeableness
Holtzman Inkblot Costa & McCrae (1986) No personality dimen-
sionsfound
Eysenck Personality McCrae & Costa (1985a) No Openness found
Inventory
Eysenck Personality Costa & McCrae (1995b) No Openness found
Profiler and EPQ-R
Cloniger's Tridimensional Costa & McCrae (in press) Agreeableness weakly
Personality Ques- represented, scales
tionnaire combine elements of
the other four domain
Multidimensional Piedmont (1994) All five factors recovered
Personality Ques-
tionnaire
Self-Directed Search Costa, McCrae, & Holland Neuroticism not well
(1984) represented
Myers-Briggs Type McCrae & Costa (1989a); Neuroticism not found
Indicator Furnham (1994)
Myers-Briggs Type Chae, Piedmont, Estadt, & Neuroticism not found
Indicator (Korean) Wicks (1995)
California Q-Set McCrae, Costa, & Busch All five factors recovered
(1986); Lanning (1994)
 PSYCHOMETRIC OVERVIEW 45

TABLE 2-3. (Continued)

Instrument Study Findings
Comrey Personality Scales Boyle (1989) CPS & 16PF evidence the
/16PF /Eysenck five major domains EPI
Personality Inventory corresponds to N & E
Personality Psycho- Trull, Useda, Costa, & All five factors recovered
pathology 5 McCrae (1995)
Basic Personality Levin & Montag (1991) All five factors recovered
Inventory (Israeli sample)
MMPI PD scales/Person- Wiggins & Pincus (1989) All five factors needed to
ality Adjective Check capture range of per-
List/Interpersonal sonality disorders
Adjective Scales
Guilford-Zimmerman Costa & McCrae (1985); N & E well represented,
Temperament Survey McCrae (1989) 0, A, & C weakly
present
MCMI I & II Costa & McCrae (1990) All five factors recovered
Personality Assessment Levin & Montag (1994) All five factors recovered
Inventory
Holden Psychological Holden (1992) All five factors repre-
Screening Inventory sented
Interpersonal Style Lorr, Youniss, & Kluth All five factors recovered,
Inventory (1992); McCrae & Costa but Openness only
(1994) weakly
16PF Gerbing & Tuley (1991); All five factors found
Hofer, Hom, & Eber
(1997)
Comrey Personality Scales Hahn & Comrey (1994) Openness not represented

entific pedigree is no guarantee of comprehensiveness. For example, re-

search with the Myers-Briggs Type Indicator (MBTI) shows that it lacks
any indicator of Neuroticism (Chae et al., 1995; McCrae & Costa, 1989a).
This may partially explain the popularity of this instrument: It does not
provide any potentially disturbing feedback to individuals about their lev-
els of negative affect and emotional dysphoria. The California Psycholog-
ical Inventory does not have any strong measure of Agreeableness. This is
curious because the instrument is based on a folk conception of personal-
ity (H. G. Gough, 1987). Folk concepts are "the kind of everyday variables
that ordinary people use in their daily lives to understand, classify, and
predict their behavior and that of others" (p. 1). This approach is similar to
the lexical approach that spawned the five-factor model. It is surprising
46 CHAPTER 2

and important to note that this omnibus test does not provide adequate
coverage of this personality dimension.
It is also interesting to note that attitudinal qualities, such as voca-
tional interests, are also related to the five-factor model. That the types of
jobs that interest us are related to personality underscores the contention
that the dimensions of the five-factor model represent genotypic qualities
that establish trajectories that guide the general directions and patterns of
our lives. This complements a growing body of research that illustrates the
usefulness of these large personality dimensions for predicting job success
and job satisfaction (Barrick & Mount, 1991; Mount, Barrick, & Strauss,
1994; Piedmont, 1993; Piedmont & Weinstein, 1994).
Recognizing the empirical overlap between scientific conceptions of
personality and the dimensions of the five-factor model provides two im-
portant pieces of information. First, we can have increased confidence that
the five-factor model is indeed a comprehensive description of individual-
difference variables. Second, given its comprehensiveness, this modei pro-
vides an opportunity for better understanding the personological content
of any psychological measure. By examining a construct's relations with
these dimensions (which, as shown previously, are clear, univocal quanti-
ties), we can derive a better understanding of the kinds of dispositions and
motivations reflected in scores from the scale. Further, we can also deter-
mine similarities and differences among scales, regardless of their theoret-
ical origin or label, by comparing their patterns of correlations with these
same five dimensions. We now turn to this major value of the five-factor
model.

WHAT THE NED PI-R CAN TELL Us ABOUT OTHER SCALES

It can certainly be said without hyperbole that thousands of person-
ality variables are being assessed by hundreds of assessment tools. The
field of personality is a veritable cornucopia of constructs, measuring di-
verse individual differences. To a novice, this diversity can be confusing,
and even to professionals, this landscape appears cluttered and disjointed.
With so many constructs proposed by the varied theoretical orientations in
the field, it is a challenge to discern any coherence among these variables.
Despite a plethora of scales, one cannot be certain that a comprehensive
description of personality can be made, or that all relevant aspects of per-
sonality have been adequately sampled.
The ability to assemble a collage of variables that one can be sure are
minimally redundant and comprehensively descriptive underscores the
value of the five-factor taxonomy. In his critique of the five-factor model,].
Block (1995) made reference to the "jingle fallacy" and the "jangle fallacy."
 PSYCHOMETRIC OVERVIEW 47

These terms refer to, respectively, the tendency to see scales as being simi-
lar, or different, on the basis of their label rather than on any empirical ev-
idence. The former sees convergence where none may exist, and the latter
allows useless redundancy to develop. Perhaps the least useful place to
look for a scale's meaning is its name. But these "fallacies" too often char-
acterize the field of assessment. Without any useful, larger framework
within which to evaluate scales, there is no way to disentangle the person-
ological content of scales.
An example of this arose in my own work during graduate school.
During my first years of school, I was interested in evaluating those per-
sonality variables that predicted performance on cognitive tasks. My first
set of variables included measures of anxiety and achievement motivation
(see Piedmont, 1988b). In selecting measures of achievement, I was very
much interested in using Murray's need for achievement. Being the in-
dustrious graduate student I was, I wanted to make sure that I was doing
good science by ensuring that I had multiple measures of my constructs.
Thus, I used two need achievement scales: one from the Edwards Personal
Preference Schedule (EPPS; Edwards, 1959) and another from the Adjec-
tive Check List (ACL; H. B. Gough & Heilbrun, 1980).
These scales were engaging because they were both established mea-
sures of Murray's need-press theory. Working from Murray's earlier work,
Edwards constructed carefully worded definitions for each need scale and
then selected items that clearly reflected those dimensions. H. B. Gough
and Heilbrun (1980) were also interested in developing measures of Mur-
ray's needs and they used Edwards's original definitions and items as
guidelines for selecting their adjectives. In short, these two scales share not
only a common theoretical pedigree, but are methodological cousins as
well. If there were any two scales that I would expect to be related mea-
sures of the same construct, these were the two. Thus, once my data were
collected, my first "validity check" was to correlate these two scales to
make sure that I found the high positive correlation that putatively should
be there. When I performed the correlation, I was horrified to find that
these two scales had a correlation of "0.00"! After several attempts at
rechecking my scoring, data entry, and computer commands, I realized
that this value was accurate. A review of the literature showed why.
As early as 1968, researchers had noted that these two scales were not
related and that they should not be considered equivalent measures
(Bouchard, 1968; Megargee & Parker, 1968). Yet despite this glaring lack of
construct validity, professionals continued to use widely both of these in-
struments. In fact, the ACL and the EPPS were listed by Buros (1978)
among the top 10 tests cited in the research literature, and the EPPS was
one of the most popular tests used in both clinical (Lubin, Larsen, &
48 CHAPTER 2

Matarazzo, 1984) and counseling settings (Watkins & Campbell, 1989). De-
spite the empirical evidence and calls to the contrary (e.g., Entwisle, 1972),
both researchers and clinicians widely employed these measures and rou-
tinely interpreted them in ways consistent with their putative definitions.
There are many reasons for this, but perhaps the most salient for me
was that researchers at the time had no way to resolve this issue. Empiri-
cally both these scales predicted achievement-related outcomes, but they
failed the test of convergent validity. One or both of these scales was in-
valid. Yet there was no independent, empirical criterion to which an appeal
could be made in order to determine which one was the culprit. Although
rational expla_nations were possible (e.g., Piedmont, DiPlacido, & Keller,
1989), they lacked the kind of persuasiveness that only data can provide.
The five-factor model offers such an empirical framework for person-
ologically parsing scales. The dimensions of the model are clear and struc-
turally robust. As the data in the previous section have shown, the model
is conceptually comprehensive, capable of organizing information from di-
verse theoretical orientations. These empirical realities provide a useful
medium for evaluating scales. Correlations between a scale and these five
factors can help to conceptually locate the measure within this well-
defined space. Second, the pattern of correlations can serve as the scale's
"fingerprint," so to speak, its unique blend of personological material.
From here, the similarity of the scale to other measures can be determined
by an inspection of each construct's pattern of correlates. Two scales can be
said to be related to the degree to which they share common correlations to
the five factors. Separate patterns of five-factor correlates would be evi-
dence of discriminant validity.
Locating scales within the five-factor space provides an opportunity
for overcoming the jingle and jangle fallacies by helping to identify areas
of overlap among scales as well as areas that have been overlooked. Given
the fact that some dimensions are better represented in the natural lan-
guage than others (i.e., Agreeableness is most represented, Openness is the
least; Goldberg, 1990, 1992; McCrae, 1990; Peabody & Goldberg, 1989), it is
possible that the development of personality scales has followed a similar
course. This would mean that some aspects of personality are better as-
sessed than others, construct names notwithstanding. Correlations with
the five-factor model would help to determine whether, and to what de-
gree, such a distinct pattern exists.
Ozer and Riese (1994) argued that determining a scale's pattern of
correlates with the five-factor model will become an essential part of de-
termining its construct validity. As they stated, "[those] who continue to
employ their preferred measure without locating it within the five-factor
model can only be likened to geographers who issue reports of new land
but refuse to locate them on a map for others to find" (p. 361).
 PSYCHOMETRIC OVERVIEW 49

An empirical example may help to highlight the value of the five-fac-

tor model. The data to be discussed are distilled from a study that looked
at the relationships between the ACLand the NEO PI (Piedmont, McCrae,
& Costa, 1991). The ACL was selected because it contains scales from sev-
eral different theoretical perspectives. First, there are the need scales de-
veloped from Murray's (1938) theory. There are also scales developed by
Gough that represent "folk concepts"; scales designed to assess Berne's
(1961) transactional analysis constructs; and, finally, scales developed by
Welsh (1975) to measure origence and intellectence (measures of creativity
and intelligence). These theories have little in common with lexical usage
and their underlying concepts appeared to have no parallels in the lay lan-
guage. It was thought that they may represent factors distinct from the
five-factor model.
A joint factor analysis was performed between the NEO PI and the
ACL in a sample of 244 adults. These individuals were part of the Balti-
more Longitudinal Study of Aging (BLSA; Shock et al., 1984). This is a sam-
ple of predominantly white, community-dwelling individuals who have
agreed to return for periodic biomedical and psychological testing. Most
have at least a college degree and work (or are retired from) scientific, pro-
fessional, or managerial occupations.
A principal components analysis was performed followed by a vari-
max rotation. A scree test indicated that five factors, accounting for 74% of
the variance, would adequately represent the data. The factor loadings are
presented in Table 2-4. Three observations emerge. First, the five factors
that are identified in this solution are each defined by one of the domains
of the NEO PI. This underscores the contention that the five-factor model
represents broad domains of personality and defines the majority of the
variance contained in personality assessment inventories.
The second observation is that all of the ACL scales correlate with at
least one of the factor dimensions. This shows that the diverse personologi-
cal content of these scales can be adequately represented by the dimensions
of the five-factor model and provides support for the comprehensiveness of
the model. If the model were not comprehensive, then some of the ACL
scales would not be correlated with these five dimensions, indicating that
additional factors are necessary for representing that information. But these
results show that at their base, all languages used to describe personality
share a common foundation; scientific conceptions of individual differences
are variants of the terms found in lay vocabulary.
The third observation of Table 2-4 is that the correlations of each of the
ACL scales to the five-factor model provide a personological sketch. For ex-
ample, the Achievement scale correlates with the Extraversion and Consci-
entiousness domains. The latter correlation makes conceptual sense: high
achievers should be organized, disciplined, and goal-oriented. The correla-
50 CHAPTER 2

TABLE 2-4. Joint Factor Analysis of the NED PI Domains and the
Adjective Check List
Factor
Scale N E 0 A C
NEO-PI Factors
Neuroticism (N) .80
Extraversion (E) .70
Openness (0) .79
Agreeableness (A) .68
Conscientiousness (0 .75
ACLScaies
Unfavorable -.59 -.61
Favorable .57 .57
Communality .44
Achievement .51 .73
Dominance .79 .44
Endurance .91
Order .87
Intraception .54 .63
Nurturance .83
Affiliation -.40 .66
Heterosexuality .79
Exhibition .82
Autonomy .45 -.71
Aggression .58 -.65
Change .59 .48
Succorance .49 -.61
Abasement -.67 .40
Deference -.52 .67
Self-control -.71
Self-confidence .77 .45
Personal Adjustment .62 .50
Ideal Self -.48 .50
Creative Personality .43 .71
Military Leadership .80
Masculinity .58 -.41
Femininity -.64
Critical Parent -.80
Nurturant Parent .57 .65
Adult -.40 .80
Free Child .80 .42
Adapted Child .50 -.71
Welsh'sA-1 -.50
Welsh'sA-2 .47 -.40 -.48
Welsh'sA-3 .70
Welsh'sA-4 .SO
Note: N = 244. Only Loadings> 1.40 I are given.
 PSYCHOMETRIC OVERVIEW 51

tion with Extraversion also shows high scorers to be energetic and domi-
nant. In a group they would like to be leaders and may exude positive
"vibes" that others may find inviting and charismatic. Individuals high on
these two dimensions are described by others as being ambitious, proud,
persistent, active, and competitive (Hofstee, de Raad, & Goldberg, 1992).
It is also interesting to note the pattern of correlates found for the Dom-
inance scale. It, too, loads on both Extraversion and Conscientiousness, al-
though it emphasizes the surgency while the Achievement scale highlights
the ambition aspects of personality. But, despite their very different labels,
these two scales are measuring redundant aspects of personality.
Similar evaluations can be done for each of the ACL scales. Each time,
an evaluation of their construct validity and personological uniqueness
can be made. For example, the Adapted Child scale seems not to be what
it claims. The high loading on Neuroticism and the low association with
Conscientiousness portrays an individual who is emotionally distressed,
selfish, and needy of others. This pattern of correlates hardly reflects
"adapted" qualities. Thus, the name for this scale is misleading. Again,
scale names should not be relied on for interpreting scale content.
Finally, the pattern of correlations shows that many of the ACL scales
correlate with the Extraversion, Agreeableness, and Conscientiousness do-
mains. Fewer correlations are found for Neuroticism and Openness. The
ACL does provide a comprehensive description of personality, but does
not evidence equal personological weight across the different domains.
One may wish to complement the ACL with additional measures of these
domains in order to ensure an adequate sampling of all relevant personal-
ity dimensions.
These factor analytic results are similar to the findings of other studies
using different instruments. The studies in Table 2-3 show that the five-fac-
tor model is indeed comprehensive: Personological material from many
different theoretical positions can be substantively organized under the
umbrella of the five-factor model. As Ozer and Riese (1994) suggested, the
correlations of scales with these domains goes far in outlining the scales'
construct validity. However, the comprehensiveness of the five-factor
model should not be construed as suggesting that these domains are all that
are necessary for understanding personality: The five-factor model and the
NEO PI-R are not intended to supplant other measures of personality. Their
inclusion in an assessment battery can ensure that a broad, comprehensive
description of personality is obtained, but other scales may be necessary to
provide more precise measures of specific psychological phenomena.
By way of completeness, I should also add that I conducted a similar
joint factor analysis of the NEO PI with the EPPS in order to establish those
scales' relations to the five-factor model (Piedmont, McCrae, & Costa,
52 CHAPTER 2

1992). It was interesting to note that the achievement scale correlated with
low Agreeableness. This is not the kind of association one would expect to
find. To some degree this result was due to the item content, where high
scorers sought high-status positions so they could influence others. This is
consistent with low Agreeable individuals, who are manipulative and con-
trolling. Yet it was also believed that this unexpected finding was due to
the ipsative nature of the instrument (ipsatization is a relatively complex
mathematical process that puts the test results into a certain form-one
that is characterized by the sum of all scores adding to constant value). But
whatever the reason, it became clear why these two instruments did not
correlate-they were assessing independent dimensions of personality!
Only the presence of a larger, taxonomic structure could provide a way of
understanding why these two achievement scales relate to each other in
the ways they do. Further, the correlations of the EPPS with the NEO PI
provide a framework for reinterpreting the previous literature employing
this instrument.

THE QUESTION OF SELF-DISTORTION

The data presented have shown that the dimensions of the NEO PI-R
are internally consistent, structurally robust, and stable over time. Further,
these dimensions do in fact represent salient individual-difference qualities;
they are present not only in our general language but also among the con-
structs developed by social scientists. However, the majority of the data
presented so far rely on self-reports. It could be argued that the overlap
demonstrated in the research just described is entirely the result of self-re-
ported distortions. In other words, measures of the five factors overlapped
with other personality measures not because of some substantive relation-
ship but rather because of the presence of correlated error. The dimensions
of the five-factor model may represent only individuals' implicit personal-
ity theories, internal fictions used to help explain the behaviors of others.
These dimensions may not have any real correspondence to actual psycho-
logical dynamics (e.g., Mischel, 1968).
Determining whether the dimensions of the five-factor model repre-
sent a cognitive fiction or are real psychological quantities requires appeal
to multiple information sources.

SELF-PEER CONGRUENCE

There is no doubt that self-report data is fallible; it has its strengths

and, importantly, its weaknesses. Sole reliance on self-report data creates
PSYCHOMETRIC OVERVIEW 53

real limitations to the level of accuracy one experiences in drawing infer-

ences. In order to establish a strong assessment foundation, a test user
needs to draw information from other sources and look for where the
threads of information begin to converge. Observer ratings are an ideal
counterpoint to self-reports (see McCrae, 1994a); they do not share the
same sources of bias as self-report data. Although observer ratings have
their weaknesses (e.g., halo effects, stereotypes), these biases do not over-
lap with the errors inherent in self-reports. Thus, convergence between
peer- and self-ratings cannot be attributable to correlated error but to a re-
liable effect (see McCrae, 1982; Wiggins, 1979).
A number of research studies have evaluated the cross-observer con-
vergence of the five-factor model and found it to be quite robust (e.g.,
Costa & McCrae, 1992d; McCrae & Costa, 1987). An example of these data
is presented in Table 2-5. These data are a continuation of the ACL study
discussed previously. Asubsample of individuals who completed the NEO
PI and the ACL also had their spouses and friends rate them on the NEO
PI. Factor scores derived from the information presented in Table 2-4 were
then correlated to those ratings.
The values in bold are the convergent validity coefficients. For exam-
ple, the .43 correlation between Spouse ratings of Neuroticism and Self rat-

TABLE 2-5 Correlations between Peer and Spouse Rated NEO-PI

Factors and Self-Reported Joint Factor Scores
Factor"
NEO-PI Rating N E 0 A C
Spouse (N = 94)
Neuroticism .43' .00 -.16 -.19 -.20~
Extraversion .17 .4"- -.11 .16 -.02
Openness -.10 -.09 .59' .03 .02
Agreeableness .14 .05 .06 .58' -.08
Conscientiousness .08 -.08 .01 .03 .40'

Peers (N = 145)
Neuroticism .41' .10 .03 .03 -.14
Extraversion -.03 .4"- .00 .11 -.11
Openness .07 .10 .71' .01 -.02
Agreeableness .01 -.24~ -.01 .52' -.22~
Conscientiousness .05 -.13 -.05 -.08 .46'
'Based on factors from Table 2-4. Based on "Adjective Check List Scales and the Five-Fac-
tor Model," by R. L. Piedmont, R. R. McCrae, and P. T. Costa, Jr., 1991,/ournal of Personal-
ity and Social Psychology, 60, p. 630-637.
'p< .05
'p < .001
54 CHAPTER 2

ings shows that how individuals perceive themselves on this dimension

corresponds to how others perceive them. As can be seen in Table 2-5,
these cross-observer, cross-method convergent validities are all statistically
significant and of a moderate to high magnitude. The off-diagonal items
are the discriminant validity coefficients. They represent the degree to
which ratings reflect one and only one personality dimension. As can be
seen, all the discriminant validity coefficients are quite small indicating
that when individuals rate a target on these five dimensions, raters have a
clear and specific idea of what these dimensions are. The ratings reflect
specific aspects of personality that agree with the particular self-ratings.
These results show that the ways individuals think about themselves
temperamentally is consistent with how others construe them. Thus, self-
ratings cannot be seen as merely the result of some type of personal dis-
tortion or bias. In a review of the cross-observer literature, Borkenau (1992)
concluded that the data clearly showed that traits are real entities that can
be reliably assessed by raters. Because these ratings correspond very well
with self-assessments, one can have confidence that self-reports are not
solely the product of bias, distortion, or both. Individuals, at least normal
volunteers, do seem to provide accurate assessments of themselves when
asked.

THE REVERSE ACQUAINTANCESHIP EFFECT

One issue that comes up in connection with observer ratings is the

length of time the rater has known the target. Research has shown that the
more visible the trait, the better it can be rated (Funder & Colvin, 1988), and
that the better the rater knows the target, the more accurate the rating
(Borkenau & Liebler, 1993; Colvin & Funder, 1991; Watson & Clark, 1991).
This latter finding is called the acquaintanceship effect, and is a rather
straightforward dictum: The longer you know someone and the more you
know about them, the more accurate your assessments of them. However,
there are caveats to this in that given the right information, even strangers
can make accurate assessments of a target (Borkenau & Liebler, 1992, 1993).
The acquaintanceship effect seems to be a natural corollary of rater
phenomenon when the assessments are being made during a period of
personality stability. However, in a recent paper (Piedmont, 1994), I ques-
tioned whether during a period of personality change and development
(e.g., adolescence, early adulthood) such long-term knowledge of a target
would prove more of a liability than an asset. It may be possible that a
long-term acquaintance may provide less accurate ratings than a more re-
cent acquaintance because the images of the target the former holds may
not be keeping pace with the real personality changes that are occurring.
PSYCHOMETRIC OVERVIEW 55

The older acquaintance may be more likely to attribute initial changes in

the target to situational factors rather than to personality development.
Long-term acquaintances may also be disposed to perceive more stability
in the target's personality than what may actually exist in order to bring a
better sense of continuity to their relationship. On the other hand, newer
friends, whose perceptions are still evolving, may be able to utilize these
perceptions of variability in the target better when forming their person-
ality impressions.
In order to evaluate this hypothesis I had 101 college students complete
the NEO PI-R for themselves and had two individuals who knew them for
at least 6 months-rate them on the rater version of the NEO PI-R. Raters
knew the targets, on average, 8.6 years (SD =8.1; range three months to 25
years). Raters were also believed to have known the subjects very well.
In evaluating the cross-observer convergence, raters were divided
into two groups: One group knew the target for 2 years or less, the other
for more than 2 years. The results showed a modest effect: The average
convergent correlation for the 2-year-or-Iess group was .56, while for the
more established raters it was approximately .47. During this period of
personality transition, those who knew the targets longer were slightly less
accurate in their perceptions than those who were more recent friends.
This effect may result from the same dynamics that underlie the fun-
damental attribution error-the tendency to attribute dispositional causes to
the behavior of others and situational causes to our own behavior (Jones &
Davis, 1965; Jones & Nisbett, 1972). Long-term acquaintances have formed
well-developed and detailed conceptualizations of the targets. During pe-
riods of personality change, targets will begin to evidence dispositional
shifts in their attitudes, feelings, and behaviors. The resulting lack of cor-
respondence between the person raters believe the targets to be and the
subjects' changing behaviors may lead raters to attribute these shifts to sit-
uational forces. Such a strategy guards against the formation of erroneous
impressions when the target's personality is not in transition. However,
during a period of change, it may lead long-term acquaintances to ignore
real dispositional shifts.
More recent acquaintances, on the other hand, are still actively devel-
oping their understanding of the targets. Each new observation is used to
elaborate their developing dispositional impressions. Thus, recent ac-
quaintances are more likely to attribute shifts in behavior to dispositional
dynamics. During periods of personality change, this lower threshold for
attributing behavioral inconsistencies to personality characteristics enables
newer acquaintances to keep pace with the actual changes that are occur-
ring. However, when personality is stable, this tendency to make disposi-
tional attributions results in the fundamental attribution error.
56 CHAPTER 2

When using observer ratings of personality one needs to be sensitive

to the age of the target. Younger subjects are more likely to be in a process
of personality development and change. Ratings obtained from long-term
informants, such as parents, may not be as accurate as the impressions gar-
nered from newer acquaintances, such as teachers or friends.

THE LOGIC OF ASSESSMENT USING THE NEO PI-R

There is no doubt that the use of psychological testing for any purpose
is complex and fraught with much potential for bias and error. Also, there
are no simple solutions for addressing this formidable array of complex is-
sues. In the previous section I addressed the issues of distortion as they may
relate to self-reports. As can probably be discerned from the earlier discus-
sion, the best way for determining whether a self-report provides accurate
information or is merely the by-product of some type of cognitive distor-
tion is to obtain an independent source of information: an informer report.
Because observers may not have the same motivations to present the
target in a good or bad light, and certainly do not share the same cognitive
framework, an observer rating provides a different perspective on a target.
At a minimum, an observer rating can give an insight into the kinds of so-
cial impressions the target makes on others. Finding convergence between
a self-report and an observer rating lends greater strength to the inferences
one can draw from the profiles. If a person admits to being interpersonally
awkward, and several others who know the target also indicate him or her
to be socially awkward, then you can be confident that the target is socially
awkward.
The value of observer ratings cannot be overstated. It is the value of
such information that led Costa and McCrae to develop a rater version of
the NEO PI-R (Form R). Items are phrased in the third person and there
are separate forms for rating men and women. Psychometric analyses of
this form show the scales to be very internally consistent (alpha for the do-
mains ranges from .89 to .95, and for the facets from .60 to .90). The factor
structure of Form R is comparable to its self-report sibling (Costa & Mc-
Crae, 1992c; Piedmont, 1994). Finally, personality ratings on Form R have
also been shown to evidence substantial temporal stability (Costa & Mc-
Crae, 1992c).
Observer data should be obtained whenever possible. It can do much
to help determine whether a self-report contains bias as well as provide a
different perspective for understanding a client. Later chapters in this book
outline the usefulness of rating data and how to incorporate such infor-
mation into the interpretive process.
 PSYCHOMETRIC OVERVIEW 57

NORMAL VERSUS ABNORMAL PERSONALITY

OVERVIEW

In evaluating results from the NEO PI-R it is important to keep in

mind that the qualities represented by these scales pertain to the normal
spectrum of psychological functioning. Scores on the NEO scales reflect
the motivations, tendencies, and capacities that characterize individuals'
ongoing interactions with their environments. Specific configurations of
scores on these domains reflect the creative adaptations individuals have
made to the demands imposed on them by their life experiences. In short,
when we speak of"personality," we refer to the consistent ways in which
individuals perceive the universe, their role or place in it, and the direc-
tions in which they desire to move.
These broad strokes at defining personality attempt to highlight the
fact that our intrinsic organization developed in response to the unique
problems of adaptation we have faced. Our personalities represent cre-
ative solutions to these problems. When interpreting scores from the NEO
PI-R it is important to keep this notion in mind and not impose any value
judgments on the scores. One frequent mistake committed by those pro-
fessionals I have trained in the use of the NEO PI-R has been to see some
of these qualities as better than others.
For example, one may perceive that being low on Neuroticism is
preferable to being high on the domain, or that having a high standing on
Agreeableness is better than scoring low. For most of the domains one can
think of "preferable poles." But this is not a productive way of construing
scores on the NEO PI-R. Per~onalities should not be thought of as "good"
or "bad." Personality is an adaptive structure; we develop the qualities we
have because they enable us to survive and thrive. The opposing poles for
each domain have their strengths and their liabilities, depending on the
circumstances (T. Widiger & Corbitt, 1994).
One may argue that being high on Agreeableness is a positive virtue.
Being compassionate, caring, helpful, and responsive to the needs of oth-
ers are certainly socially desirable characteristics. In general, we all value
these qualities, and respond warmly to those who seem to typify them,
such as Mother Teresa or Gandhi. However, if you were to invest your life
savings in the stock of a company, hoping for it to grow, would you be in-
clined to entrust your money to a company run by Mother Teresa or
Gandhi? Or would you prefer to have someone running the company who
is much more manipulative and cunning, able to outmaneuver the com-
petition and undermine the efforts of future competitors-say, someone
like Bill Gates?
58 CHAPTER 2

Thus the question to be asked from a personality inventory is not

whether an individual has a good or bad personality, but rather how well
the individual's personality fits with the demands of a given situation or
environment. There is a place for all different types of people, and we use
personality assessment inventories to assess needs and identify comple-
menting environments.
Because the NEO PI-R scales capture these normal, adaptive qualities,
high or low scores should not be seen as representing anything abnormal.
This is not a measure of psychopathology, nor is it designed to capture such
types of processes. Therefore, it is inappropriate to draw such inferences from
the scales. This.is particularly true on the Neuroticism domain. This term was
selected to define this dimension because of its historical place in the psy-
chologicalliterature. However, it does not refer to the kinds of dynamics the
term originally reflected. High scores on these facets (e.g., Anxiety, Hostility,
Depression, Self-Consciousness, Impulsiveness, and Vulnerability) do not in-
dicate any type of Axis I disorder. For example, an individual can have a very
high score on the Depression facet, but this does not mean that the person is
suffering from clinical depression. However, as we shall see later, those suf-
fering from clinical depression will tend to score high on this facet. Scoring
low on Agreeableness and Conscientiousness does not mean that someone
has an Antisocial Personality Disorder; although those with an Antisocial
Personality Disorder do score low on those domains.
What this tells us is that the scores from the NEO PI-R need to be in-
terpreted from a nonclinical vantage point. Scores from these scales speak
to the regularities in an individual's ongoing behavior. However, to the ex-
tent that the five-factor model provides a complete description of funda-
mental personality dimensions, one might expect some relationship to
those dynamics that define characterological impairment as well (Widiger
& Frances, 1985; Widiger & Trull, 1992). The next section evaluates the re-
lations between the five-factor model and dimensions of clinical dysfunc-
tion. The value of such correspondence for future clinical and research
practices is highlighted.

THE FIVE-FACTOR MODEL AND ITs RELATIONS TO CLINICAL BEHAVIOR

Given that the five-factor model is a comprehensive description of
normal personality dispositions capable of organizing individual-differ-
ence constructs from a wide array of theoretical models, and able to pro-
vide a high degree of empirical power in predicting numerous life
outcomes, it is not surprising that researchers have examined the role these
dimensions may play in abnormal behavior as well. Although the five fac-
tors are descriptive of nonclinical functioning, these dimensions may
nonetheless have something to contribute to our understanding of abnor-
 PSYCHOMETRIC OVERVIEW 59

mal functioning. In fact, several recent volumes have appeared that at-
tempt to outline theoretically the expected linkages between traits and
clinical nosology (Costa & Widiger, 1994; Strack & Lorr, 1994).
Widiger and Trull (1992) suggested four ways in which traits may be
related to Axis I disorders: (a) they may predispose individuals to a disor-
der; (b) they may be a result of the disorder; (c) they may affect the expres-
sion of the disorder; or (d) they may share a common etiology with the
disorder. McCrae (1994c) contended that the five-factor model can provide
a framework for evaluating person by treatment interactions (see also
Miller, 1991). Costa and McCrae (1992a) have also argued that the assess-
ment of these normal personality dimensions can provide valuable, nonre-
dundant information to clinical assessment protocols that rely solely on
measures of symptomatology. Finally, the five-factor model is increasingly
being perceived as a dimensional alternative to the current categorically
based classification system (Costa & Widiger, 1994).
There are four major questions that need to be addressed when con-
sidering the utility of the NEO PI-R in a clinical context. First, does the
psychometric integrity of the instrument continue to be evidenced when
used clinically? Second, what are the relations between these five domains
and Axis I symptomatology? Third, what are the relations between the
five-factor model and Axis II symptomatology? Finally, what is the incre-
mental validity of the NEO PI-R? Does it provide any additional, clinically
relevant information over what is already available with current instru-
ments? Each of these questions is discussed in tum.

The Psychometric Integrity of the NEO PI-R in a Clinical Setting

Given the relatively recent appearance of the NEO PI-R, there have
not been many studies conducted to date evaluating the psychometric
qualities of the instrument in a clinical context. However, those that are
available have much to encourage its usage.
Piedmont and Ciarrocchi (in press) gave the NEO PI-R to a sample of
132 outpatient, substance-abuse clients. Descriptive statistics showed this
sample to be high on Neuroticism and low on Agreeableness and Consci-
entiousness, a pattern characteristic of a substance-abuse disorder
(Brooner, Herbst, Schmidt, Bigelow & Costa, 1993). Alpha reliabilities for
the domains were all quite high (ranging from .83 to .89). Comparable val-
ues for the facets were lower than those found normatively (range from .44
for Excitement Seeking to .75 for Hostility; median = .61). A factor analy-
sis of the scores clearly revealed the five-factor structure. In order to deter-
mine whether this obtained structure was identical to the normative factor
structure, congruence coefficients were calculated (Gorsuch, 1983). Con-
gruence coefficients are like correlation coefficients; they determine the de-
60 CHAPTER 2

gree of similarity between two sets of factor loadings. Using critical values
provided by McCrae, Zonderman, Costa, Bond, and Paunonen (1996), it
was determined that the obtained factor structure can be considered iden-
tical to the normative structure. This tells us that the NEO PI-R in a clini-
cal context continues to manifest the same underlying factor structure that
it does with nonclinical volunteers.
Retest reliability coefficients were also evaluated. Six weeks after ad-
mission to the outpatient program, participants completed another NEO
PI-R as part of their discharge process. In evaluating these values, one
needs to keep in mind that the respondents had just gone through a treat-
ment program designed to impact their ways of perceiving, interpreting,
and responding to the world. Correlations between scores taken pre- and
postcounseling were quite high for both the domains (range .52 for Neu-
roticism to .79 for Openness) and facets (range .42 for Order to .72 for Aes-
thetics, median = .58), although these values were lower than normative
values. This may have been a result of real individual-difference changes
among the participants as a consequence of their treatment.
Two other studies have evaluated the reliability of the NEO PI in a
clinical context. Trull and colleagues (1995) found 3-month retest coeffi-
cients averaging .85 for a sample of university-based clients receiving out-
patient services. Similar findings were reported by Bagby and colleagues
(1995) in a sample of depressed outpatients; they found retest reliabilities
averaging about .73. In all studies, mean level changes on the domains
were noted. The Piedmont and Ciarrocchi (in press) study evidenced large
(about one-half standard deviation) changes on all five domains. The other
two studies showed smaller and more circumscribed changes. These dif-
ferences may have been a function of the type of sample and intensity of
the treatment. Nevertheless, these three studies show that the NEO scales
remain structurally valid and reliable in a clinical context.

The NEO PI-R's Relations to Axis I Disorders

The impetus for evaluating the correspondence between Axis I symp-
tomatology and the five-factor model is not so much to identify the model
as a clinical paradigm, but to emphasize that psychological distress does
not develop in a vacuum. Adient's presenting problems find a foundation
in his or her larger motivational patterns. In this way, many psychosocial
difficulties may be seen as problems in living associated with specific
types of personality configurations. From this perspective, measures of the
five-factor model can have two uses. First, obtaining a profile of a client on
the five factors can help better outline the larger internal forces that have
brought the individual to his or her current situation. Is the presenting
 PSYCHOMETRIC OVERVIEW 61

problem a reaction to recent events or part of a larger, ongoing distressed

lifestyle? Second, measures of the five-factor model may be useful for iden-
tifying potential psychological risks a person may face. For example, being
high on Neuroticism may predispose one toward experiencing a clinical
depression (Zonderman, Herbst, Schmidt, Costa, & McCrae, 1993).
Two studies have evaluated the relations between the MMPI and the
NEO PI (Costa, Busch, Zonderman, & McCrae, 1986; Costa, Zonderman,
McCrae, & Williams, 1985). Both studies showed that the content of the
MMPI overlaps significantly with the dimensions of the five-factor model.
Further, it was noted that the MMPI contained a heavy emphasis on Neu-
roticism in its content and had an underrepresentation of Conscientious-
ness. This is important to note because the MMPI is frequently relied on to
provide a description of a client's personality. Because it lacks any mea-
surement of Conscientiousness, it cannot be counted on to provide a com-
plete description of personality.
Piedmont and Ciarrocchi (in press) provided correlations between the
NEO PI-R and self-reported scores on the Brief Symptom Inventory (BSI;
Derogatis, 1993), a measure designed to capture Axis I symptomatology,
and the Personal Problems Check List (PPCL; Schinka, 1985) a measure of
the extent and type of psychosocial difficulties encountered by the client.
High Neuroticism and low Conscientiousness were the strongest and most
consistent correlates of the BSI, indicating that the experience of such
symptoms as Hostility, Phobic Anxiety, Interpersonal Sensitivity, and Para-
noid Ideation had a strong motivational foundation. High Neuroticism
and low Conscientiousness characterize an individual who experiences
high levels of affective dysphoria and seeks succorance for this distress
through any type of immediate gratification.
Interestingly, scores on the PPCL also correlated with high Neuroti-
cism and low Conscientiousness. Clients with this personality configura-
tion reported experiencing numerous problems related to social, religious,
emotional, sexual, and health issues. These correlations highlight the mal-
adaptive aspects to these five factors. Certain personality dispositions tend
to increase the likelihood that individuals will experience certain types of
problems with their environment. In particular, Neuroticism seems to be
the single best predictor of the level and extent of problems experienced.
On one level this is not very surprising, given that the dimension represents
high levels of negative affect. But what is interesting is that these character-
istic levels of distress will generalize into an individual's life path and cre-
ate difficulties in many other areas. Research has shown that individuals
high on Neuroticism are at risk for receiving a psychiatric diagnosis or ex-
periencing depressive symptomatology (Saklofske, Kelly, & Janzen, 1995;
Zonderman et ai., 1993). Those high on Neuroticism are also likely to
62 CHAPTER 2

experience a lowered sense of job satisfaction and have an increased risk of

burning out (Piedmont, 1993).
Scores on the NEO also have been shown to be useful in distinguish-
ing motivationally between various nosological types. For example, scores
on the NEO PI have been shown to discriminate between those with a sex-
ual disorder and those with a sexual dysfunction (Fagan et al., 1991) and
are predictive of one's level of sexual functioning (Costa, Fagan, Piedmont,
Ponticas, & Wise, 1992). Trull and Sher (1994) showed that scores from the
NEO-Five Factor Inventory (NEO-FFI; this is a 60-item short form of the
NEO PI-R-it assesses only the five global domains) were differentially
sensitive to various Axis I diagnoses, such as drug addiction, anxiety dis-
orders, major depression, and phobias.
Overall, the results of this growing body of research support the util-
ity of the five-factor model in clinical assessment. Scores on the five factors
are related to the type and amount of symptom expression. Appreciating
the motivational basis of many psychological symptoms can assist a clini-
cian in two ways. First, scores from the NEO PI-R can help establish a
broader context for evaluating symptoms. Are the presenting problems a
reaction to recent events or are they symptoms of enduring and pervasive
maladjustment? Being able to disentangle the motivational basis of the
presenting problem may aid in selecting appropriate treatments.
A second value of the NEO PI-R is that it can help establish realistic
and appropriate therapeutic goals. For example, the Piedmont and Ciarroc-
chi (in press) study showed that individuals who were experiencing many
social problems were also high on Neuroticism and low on Agreeableness
and Conscientiousness. Such individuals feel very insecure and anxious in
relationships with others. Their low Agreeableness and Conscientiousness
indicates an interpersonal style that is selfish and manipulative. These indi-
viduals may see others as objects whose function is to provide physical grat-
ification to assuage the individual's affective distress. Any treatment
intervention needs to appreciate that individuals with this type of person-
ality profile will always maintain an egocentric view of the world that will
be characterized by a "me first" attitude. Attempting to rework this charac-
ter style into one that is defined by trust and mutuality may not be possible.

The NEO PI-R and Its Relations to Personality Pathology

The application of the five-factor model to personality pathology
seems a natural and logical step. Because a personality disorder is "an en-
during pattern of inner experience and behavior that deviates markedly
from the expectations of the individual's culture, is pervasive and inflex-
ible, . . . is stable over time" (American Psychiatric Association, 1994,
 PSYCHOMETRIC OVERVIEW 63

p. 629), it is reasonable to believe that these ongoing, stable patterns find

root in one's personal dispositions. Therefore, charting these maladaptive
qualities in the five-factor taxonomy may help to shed additional etiologi-
cal, motivational, and prognostic light on this aspect of functioning.
The interest in applying a dimensional model to the Axis II disorders
stems from a dissatisfaction with the current categorical nosology of the
Diagnostic and Statistical Manual of Mental Disorders '(DSM-IV; American
Psychiatric Association, 1994). Anumber of critics have appeared, finding
fault with many aspects of the document, including its high prevalence of
boundary cases, its descriptive nature, and the putative independence of
the Axis I and Axis II disorders (Frances et al., 1991; Livesley, Schroeder,
Jackson, & Jang, 1994; Widiger & Shea, 1991). A growing controversy has
developed over whether characterological pathology is better conceptual-
ized by a dimensional or categorical model (Widiger & Costa, 1994; Widi-
ger & Frances, 1994).
One basic assumption of the categorical model is that pathology rep-
resents a mode of functioning separate and distinct from normality. Those
individuals who exhibit more than the threshold number of criteria are be-
lieved to be qualitatively different from those who meet fewer criteria.
Thus, the distributions of the phenotypic features of personality disorder
are discontinuous and show either a bimodal distribution or a point of rar-
ity (Livesley et al., 1994). A dimensional model, on the other hand, as-
sumes that the underlying qualities of personality pathology are the same
as those found among nonclinical individuals. Thus, there is a continuous,
normal distribution of scores for each of the traits.
Given that these two models generate very different expectations,
they can be readily tested. Anumber of studies have shown that a dimen-
sional model does better represent the qualities underlying the personality
disorders than does a categorical model (Helmes & Jackson, 1994; Livesley,
Jackson, & Schroeder, 1992; Livesley et al., 1994; Watson, Clark, & Hark-
ness, 1994). As Schroeder, Wormworth, and Livesley (1992) have con-
cluded, "personality disorders are not characterized by functioning that
differs in quality from normal functioning" (p. 52). Not only do the same
traits that describe normal behavior also describe pathological behavior,
but research continues to show that the traits of the five-factor model of
personality provide a strong paradigm for organizing characterological
dysfunctioning-measures of the Big Five overlap heavily with ratings of
personality dysfunctioning from a wide variety of Axis II assessment tools
(e.g., Duijsens & Diekstra, 1996; Trull, 1992; Trull et ai, 1995; Widiger &
Costa, 1994; Wiggins & Pincus, 1989).
Widiger and Frances (1994) outlined three advantages to using a di-
mensional model for conceptualizing Axis II disorders. The first is its
 64 CHAPTER 2

flexibility. Using continuous scores to describe functioning allows clini-

cians to establish various cutoff points for determining the presence of
pathology that are sensitive to the context and type of clinical issues ad-
dressed. In the current clinical reality, decisions need to be made about
clients concerning hospitalization, responsiveness to medication, insur-
ance coverage, type of therapy, and so on. The current categorical nature of
the DSM-IV provides only a single set of thresholds that may not be opti-
mal for different types of decisions. The dimensional model allows one to
determine different cutoffs that are relevant to those outcomes.
A second benefit of the dimensional model noted by Widiger and
Frances (1994) is the retention of information. The polythe tic nature of
the DSM-IV gives rise to very heterogeneous diagnostic categories. For
example, there are 256 different ways to meet the DSM-JV criteria for
Borderline Personality Disorder and 99 ways to meet the criteria for An-
tisocial Personality Disorder, yet all receive the same diagnostic label. Al-
though the categorical model can appreciate the diverse expression of a
single pathologic process, it fails to appreciate the unique ways the dis-
order is expressed in a given client. And the degree to which a client may
be atypical for a category will proportionally reduce the utility of the di-
agnostic label.
The dimensional model can reduce such stereotyping by providing
much more precise information about a client. Subtleties and nuances can
be detected by evaluating how much of each personality trait a client ex-
hibits. The graduations provided by the ordinal-interval level personality
scales can provide a rich source of information addressing issues of etiol-
ogy, level and style of functioning, and prognosis.
The final, and perhaps most useful, advantage of the dimensional
model is its ability to improve differential diagnosis. DSM-IVestablishes
behavioral thresholds for inclusion in a diagnostic category. However, the
degree to which those criteria overlap will undermine one's ability to as-
sign a client to one and only one category. Problems surrounding comor-
bidity and differentiation have long plagued the DSM and explain why the
diagnosis Personality Disorder, Not Otherwise Specified is so frequently
given for clients on Axis II.
For example, Livesley, West, and Tanney (985) have noted the diffi-
culty in differentiating between the avoidant and schizoid personality dis-
orders. Both groups prefer solitary activities, have few social contacts or
relationships, and remain interpersonally aloof. On a phenotypic level,
both disorder categories may appear quite similar; the one distinguishing
characteristic is that the avoidant feels his or her loneliness while the
schizoid may be content with the isolation. Although DSM-IV has reduced
the number of overlapping criteria between these two categories evi-
 PSYCHOMETRIC OVERVIEW 65

denced in previous editions, it still may be difficult for a clinician to dis-

cern whether the reticence concerning social interaction is a result of a per-
sonal sense of inadequacy or a constricted interpersonal style.
One way to resolve this issue is to plot the personality disorders in the
space defined by the five major personality disorders. Then, each disorder
can be understood by its characteristic pattern of loadings. As we did with
understanding personality scales previously, triangulating the personality
disorders with the five-factor model can not only highlight the kinds of
personality qualities captured by these diagnostic categories, but also de-
lineate the similarities and differences among the categories.
An example of this is given by Trull (1992), who presented correla-
tions between the NEO-PI and the MMPI Personality Disorder (MMPI PD)
scales in a sample of 54 psychiatric outpatients. The results of this analy-
sis are presented in Table 2-6. Several observations can be noted from these
data. The first concerns the utility of the five-factor model for making dif-
ferential diagnoses. Consider the pattern of correlates for the Avoidant and
Schizoid personality scales. Both have strong, negative loadings on Extra-
version, highlighting the fact that both groups find themselves interper-
sonally distant and aloof; they tend to express little positive affect and seek
solitary pursuits. Such similar interpersonal styles makes it difficult to dif-
ferentiate between the two phenotypically. However, correlations with the
NEO-PI domains show that the Avoidant tends to have higher levels of
emotional distress than the schizoid, who may appear more placid and
content. The Avoidant has inner feelings of inadequacy, anxiety, and fear.
The Avoidant may wish to be in a group but is too socially phobic to initi-
ate any interactions. The Schizoid, on the other hand, may feel perfectly
content with interpersonal isolation. Because both types of individuals
may be taciturn when it comes to clinical interviews, an effort needs to be
made to measure levels of emotional distress.
Another example concerns the Histrionic and Narcissistic personality
disorders. Both groups have strong positive correlations with Openness
and Extraversion. This pattern reflects the attention-drawing, self-aggran-
dizing styles of these two groups. However, the Narcissistic claims a level
of emotional stability and placidity that may not be present in the Histri-
onic, who may evidence some level of insecurity. It should also be noted
that the amount of personological overlap among the personality diag-
noses raises questions about the amount of redundancy reflected in these
categories. Are all necessary to capture this type of disordered function-
ing? Could some diagnostic categories be meaningfully collapsed into oth-
ers? Are there other disorders that have yet to be discovered? The relative
absence of Openness among the personality disorder scales makes this a
viable empirical question.
66 CHAPTER 2

TABLE 2-6. Correlations between the MMPI-PD Scales and the Five
Domains Scores of the NEO PI
Personality
NEO PI Domain Score
Disorder
Symptoms N E 0 A C R
Paranoid .46' -.06 .04 -.48' -.05 .61'
Schizoid .00 -.72' -.27" -.19 .02 .73'
Schizotypal .45b -.49' -.21 _.40b -.12 .66'
Obsessive-compulsive .52' -.29" .03 -.27 -.14 .58'
Histrionic -.01 .61' .39b -.14 -.14 .71'
Dependent .64' -.21 -.01 -.08 -.33" .67'
Antisocial .29" -.22 -.02 -.42& -.27" .50"
Narcissistic -.30" .59' .43& -.06 .20 .73'
Avoidant .55' -.63' -.27 -.16 -.19 .78'
Borderline .61' .13 .18 -.45& -.24 .75'
Passive-Aggressive .56' -.19 .11 -.45& -.37& .68'
mUltiple R .82' .80' .54 .66& .66b
'p<.05
&p < .01
'p < .001, two-tailed
Note: From "OSM-III-R Personality Disorders and the Five-Factor Model of Personality," by T. J.
Trull, 1992, Journal of Abnormal Psychology, 101, p. 557. Copyright 1992 by the American Psycholog-
ical Association. Adapted with permission.
N = 54. All correlations are two-tailed. Correlations are based on raw NEO PI scores and raw
MMPI-PD scores. R's in the bottom row indicate the correlations between the 11 personality dis-
order scores and each NEO PI domain. R's in the last column indicate the correlation between the
five NEO PI domain scores and each personality disorder. MMPI-PD = Minnesota Multiphasic Per-
sonality Inventory-Personality Disorder.

Another observation from Table 2-6 is that each of the personality dis-
order scales correlates with at least one of the five personality domains,
demonstrating that personality qualities underlying the personality disor-
ders are related to those dimensions that describe normal functioning. The
degree of overlap is quantified by the multiple Rs that are presented in the
last column and row of the table.
The multiple Rs in the last column were obtained by using each MMPI
PD scale score as the dependent variable in a multiple regression analysis.
The five personality factors of the NED-PI were entered as predictors. The
resulting multiple R indexes the degree of relationship between each per-
sonality disorder and the five-factor model. Squaring these values will
provide the amount of shared variance. As can be seen, each of the MMPI
PO scales is strongly associated with the personality domains, sharing be-
tween 25% and 61 % of their variances.
The multiple Rs in the last row index the degree to which each of the
personality domains is involved in the expression of deviant charactero-
logical functioning. These values were obtained by using each NEO-PI
 PSYCHOMETRIC OVERVIEW 67

domain as the dependent variable in a regression analysis, with the MMPI

PD scales entered as the predictors. As can be seen, the personality do-
mains are heavily related with Axis II functioning. Between 29% and 67%
of the variance in the NEO-PI domains overlaps with the MMPI PD
scales. There is no doubt that the five-factor model of personality in gen-
eral, and the NEO scales in particular, have much in common with Axis
II functioning (see Widiger & Costa, 1994, for a review of several similar
studies).
Another point of interest is that Openness to Experience is least in-
volved with Axis II symptomatology; it correlated with only the Schizoid,
Histrionic, and Narcissistic disorders. This may suggest, as noted previ-
ously, that there may be other disorders, as yet undiscovered, that involve
facets of this domain. Because Openness is the most underrepresented of
the five domains in the psychological literature, future research may wish
to further evaluate how this domain may play into characterological dys-
functioning. A related observation is the strong presence of Neuroticism
and Extraversion. These two domains seem to play an important role in
describing Axis II functioning. Essentially, high levels of emotional dys-
phoria and a withdrawn, aloof interpersonal style characterize a wide va-
riety of psychological syndromes. This may partially explain why making
differential diagnoses is difficult-because many of the disorders share
similar features. One way to disentangle this indistinctiveness would be to
evaluate the relatedness of the facet scales to the personality disorders
(Clark, 1990; Harkness, 1992; Widiger & Costa, 1994).
Facet scales can provide a more fine-grained analysis of the persono-
logical qualities expressed in the personality disorders. For example, low
Agreeableness is a correlate of both the Antisocial and Borderline disor-
ders. It may be possible that the elements of low Straightforwardness and
low Compliance are defining of the Antisocial disorder while the elements
of low Trust and low Altruism may define the Borderline disorder. The for-
mer reflects the manipulative self-centeredness of the sociopath while the
latter captures the interpersonal alienation of the borderline. Thus, disor-
ders that appear similar personologically at a global level of description
may become distinct at a more specific level.
This leads to a final observation, that the five-factor model can be very
useful for understanding Axis II functioning. The correlations of the per-
sonality disorder scales with these dimensions help to flesh out the per-
sonological implications of each. For example, the Paranoid disorder scale
correlates with high Neuroticism and low Agreeableness. Such individu-
als are perceived as being suspicious, ill-tempered, demanding, angry, in-
tolerant, faultfinding, quarrelsome, and cranky (Hofstee et al., 1992).
Certainly, the Paranoid personality possesses a high level of affective
 68 CHAPTER 2

distress that fuels an active suspicion and distrust of others. The Histrionic
personality reflects qualities of high Extraversion and high Openness.
Such an individual can be described as being theatrical, intense, dramatic,
expressive, and eloquent.
It is clear that both the five-factor model of personality and the NEO
PI-R have something important to contribute to our understanding of Axis
II functioning. Those personality qualities that define normal functioning
are the same as those that describe abnormal functioning. As to where one
may draw the line, that is a question for future research. At this point there
is no cutoff score that clearly demarcates the boundary between the two
levels. Scores. on the NEO PI-R reflect only normal personality dynamics.
An extreme score on any dimension does not indicate the presence of any
type of maladaptiveness. It may be necessary to develop additional mea-
sures that capture individual-difference qualities that are more germane
diagnostically. Nl)netheless, scores on the NEO PI-R can be used to ad-
dress Axis II-type issues in two ways. First, as Costa and McCrae (l992a)
have argued, scores on these scales can be useful in making differential di-
agnoses. An individual high on Agreeableness is not likely to have an An-
tisocial Personality Disorder. Relatedly, high scores on Neuroticism can
help rule out a Schizoid disorder. Second, an individual's pattern of scores
on both the domains and facets can be compared to patterns of NEO PI-R
profiles obtained from individuals known to possess various personality
disorders. An empirical index of profile agreement can then be calculated.
Very high congruence coefficients may suggest the presence of a personal-
ity disorder, or, at the least, indicate the presence of certain character styles
similar to those whose disordered profile signature is being mimicked. The
computer report generated for the NEO PI-R will conduct such analyses,
systematically comparing the obtained profile to prototype profiles for
each of the personality disorders. In this way, clinicians can be alerted to
the presence of salient character traits that could significantly impact the
therapeutic process.
In any event, the research findings make it clear that Axis II function-
ing may be better captured by a dimensional model of personality than by
the current categorical system. As the research evidence mounts, future re-
visions of the DSM may move in that direction. In the meantime, the NEO
PI-R has much to contribute to the clinical context; it is a very useful com-
ponent of the assessment process. As the research presented in this section
has shown, scores on the NEO PI-R overlap significantly with a number of
clinical measures. The next question to be addressed is, liTo what degree
does the NEO PI-R provide information about clients that cannot be ob-
tained from standard clinical instruments?" It is to this question of incre-
mental validity that we now tum.
 PSYCHOMETRIC OVERVIEW 69

The Incremental Validity of .the NEO PI-R

The preceding sections have demonstrated the value of the five-factor

model for conceptualizing psychopathology. Although originally based in
language, the five-factor constructs not only comprehensively outline nor-
mal personality dynamics, but also have relevance for organizing and un-
derstanding clinical characteristics. This linkage provides a sense of
parsimony to the psychological literature: A wide range of dynamics can
be usefully described by a limited set of constructs. However, efforts at
linking the five-factor model to psychopathological dynamiCS has not been
without its critics.
Ben-Porath and Waller (1992) have argued that the enthusiasm over
the five-factor model earned in normal assessment should not persuade
clinicians to uncritically accept such measures for clinical measurement.
They argue that there already exist a large number of clinical measures that
have been established as valid for such populations. These measures have
been derived from extant theories of abnormal functioning and have been
demonstrated to be useful in assessing such clients. They note many prob-
lems in using an instrument derived from "normal" constructs in a clinical
context, not the least of which is the need to demonstrate that a scale such
as the NEO PI-R "contributes incrementally to the procurement of diag-
nostic information beyond that which is obtained from current clinical
measures" (Ben-Porath & Waller, 1992, p. 17). In other words, what is it
that the NEO PI-R adds to a clinical assessment protocol that is not already
obtained from existing measures?
A growing literature is developing that addresses this issue of incre-
mental validity and consistently demonstrates that the NEO PI-R does
provide clinically relevant information a~ove and beyond what is already
obtained by more standard measures of psychopathology. Trull and Sher
(1994) have shown that the domains of the NEO were related to Axis I di-
agnoses of substance abuse disorder, anxiety disorder, and major depres-
sion, even after the effects of gender and current psychopathology were
controlled. The dimensions of the five-factor model were also shown to ex-
hibit unique patterns of relations with the various disorders. For example,
Neuroticism and Extraversion were particularly sensitive to the social
phobia diagnosis, while Agreeableness was less sensitive. The findings of
unique personality by diagnosis interactions led Trull and Sher to con-
clude that the five-factor model may aid in the differential diagnosis of
Axis I disorders.
Trull and colleagues (1995) evaluated the incremental validity of the
NEO PI with regard to the personality disorders. Scores for each of the per-
sonality disorders were obtained using both the results of the Structured
70 CHAPTER 2

Interview for DSM-III-R (DISP-R) and a self-report measure of personality

pathology, the Personality Disorder Questionnaire-Revised (PDQ-R).
These scores served as the criterion measures in two separate hierarchical
stepwise regression analyses. On the first step of the regression, scores
from the Beck Depression and Anxiety scales were entered to remove vari-
ance attributable to acute mood variation. On the next step, scores from the
NEO PI domains were entered. Partial F-tests determined whether the in-
crease in explained variance due to the five-factor domains was signifi-
cant. For the SIDP-R, 8 of 13 incremental validity coefficients were
significant; all 13 were significant for the PDQ-R scales. Again, the NEO
domains scores provided additional explanatory power to the Axis II dis-
orders over and above information provided by clinical measures of symp-
tomatology.
Finally, Piedmont (1996a) used the Global Pathology Index score ob-
tained from counselor ratings on the Derogatis Psychiatric Rating Scale
(formerly known as the Hopkins Psychiatric Rating Scale) as the criterion
measure in a hierarchical regression analysis involving outpatient sub-
stance abusers. On the first step of the analysis, self-report scores from the
Brief Symptom Index (Derogatis, 1993), a measure of Axis I symptomatol-
ogy, were entered. On the next step, domain scores from the NEO PI-R
were put in. The partial F-test again revealed a significant increase in ex-
plained variance.
The results of these three studies show that clinical distress does not
occur in a vacuum. The types of problems individuals encounter as well as
the intensity of distress arise from underlying, stable characterological dis-
positions. The NEO PI-R provides a wealth of information about individ-
uals, including their motivational aspirations, coping styles, interpersonal
approaches, character styles, and levels of personal well-being. All of this
information contributes significantly to clinical outcome indices above and
beyond information obtained from symptom-based scales. The five-factor
model can provide a salient framework for contextualizing clients' pre-
senting problems; different personality types may be prone to experience
certain types of predicaments. Therefore, merely noting the particular dif-
ficulties that clients experience without appreciating their motivational
basis may undermine not only one's clinical understanding, but the effec-
tiveness of any therapeutic intervention.
All the information presented in this section shows that the NEO PI-R
can be a very useful clinical instrument. Psychometrically sound in a clin-
ical milieu, the NEO scales can provide not only a tremendous amount of
information about the personological context of a client, but can also help
organize information in clinically relevant ways. Issues about differential
diagnosis, intrapsychic dynamics, and treatment prognosis can all be
 PSYCHOMETRIC OVERVIEW 71

linked together. As Miller (1991) has noted, "the five-factor model can re-
late patient personality, presenting complaint, treatment plan, and treat-
ment outcome to each other in a reasonable, systematic way, without loss
of empathy or compassion for the patient" (p. 432).

ROBUSTNESS OF THE FNE-FACTOR MODEL

The data presented throughout this chapter clearly show that the five-
factor model of personality is a sound paradigm for assessing individual
differences. Although originally derived from the English language, the
constructs represented by this model are reflected in numerous instru-
ments that were derived from diverse theoretical orientations. The NEO
PI-R provides a psychometric articulation of these constructs that is quite
robust, useful in describing both normal and abnormal groups. However,
the model is not without its critics (J. Block, 1995; McAdams, 1992).
J. Block (1995) criticized the five-factor model and NEO PI-R for their
non theoretical basis; the model and the instrument were founded in fac-
tor analysis. Block believed that the model, as a statistical tool, is unable to
capture the subtleties of personality. Further, as with any statistical analy-
sis, what you put into it very much determines what you pull out. If you
put in sufficient information for five dimensions, then the factor analysis
will retrieve only five dimensions. Broadening the inclusion criteria may
result in the retrieval of additional factors. There are those who believe that
there are more than five factors (e.g., Benet & Waller, 1995; Hahn & Com-
rey, 1994; Hogan & Hogan, 1992; Hough, 1992; Tellegen, 1993). Thus, J.
Block (1995) believed that the five-factor model does not provide as com-
prehensive a sampling of personological content as it argues.
McAdams (1992) saw a number of limitations to the five-factor model.
The bottom line here is that because of its trait-based nature, the five-fac-
tor model provides a shallow and superficial description of personality.
McAdams believed that the five-factor model is essentially a "psychology
of the stranger"; it provides a broad-band description of people that is use-
ful when you do not know anything else about them.
Critics of the five-factor model argue that, at best, this paradigm cap-
tures superficial, relatively descriptive aspects of the individual that, taken
at face value, can provide some useful information. However, the model
lacks depth and interpretive insight. It overlooks many of the more im-
portant qualities of the individual that drive and direct the ongoing course
of a person's life. In response to some of these issues, the next sections at-
tempt to demonstrate the value and depth of the model. Rather than being
surface descriptors, the five dimensions reflect fundamental dispositions
72 CHAPTER 2

of an individual to think, act, and feel. The importance of these qualities is

found in the fact that they are genetically based and have also been found
to be cross-culturally valid. The final section outlines McCrae and Costa's
(1996) theoretical conceptualization of the five-factor model-its place in
a larger meta theoretical framework.

HERITABILITY

The field of behavior genetics attempts to discover the degree to which

the observed psychological qualities of individuals are linked to one's ge-
netic makeup:- Research in this area has shown that between 25% and 50%
of the variance in observed personality variables is linked to genetic factors
(Bouchard & McGue, 1990; Hershberger, Plomin, & Pedersen, 1995; Plomin,
Chipuer, & Loehlin, 1990). These data suggest that personality traits are not
mere fictions in the eyes of observers or the product of an individual's own
implicit personality theory. Rather, traits spring from the person's biologi-
cal foundation and hence represent a more "objective" reality.
Research on the dimensions of the five-factor model have shown that
they possess significant levels of heritability (Bergeman et al., 1993; Heath
et al., 1992). Jang, Livesley, and Vernon (1996b) evaluated the genetic basis
of all of the NEO PI-R scales. Heritability coefficients from 41% (Neuroti-
cism) to 61 % (Openness) were found. Interestingly, the genetic component
varied tremendously over the various facet scales, with Order, Self-Disci-
pline, and Deliberation largely determined by environmental influences.
Research has shown that genetics contribute to the stability of per-
sonality in adulthood (Jang, Livesley, & Vernon, 1996a; Viken et al., 1994).
Thus, the observed consistency in personality after age 30 cannot be re-
duced to mere consistency in self-presentation. Rather, stability of tem-
perament reflects the emergence of underlying neurobiological processes
that guide the ongoing course of development.
Given the biological basis of the five-factor model, it is difficult to in-
terpret these factors as mere statistical artifacts; these are qualities that will
certainly emerge from any comprehensive description of individuals. Fur-
ther, given their biological nature, it seems equally unlikely that these di-
mensions reflect mere superficial aspects of the individual. It is unlikely
that such peripheral features, with limited motivational implications,
would become genetically imprinted into the species. Rather, the heri-
tability of traits argues for their recognition as important qualities of the in-
dividual that serve to enhance one's ability to adapt and survive (Buss,
1991b). Not mere summary descriptions of behavior, traits represent un-
derlying sources of motivation that organize and direct the ongoing course
of human development and activity.
 PSYCHOMETRIC OVERVIEW 73

If there is a genetic basis to personality it also stands to reason that these

dispositions should find universal expression in the human experience. The
next section shows that the dimensions of the five-factor model in general,
and NEO PI-R in particular, are also readily recoverable in diverse cultures.

CROSS-CULTURAL GENERALIZABILITY

The five-factor model finds its origins in the English language. These
dimensions certainly represent important personality qualities for those
individuals who speak English and share in its derivational experiences.
This leaves the door open for the criticism that these dimensions are cul-
turally specific and may not even generalize to other cultural contexts. If
this is so, then the five-factor model represents circumspect, superficial
qualities of the individual. However, if these dimensions do generalize
cross-culturally, then this would be strong evidence that these dimensions
address human functioning at a much more basic, fundamental level.
In addressing issues of cross-cultural generalizability, there are two
major approaches: emic and etic. The first, ernie, refers to identifying cul-
ture-specific constructs. The second, etie, is concerned with identifying
similarities among cultures, to identify universal aspects of human func-
tioning. Research has shown that the dimensions of the five-factor model
represent discernible constructs in a variety of societies and can be useful
for understanding culture-specific phenomena (Capara, Barbaranelli, Bor-
gogni, & Perugini, 1993; Heaven et at., 1994; John, Goldberg, & Angleitner,
1984; Paunonen et at., 1992).
Numerous researchers have shown that the dimensions of the five-
factor model generalize quite well to a number of different cultures, in-
cluding European (Italian-Capara et al., 1993; Capara, Barbaranelli, &
Comrey, 1995; German-Borkenau & Ostendorf, 1990; Finnish and Polish-
Paunonen et at., 1992; Spanish-Avia et at., 1995), Indian (Narayanan,
Shanka, & Levine, 1995), and Asian (Japanese-Bond, Nakazato, & Shi-
raishi, 1975, Isaka, 1990; Chinese-Bond, 1979; Korean-Piedmont & Chae,
1997; Filipino-Katigbak, Church, & Akamine, 1996). Research continues to
document the presence of these five factors in both self-report and rater
data. It is of particular interest to note that the five-factor model is recov-
erable in languages that do not share a common derivational or experien-
tial history with English (see McCrae & Costa, 1997). The diversity of
cultures that have developed and applied personality dispositions identi-
cal to Western-based ones is exciting evidence of the unity of human psy-
chosocial functioning.
An even more exciting fact has been the finding that these domains
operate maturationally in similar ways across cultures (McCrae, Costa,
 74 CHAPTER 2

Piedmont, et ai., 1996). It has long been known that adolescents and young
adults (ages 17-21) have higher levels of Neuroticism, Extraversion, and
Openness than older (age 30 and over) adults and lower levels of Agree-
ableness and Conscientiousness. There are real changes in personality that
occur over the early adult years. These maturational changes have also
been noted in several other cultures where the NED PI-R has been trans-
lated (e.g., Korea, Italy, and Croatia). Finding similar developmental pat-
terns across diverse cultures mutes criticisms that these factors are merely
structural artifacts that emerge from the questionnaires themselves. If this
were so, it would be unlikely that changes in the levels of these domains
over the life span would follow similar forms in different cultures.
Rather than superficial descriptive qualities or statistical artifacts, the
five-factor model represents broad, generalizable qualities of individuals.
Measuring people on these constructs enables one to anticipate important
behavioral outcomes in a variety of cultures. These data support the view
that there is tremendous parsimony to human personality. That much of
personality can be described by a finite set of constructs sets the stage for
the development of a more inclusive and complete understanding of indi-
vidual differences.

THEORETICAL FOUNDATION
McCrae and Costa (1996) have outlined a general theoretical frame-
work within which they place the five-factor model of personality. The
value of this model is that it attempts to put forth the underlying, motiva-
tional role these five dimensions play in shaping personality. Although they
note that their model is only one of a large number of theories that are con-
sistent with research on the five-factor model, its value is that it provides a
series of testable hypotheses that can help extend research in this area.
There are five components of this model and they are outlined in Figure
2-1, The first is Basic Tendencies. This refers to the fundamental raw material of
personality, the "stuff" with which we are born and from which the individ-
ual emerges. This material is genetically received and not shaped by envi-
ronmental forces (e.g., Plomin & Daniels, 1987). Rather, it is internally driven
through its development, with final form reached in adulthood. The content
of these basic tendencies includes the five broad dimensions of personality:
Neuroticism, Extraversion, Openness, Agreeableness, and Conscientious-
ness. These domains are hierarchically organized, with more specific facets
defining more specific levels of organization. There are other aspects of func-
tioning located here as well, including physical characteristics (e.g., physical
appearance, gender, health), cognitive capacities (e.g., perceptual styles, gen-
eral intelligence), and physiological drives (e.g., need for food and oxygen).
 PSYCHOMETRIC OVERVIEW 75

Objective
Biography
External
Inftuences

Cullund_.
LIIe _ _:
SlluetJon
B.slc
Tendencle. hi:!:=~

NIIIIIOIiciIm.
Extra-*,.
Openneu.
~.

-
CCJIIIC:ienIIcMMu

.
FIGURE 2-1. General Theoretical Framework for the Five-Factor Model of Personality.
Note. From "Toward a New Generation of Personality Theories: Theoretical Contexts for the
Five-Factor Model," by R. R. McCrae and P. T. Costa, Jr., 1996. In J. S. Wiggins (Ed.), The Five
Factor Model cfPersonality: Theoretical Perspectives (p. 73), New York: Guilford Press. Copyright
1996 by Guilford Press. Reproduced with permission.

The second component is referred to as Characteristic Adaptations. This

component comprises the acquired skills, abilities, habits, and attitudes
that emerge out of the interaction between the individual and his or her
environment. It is here that the basic tendencies find their specific mani-
festations. Characteristic Adaptations include specific competencies, such
as language, social skills, and religious beliefs and attitudes. An important
element of this member is the third component-Self-Concept, the implicit
and explicit views of the self, personal valuations of worth and value, our
sense of identity. This mechanism contains an individual's self-schema,
that particular affective-cognitive view of the self that is available to con-
sciousness. This framework provides for a selective perception of one's
environment and explains the unique ways that individuals orient them-
selves to the world.
It is important to realize this distinction between Basic Tendencies and
Characteristic Adaptations. The former reflects the basic, broad qualities of
human personality that are common to all people. As these qualities
emerge, the individual refines their expression so that they are most adap-
tive to the specific demands of the immediate context. For example, differ-
ent people, in different cultures may exhibit a wide variety of habits, tastes,
 76 CHAPTER 2

and preferences, all reflecting the specifics of differing environments.

However, it is the same set of needs that is driving these behaviors.
The third component of this model is the Objective Biography. This con-
sists of every significant thing that a person has ever done. It contains the
reactions one has to life events, the life paths followed, the life outcomes
experienced. It is, in a sense, the testament of one's outward life. The vari-
ables in this category are usually seen as the outcomes that psychologists
wish to predict. These events are the product of multiple sources of deter-
mination; factors from both within the individual and without converge to
shape the direction and content of his or her life. Such variables would in-
clude wheth& or not one graduated college, got married, was divorced,
the number of children one had, number of hospitalizations, and so on.
The fourth component of the model is External Influences, and refers to
one's psychological environment. It includes such factors as parent-child
interactions, peer socialization, and one's cultural, socioeconomic, and
family groups. These outer forces exert influences on the direction and
pace of development. However, individuals also select their environments
and influence them in significant ways (e.g., Bandura, 1989; Hartmann,
1939/1958; M. Snyder, 1983). Thus there is an important reciprocity be-
tween the individual and the environment; people create situations that
enable them to express and gratify important personal needs. The form
and manner of this expression is certainly impacted by the external forces
that are present.
The final component of McCrae and Costa's (1996) model is Dynamic
Processes. This refers to the ways each of the above components interact
with one another. Such processes include perceptive capacities, operant
conditioning, coping and defense mechanisms, hedonic adaptation, inter-
personal processes, and identity formation. This component highlights the
fact that the person is an active, emergent property, always in a state of
movement, development, and continual adaptation. The life process is a
trajectory through time; the individual is never in the same place.
There are two types of dynamics postulated here. The first refers to
universal dynamics, and reflects the fact that the "ongoing functioning of
the individual in creating adaptations and expressing them in thoughts,
feelings, and behaviors is regulated in part by universal cognitive, affec-
tive, and volitional mechanisms" (McCrae & Costa, 1996, p. 75). The other
is differential dynamics, and refers to the reality that "[s]ome personality
processes are differentially affected by basic tendencies of the individual,
including personality traits" (p. 75).
This model provides for a rough sketching of how and where the five-
factor model finds a home in the larger schema of personality organiza-
tion. It is important to point out here that the five-factor model is not seen
PSYCHOMETRIC OVERVIEW 77

as a mere summary description of personality traits. Rather, these dimen-

sions serve as basic temperaments that impact the entire enterprise of per-
sonality development and adaptation. The five factors take motivational
precedence over all endeavors, although how these qualities become ex-
pressed in particular behaviors is certainly moderated both by external cir-
cumstances and by other intrapersonal mechanisms (e.g., self-concept).
Therefore, measuring these five domains and their more specific facets en-
ables one to anticipate a wide range of important life outcomes. It is not
surprising, then, that these qualities have been found cross-culturally and
are relevant for understanding behavior in diverse societies.
Yet it needs tObe mentioned that measuring the five factors does not
provide a complete description of personality. There are many other as-
pects that are of value and need to be assessed. As the model just presented
noted, one's characteristic adaptations and self-concept are other impor-
tant aspects of functioning that have an important role to play. Under-
standing a person's life tasks and goals (e.g., Cantor et al., 1991) can
provide a real texture to the nuances of the individual's life. These con-
structs are predictive of important outcomes in their own right, even
though they are related to one's basic tendencies (Cantor, 1990).
Nonetheless, as the information in this chapter has shown, measuring
the five domains of personality provides a useful, predictively powerful
way of describing personality at a broad level. The NEO PI-R can provide
useful information that can help the test user anticipate a wide range of life
outcomes with a relatively high degree of predictive power. The next chap-
ter provides more interpretive information about the scales and begins to
demonstrate how scores on the NEO PI-R can be used to gain useful in-
sights into others.
 CHAPTER 3

INTERPRETING THE NEO PI-R

Psychological test interpretation is an art. It requires of the test evaluator

solid grounding in personality theory, an appreciation of the construct va-
lidity of the test being used, and a firm understanding of psychometrics.
All of this information is frequently combined in ways that defy empirical
quantification. No matter how advanced or skilled the interpreter, if the
quality of the data for analysis is poor or invalid, then no amount of psy-
chological acumen can create a valid interpretation. Therefore, before be-
ginning to learn NEO PI-R interpretation, some consideration of profile
invalidity and its detection is in order. The next section will first consider
the role and value of validity scales for assessing profile validity.

THE USE OF VALIDITY SCALES IN ASSESSMENT

One of the most important undertakings in the enterprise of test de-

velopment is the need for the test maker to demonstrate whether scores on
his or her scale actually reflect important, identifiable personological con-
structs or are merely the by-product of response distortions, intended or
otherwise. Great effort continues to be expended in order to develop scales
aimed at second-guessing the response patterns of individuals on self-
reports (e.g., Arbisi & Ben-Porath, 1995). The goal of these measures is to
alert test interpreters to the possible presence of various response distor-
tions that may be operating to invalidate the protocol. Such distortions can
be as benign as a subject taking the test process for granted by responding
randomly to items, or as serious as deliberate attempts by a respondent to
manipulate his or her psychological presentation to maximum advantage.
Such "validity scales" have been widely embraced by test users, and their

79
80 CHAPTER 3

presence in a test protocol can be quite reassuring (see Ben-Porath &

Waller, 1992) if not absolutely necessary (Butcher & Rouse, 1996). The use
of validity scales does raise several important questions: Do such scales
provide sufficient information to justify the inclusion of additional test
items, many of which can appear quite odd to test takers? Are they effec-
tive in identifying individuals who may be distorting their responses? Do
the adjustments made to test scores based on these scales actually improve
the validity of the measure?
Answers to these questions are important because the developers of
the NEO PI-R explicitly excluded any type of validity scales. Costa and
McCrae (199Tc) believed that such items do not serve any empirically de-
termined role in testing. Here we examine some issues related to the use of
validity scales in personality assessment.

SOCIAL DESIRABILITY

Perhaps one of the more influential books written in the area of per-
sonality assessment was Edwards's (1957) tome on social desirability. His
argument that responses on personality questionnaires were influenced to
a large extent by the social desirability of the item rather than by any per-
sonological content was quickly embraced by the measurement field and
has had a lasting impact. The Edwards Personal Preference Schedule
(EPPS; Edwards, 1959), with its aim of eliminating biasing effects due to
social desirability, represented a major shift in test construction. Even
today, many test developers continue to find it noteworthy to demonstrate
their scales' independence from this bias by correlating scores on their
scale with measures of social desirability (Linehan & Nielsen, 1983).
One of the first efforts at uncorking the genie in the social desirability
bottle surrounded the MMPI. Edwards and Heathers (1962) argued that
the first factor to emerge from factor analyses of the MMPI represented a
social desirability factor. The implication of this assertion was that the ma-
jority of explainable variance in MMPI responses was attributable to re-
sponse distortions rather than to any kind of substantive psychological
phenomenon. However, the specter of the MMPI as providing nothing
more than systematic error was laid to rest with J. Block's (1965) cogent
and empirically compelling analysis of the problem. He was able to
demonstrate that scores on the first factor did, in fact, represent charac-
terological aspects of the individual.
Nonetheless, the MMPI possesses a validity scale, K, that is designed to
detect socially desirable responding by test takers. Several of the content
scales (Hs, Pd, Pt, Sc, and Ma) are mathematically adjusted, or corrected, for
K. The logic of this process is that mathematically partialing out social de-
 INTERPRETING THE NEO PI-R 81

sirability from a test score increases its validity. However, such an assump-
tion does not seem to be empirically justified. McCrae et al. (1989) have
shown that K corrected MMPI scores are less valid than their uncorrected
versions suggesting that rather than being only a suppressor variable, so-
cial desirability represents a substantive aspect of personality. Anumber of
studies have shown that adjusting measures of normal personality for so-
cial desirability will compromise rather than enhance the scale's validity
(Hsu, 1986; Kozma & Stones, 1987; McCrae, 1986; McCrae, & Costa, 1983b;
Piedmont, McCrae, & Costa, 1992; Silver & Sines, 1962). Although the issue
of social desirability as an index of response distortion has been largely re-
futed (see Nicholson & Hogan, 1990), interest remains keen among psy-
chometricians to develop measures better suited for detecting response
distortions. Instead of developing items whose content would reflect ten-
dencies to manipulate one's self-presentation, efforts have focused on "con-
tent-free" validity scales. Such measures rely on response styles rather than
on item content and represent the next generation of validity scales.

CONTENT-FREE VALIDITY SCALES

Early validity scales were themselves measures aimed at assessing

qualities of the test taker. However, the qualities being assessed were char-
acteristics related to a tendency to distort one's responses on other content-
relevant scales. For example, Social Desirability scales contain items such
as, "I always try to practice what I preach," and "I have almost never felt
the urge to tell someone off" (Crowne & Marlowe, 1960) which reflect a
high need for approval. Such a need to be seen in a good light was thought
to dispose one to distort responses on other scales. But as shown above,
measures of social desirability are confounded with personality. Need for
approval is itself a substantive individual-difference variable.
In response to this, researchers began looking for other methods that
were "content-free," scales that would identify aberrant responding with-
out having to rely on items that may reflect st~ble qualities of the re-
sponder. One popular approach is to evaluate the consistency with which
subjects respond to items (Tellegen, 1982). Tellegen (1988) outlines how
such scales are developed and applied. At the heart of this approach is the
belief that individuals will respond consistently to test items having simi-
lar content. Two such scales that are frequently used are the Variable Re-
sponse Inconsistency (VRIN) and True Response Inconsistency (TRIN)
scales. These scales have been developed for use on a number of major
psychological tests (e.g., the Multidimensional Personality Questionnaire
[MPQ1 and the MMPI-2). VRIN scales consist of pairs of items with content
that varies greatly from pair to pair but that is quite homogenous within
82 CHAPTER 3

each pair if both items are scored in the same direction. For example, one
such pair from the MPQ is "When I work with others I like to take charge"
and "I usually do not like to be a 'follower.'" Responding True to one item
and False to the other item in the pair would be scored as one inconsistent
response. High scores on the VRIN scale would indicate an individual
who is responding to the questions in an indiscriminate manner.
The TRIN scale is made up of item pairs that are opposite in content.
For example, one such item pair from the MPQ is "I would rather tum the
other cheek than get even when someone treats me badly" and "When
people insult me, I try to get even." Inconsistency is scored when an indi-
vidual answers True to both items or False to both items. High scores on
TRIN indicate a tendency to answer True indiscriminately (an acquies-
cence effect) while low scores indicate a tendency to answer False ("nonac-
quiescence" or "nay-saying"). In both cases, extreme scores would reflect
a protocol that may be invalid, uninterpretable, or both. Tellegen and
Waller (in press) have shown that the VRIN and TRIN scales are uncorre-
lated with each other and maintain low associations with the content
scales of the MPQ. Thus, these scales are unconfounded with substantive
personality and are claimed to be able to identify invalid protocols that
would not be detected by the more familiar validity scales.
Interestingly, even though the NEO PI-R does not contain any valid-
ity scales, several have been developed (Schinka, Kinder & Kramer, 1997;
Goldberg, personal communication, January 25,1995). These scales assess
the degree to which a subject is willing to claim positive and negative qual-
ities as characteristic. Others evaluate the degree of consistency in re-
sponding to the items. Regardless of the content of these validity scales,
their purpose is to identify protocols that are likely to be invalid. Put an-
other way, scores on these validity scales determine the likelihood of a
scale's validity. Scores on the validity scales moderate the validity of the
substantive scales.
In a series of analyses conducted by Piedmont and McCrae (1998), we
evaluated the validity of a number of validity scales, both those developed
for the NEO PI-R and those on the MPQ. From all the analyses that were
done in this study it was determined that the validity scales showed no
consistent, significant pattern of effects. Scores on these response-distor-
tion scales did not moderate the validity of either self-reports or observer
ratings. Their value for use among volunteer, nonclinical samples is seri-
ously questioned. Of course, further research is needed in this area to eval-
uate the utility of validity scales in clinical, forensic, and job selection
samples. Until a complete answer to the value of validity scales is ob-
tained, a reliance on self-report data will continue. In order to use such in-
formation well, some consideration of its value is necessary.
 INTERPRETING THE NED PI-R 83

THE VALUE AND LIMITS OF SELF-REpoRT DATA

The underlying assumption of a self-report is that the individual is in

the best position to provide information about him- or herself (e.g., private
beliefs and past behaviors) that may not be readily accessible from ob-
servers or life outcome data sources. There is also a certain parsimony to
the approach: If you wish to know something about people, ask them. Yet
despite such an appealing paradigm, there has always been great skepti-
cism and mistrust of self-report information. Concerns about defensive-
ness, faking, and social desirability have always created a pall over such
procedures and have led to the development of many different types of va-
lidity scales to correct for these sources of error.
However; data do exist that offer support for trust in self-reports. Ev-
idence shows that more direct and "obvious" items possess better validity
than subtle items (Furnham, 1986; Worthington & Schlottman, 1986; Wro-
bel & Lachar, 1982), suggesting that when respondents are presented with
a direct query about their internal state, they will give an honest and accu-
rate response. In perhaps the largest study to date, Hough, Eaton, Dun-
nette, Kamp, and McCloy (1990) concluded that applicants for a job tended
not to distort their responses in the absence of any instructions to do so
(see Mount & Barrick, 1995, for a review). Hogan (1991) argued that the de-
sire by some respondents to manipulate their test image may be an indi-
vidual difference variable itself worthy of study. Jackson (1989) showed
convergent correlations between self-reported scores on the Basic Person-
ality Inventory scales and professional ratings for psychiatric patients.
Muten (1991) showed, with a sample of his patients, significant levels of
agreement between self-reported scores and spouse ratings on the NEO PI.
Taken together, these findings support the trustworthiness of self-report
data. In general, people seem to respond in an honest and straightforward
manner to the testing situation.
These data do not deny that there are occasions when a self-report
may not provide useful data. An uncooperative client or a cognitively im-
paired individual may not be able to provide valid information. There may
also be situations in which an individual has great incentive to distort or
manipulate his or her presentation. In such scenarios, self-report data are
of little use, irrespective of any validity scales or mathematical manipula-
tions of the information. In such cases it would be more beneficial to rely
on other sources of information, such as an observer rating by a friend or
spouse. If you cannot trust a person to provide valid information, then do
not use a self-report measure.
Observer-based data provide a useful counterpoint to self-reports. Al-
though they have their own potential sources of error (e.g., halo effects,
84 CHAPTER 3

stereotypes), these biases do not overlap with the errors that are inherent
to self-reports (McCrae, 1982). Thus, convergence between peer and self
ratings cannot be attributable to correlated error but to a reliable effect (see
Wiggins, 1979). Because self-peer agreement seems to be the rule in mea-
suring the FFM, disagreements between these two sources of information
may be indicative of some type of distortion. McCrae (1994c) provides a
methodology for using both self and observer ratings on the NEO PI-R.
But if a self-report cannot be trusted, observer data can be a very helpful
stand-in. Data do support the validity of peer ratings in predicting useful
outcomes such as job success and marital satisfaction (Kosek, 1996a;
Mount, Barri<j<, & Strauss, 1994).
Believing that self-report data are relatively sound, Costa and McCrae
(1992c) developed the NEO PI-R without any type of validity scales (see
Costa & McCrae, 1992a). The only "validity" scale is the presence of three
short items at the end of the NEO PI-R that ask the subject if he or she has
completed all items in the proper way. Although such direct items may not
capture subtle attempts at distortion, a "No" to any of the items would
clearly indicate a questionable protocol.

NEO PI-R FACET SCALES AND THEIR

INTERPRETATIONS

Recognizing that there are inherent limitations to test data, regardless

of its source, we tum now to a presentation of the NEO PI-R facet scales
and their definitions. Scales will be presented by domain. The facet scales
can be considered partitions of the larger spectrum of qualities represented
in each domain. In developing the facets every effort was made to create
scales that were as nonoverlapping in content as possible while still re-
maining on the same domain. By minimizing redundancy among the
facets, the test developers maximized the interpretability of each scale.

NEUROTICISM

Neuroticism assesses affective adjustment versus emotional instabil-

ity. Individuals who score high on this domain are prone to experiencing
psychological distress, unrealistic ideas, excessive cravings or urges, and
maladaptive coping responses. Although high scores on this domain do
not indicate the presence of any clinical disorder, individuals with a clini-
cal syndrome do tend to have a high score here (see Costa & Widiger,
1994). In fact, high scores on Neuroticism place one at risk for receiving a
psychiatric diagnosis (Zonderman, Herbst, Schmidt, Costa, & McCrae,
 INTERPRETING THE NEO PI-R 85

1993). The six facets for this domain, and th.eir definitions are from Costa
and McCrae (1992c). Adjective descriptors for each facet are taken from a
variety of sources (McCrae & Costa, 1992; Piedmont & Weinstein, 1993).
Nl: Anxiety. Anxious individuals are apprehensive, fearful, prone to
worry, nervous, tense, and jittery. The scale does not measure specific fears
or phobias, but high scorers are more likely to have such fears, as well as
free-floating anxiety. Low scorers are calm and relaxed. They do not dwell
on things that might go wrong. Adjectives that describe high scorers on
this facet include tense, fearful, worried, apprehensive. Adjectives that de-
scribe low scorers on this facet include calm, relaxed, stable, fearless.
N2: Angry lWstility. Angry hostility represents the tendency to expe-
rience anger and related states such as frustration and bitterness. This scale
measures the individual's readiness to experience anger; whether the anger
is expressed depends on the individual's level of Agreeableness. Note, how-
ever, that disagreeable people often score high on this scale. Low scorers
are easygoing and slow to anger. High scorers on this facet are described
as being hot-tempered, angry, and frustrated. Low scorers are described as
being amiable, even-tempered, and gentle.
N3: Depression. This scale measures normal individual differences in
the tendency to experience depressive affect. High scorers are prone to
feelings of guilt, sadness, hopelessness, and loneliness. They are easily dis-
couraged and often dejected. Such individuals are described as being
hopeless, guilty, downhearted, and blue. Low scorers rarely experience
such emotions, but they are not necessarily cheerful and lighthearted-
characteristics associated instead with Extraversion. High scorers on this
facet are described as being seldom sad, hopeful, confident, and as feeling
worthwhile.
N4: Self-Consciousness. The emotions of shame and embarrassment
form the core of this facet of Neuroticism. Self-conscious individuals are
uncomfortable around others, sensitive to ridicule, and prone to feelings of
inferiority. Self-consciousness is akin to shyness and social anxiety. Such
individuals are described as being ashamed, feel inferior, and are easily
embarrassed. Low scorers do not necessarily have grace or good social
skills; they are simply less disturbed by awkward social situations. These
individuals are described as poised, secure, and feel adequate.
N5: Impulsiveness. This facet refers to the inability to control cravings
and urges. Desires (e.g., for food, cigarettes, possessions) are perceived as
being so strong that the individual cannot resist them, although he or she
may later regret the behavior. Low scorers find it easier to resist such temp-
tations, having a high tolerance for frustration. The term impulsive should
not be confused with spontaneity, risk taking, or rapid decision time. High
scorers on this facet are described as being unable to resist cravings, hasty,
86 CHAPTER 3

sarcastic, and self-centered. Low scorers are described as being self-con-

trolled and able to resist temptation.
N6: Vulnerability. The final facet of N is vulnerability to stress. Indi-
viduals who score high on this scale feel unable to cope with stress, be-
coming dependent, hopeless, or panicked when facing emergency
situations. High scorers are characterized as being easily rattled, panicked,
and unable to deal with stress. Low scorers perceive themselves as capable
of handling themselves in difficult situations. These individuals are de-
scribed as being resilient, cool-headed, and hardy.

EXTRAVERSION

Costa and McCrae (1985) have defined this domain as representing the
quantity and intensity of interpersonal interaction, the need for stimulation
and the capacity for joy. This domain contrasts sociable, active, person-ori-
ented individuals with those who are reserved, sober, retiring, and quiet.
There are two qualities assessed on this domain: interpersonal involvement
and energy. The former evaluates the degree to which an individual enjoys
the company of others and the latter reflects the personal tempo and activ-
ity level. This dimension has been shown to capture levels of positive affect.
E1: Warmth. Warmth is the facet of Extraversion most relevant to is-
sues of interpersonal intimacy. Warm people are affectionate and friendly.
They genuinely like people and easily form close attachments to others.
Such individuals are characterized as being outgoing, talkative, and affec-
tionate. Low scorers are neither hostile nor necessarily lacking in compas-
sion, but they are more formal, reserved, and distant in manner than high
scorers. Warmth is the facet of E that is closest to Agreeableness in inter-
personal space, but it is distinguished by a cordiality and heartiness that
is not part of A.
E2: Gregariousness. A second aspect of E is gregariousness-the pref-
erence for other people's company. Gregarious people enjoy the company
of others, the more the merrier. They are characterized as being convivial,
having many friends, and seeking social contact. Low scorers on this scale
tend to be loners who do not seek-or who even actively avoid-social
stimulation. These individuals are described as avoiding crowds and pre-
ferring to be alone.
E3: Assertiveness. High scorers on this scale are dominant, forceful,
and socially ascendant. They speak without hesitation and often become
group leaders. Adjective descriptions of high scorers include dominant,
forceful, confident, and decisive. Low scorers prefer to keep in the back-
ground and let others do the talking. Adjective descriptors include unas-
suming, retiring, and reticent.
 INTERPRETING THE NEO PI-R 87

E4: Activity. A high Activity score is seen in rapid tempo and vigor-
ous movement, in a sense of energy, and in a need to keep busy. Active
people lead fast-paced lives. They are described as being energetic, fast-
paced, and vigorous. Low scorers are more leisurely and relaxed in tempo,
although they are not necessarily sluggish or lazy. They are described by
others as being unhurried, slow, and deliberate.
E5: Excitement-Seeking. High scorers on this scale crave excitement
and stimulation. They like bright colors and noisy environments. Excite-
ment-seeking is akin to some aspects of sensation seeking. These individ-
uals are described as flashy, seekers of strong stimulation, and risk takers.
Low scorers feel little need for thrills and prefer a life that high scorers
might find boring. These individuals are described as cautious, staid, and
uninterested in thrills.
E6: Positive Emotions. This facet reflects the tendency to experience
positive emotions such as joy, happiness, love, and excitement. High scor-
ers on this scale laugh easily and often. They are seen as cheerful, high-
spirited, joyful, and optimistic. Low scorers are not necessarily unhappy;
they are merely less exuberant and high-spirited. They are described as
unenthusiastic, placid, and serious. Research has shown that happiness
and life satisfaction are related to both Nand E, and that Positive Emotions
is the facet of E most relevant to the prediction of happiness.

OPENNESS TO EXPERIENCE

Openness to Experience is defined as the proactive seeking and

appreciation of experience for its own sake, and as toleration for and ex-
ploration of the unfamiliar. This domain contrasts curious, original, untra-
ditional, and creative individuals with those who are conventional,
unartistic, and unanalytical. Of all the domains, this one is the most con-
troversial; it is the least developed and explored. In terms of its represen-
tativeness in the language, it has the fewest number of descriptors.
Nonetheless, Openness continues to show its personological value (Mc-
Crae & Costa, 1985b; McCrae, 1990, 1993-1994, 1994b). The facets for this
domain are as follows:
01: Fantasy. Individuals who are open to fantasy have a vivid imag-
ination and an active fantasy life. They daydream not simply as an escape
but as a way of creating for themselves an interesting inner world. They
elaborate and develop their fantasies and believe that imagination con-
tributes to a rich and creative life. These individuals are described as imag-
inative and as enjoying daydreaming. Low scorers are more prosaic and
prefer to keep their minds on the task at hand. They are described as prac-
tical and as preferring realistic thinking.
 88 CHAPTER 3

02: Aesthetics. High scorers on this scale have a deep appreciation for
art and beauty. They are moved by poetry, absorbed in music, and in-
trigued by art. They need not have artistic talent, nor even necessarily
what most people would consider good taste, but for many of them, inter-
est in the arts will lead them to develop a wider knowledge and apprecia-
tion than the average individual. They are seen as valuing aesthetic
experiences and as moved by art and beauty. Low scorers are relatively in-
sensitive to and uninterested in art and beauty. They are described by oth-
ers as being insensitive to art and unappreciative of beauty.
03: Feelings. Openness to feelings implies receptivity to one's own
inner feelings and emotions and the evaluation of emotion as an important
part of life. High scorers experience deeper and more differentiated emo-
tional states and feel both happiness and unhappiness more intensely than
others. Descriptions of high scorers include emotionally responsive, sen-
sitive, empathic, and values own feelings. Low scorers have somewhat
blunted affects and do not believe that feeling states are of much impor-
tance. Descriptions of low scorers include narrow range of emotions and
insensitive to surroundings.
04: Actions. Openness is seen behaviorally in the willingness to try
different activities, go new places, or eat unusual foods. High scorers on
this scale prefer novelty and variety to familiarity and routine. Over time,
they may engage in a series of different hobbies. They are described by
others as seeking novelty and variety and trying new activities. Low scor-
ers find change difficult and prefer to stick with the tried and true. These
individuals are perceived by others as being set in their ways and prefer-
ring the familiar.
05: Ideas. Intellectual curiosity is an aspect of Openness that has long
been recognized. This trait is seen not only in an active pursuit of intellec-
tual interests for their own sake, but also in open-mindedness and a will-
ingness to consider new, perhaps unconventional ideas. High scorers enjoy
both philosophical arguments and brain-teasers. Openness to ideas does
not necessarily imply high intelligence, although it can contribute to the
development of intellectual potential. These individuals are described as
being intellectually curious, analytical, and theoretically oriented. Low
scorers on the scale have limited curiosity and, if highly intelligent, nar-
rowly focus their resources on limited topics. These individuals are de-
scribed as being pragmatic, factually oriented, and unappreciative of
intellectual challenges.
06: Values. Openness to Values means the readiness to reexamine so-
cial, political, and religious values. High scorers on this facet are seen as
tolerant, broad-minded, nonconforming, and open-minded. Closed indi-
viduals tend to accept authority and honor tradition and as a consequence
 INTERPRETING THE NED PI-R 89

are generally conservative, regardless of political party affiliation. Low

scorers on this facet are seen as dogmatic, conservative, and conforming.
Openness to Values may be considered the opposite of dogmatism.

AGREEABLENESS

Extraversion evaluates the degree to which a person enjoys being in

the presence of others. Agreeableness examines the attitudes an individual
holds toward other people. These attitudes can be very pro-person, com-
passionate, trusting, forgiving, and soft-hearted on one end to very antag-
onistic, cynical, manipulative, vengeful, and ruthless on the other. The
broad interpersonal orientation captured here ranges from very Mother
Teresa-ish on the one hand to Machiavellian on the other. The facets for
this domain include the following:
AI: Trust. High scorers have a disposition to believe that others are-
honest and well intentioned. High scorers are characterized as being for-
giving, trusting, and peaceable. Low scorers on this scale tend to be cynical
and skeptical and to assume that others may be dishonest or dangerous.
Low scorers are characterized as being wary, pessimistic, suspicious, and
hard-hearted.
A2: Straightforwardness. High scorers on this scale are frank, sincere,
and ingenuous. These individuals are characterized as being direct, frank,
candid, and ingenuous. Low scorers on this scale are more willing to ma-
nipulate others through flattery, craftiness, or deception. They view these
tactics as necessary social skills and may regard more straightforward peo-
ple as naive. These individuals are described as being shrewd, clever, and
charming.
A low scorer on this scale is more likely to stretch the truth or to be
guarded in expressing his or her true feelings, but this should not be inter-
preted to mean that he or she is a dishonest or manipulative person. In par-
ticular, this scale should not be regarded as a lie scale, either for assessing
the validity of the test itself, or for making predictions about honesty in
employment or other settings.
A3: Altruism. High scorers on this facet have an active concern for
others' welfare as shown in generosity, consideration of others, and a will-
ingness to assist others in need of help. These individuals are seen by oth-
ers as being warm, soft-hearted, gentle, generous, and kind. Low scorers
on this scale are somewhat more self-centered and are reluctant to get in-
volved in the problems of others. These individuals are seen by others as
being selfish, cynical, cold, and snobbish.
A4: Compliance. This facet of A concerns characteristic reactions to in-
terpersonal conflict. The high scorer tends to defer to others, to inhibit ag-
90 CHAPTER 3

gression, and to forgive and forget. Compliant people are meek and mild.
High scorers are characterized as being deferential, obliging, and kind. The
low scorer is aggressive, prefers to compete rather than cooperate, and has
no reluctance to express anger when necessary. Low scorers are character-
ized as being stubborn, demanding, headstrong, and hard-hearted.
AS: Modesty. High scorers on this scale are humble and self-effacing
although they are not necessarily lacking in self-confidence or self-esteem.
These individuals are perceived by others as being humble and unassum-
ing. Low scorers believe they are superior people and may be considered
conceited or arrogant by others. A pathological lack of modesty is part of
the clinical conception of narcissism. These individuals are seen by others
as being aggressive, tending to show off, and tough.
A6: Tender-Mindedness. This facet scale measures attitudes of sympa-
thy and concern for others. High scorers are moved by others' needs and
emphasize the human side of social policies. Adjective descriptors of high
scorers include friendly, warm, kind, gentle, and soft-hearted. Low scorers
are more hardheaded and less moved by appeals to pity. They would con-
sider themselves realists who make rational decisions based on cold logic.
Adjective descriptors of low scorers include intolerant, cold, opinionated,
and snobbish.

CONSCIENTIOUSNESS

This domain assesses the individual's degree of organization, persis-

tence, and motivation in goal-directed behavior. This dimension contrasts
dependable, fastidious people with those who are lackadaisical and
sloppy. Also represented here is the amount of personal control and the
ability to delay gratification of needs. The facets include the following:
Cl: Competence. This facet refers to the sense that one is capable, sen-
sible, prudent, and effective. High scorers on this scale feel well prepared
to deal with life. These individuals are perceived by others as being effi-
cient, thorough, confident, and intelligent. Low scorers have a lower opin-
ion of their abilities and admit that they are often unprepared and inept.
These individuals are perceived by others as being confused, forgetful, and
frivolous. Of all the C facet scales, competence is most highly associated
with self-esteem and internal locus of control.
C2: Order. High scorers on this scale are neat, tidy, and well orga-
nized. They keep things in their proper places. Adjective descriptors for
high scorers include precise, efficient, and methodical. Low scorers are un-
able to get organized and describe themselves as unmethodical. Adjective
descriptors for low scorers include disorderly, impulsive, and careless.
Carried to an extreme, high Order might contribute to a Compulsive Per-
sonality Disorder.
 INTERPRETING THE NED PI-R 91

C3: Dutifulness. In one sense, conscientiousness means "governed by

conscience," and that aspect of C is assessed as Dutifulness. High scorers
on this scale adhere strictly to their ethical principles and scrupulously ful-
fill their moral obligations. These individuals are described as being de-
pendable, mannerly, organized, and thorough. Low scorers are more
casual about such matters and may be somewhat undependable or unreli-
able. These individuals are described as being lazy, absent-minded, and
distractible.
C4: Achievement Striving. Individuals who score high on this facet
have high aspiration levels and work hard to achieve their goals. They are
diligent and purposeful and have a sense of direction in life. These indi-
viduals are seen as being ambitious, industrious, enterprising, and per-
sistent. Very high scorers, however, may invest too much in their careers
and become workaholics. Low scorers are lackadaisical and perhaps even
lazy. They are not driven to succeed. They lack ambition and may seem
aimless, but they are often perfectly content with their low levels of
achievement. These individuals are seen as being leisurely, dreamy, and
disorganized.
C5: Self-Discipline. This term means the ability to begin tasks and
carry them through to completion despite boredom and other distractions.
High scorers have the ability to motivate themselves to get the job done.
Adjective descriptors include organized, thorough, energetic, capable, and
efficient. Low scorers procrastinate in beginning chores and are easily dis-
couraged and eager to quit. Adjective descriptors include unambitious,
forgetful, and absent-minded.
Low self-discipline is easily confused with impulsiveness. Both are
evidence of poor self-control, but empirically they are distinct. People high
in impulsiveness cannot resist doing what they do not want themselves to
do; people low in self-discipline cannot force themselves to do what they
want themselves to do. The former requires an emotional stability; the lat-
ter, a degree of motivation that they do not possess.
C6: Deliberation. The final facet of C assesses the tendency to think
carefully before acting. High scorers on this facet are cautious and deliber-
ate. High scorers are described as being cautious, logical, and mature. Low
scorers are hasty and often speak or act without considering the conse-
quences. At best, low scorers are spontaneous and able to make snap deci-
sions when necessary. These individuals are described as being immature,
hasty, impulsive, and careless.
The 35 scales of the NEO PI-R (five domain scores and the 30 facets)
provide a comprehensive assessment of the individual. There is certainly a
tremendous amount of information available for interpretation. As was
seen in the previous chapter, this information has implications for a wide
range of psychologically significant outcomes. The next section provides
92 CHAPTER 3

some examples in order to help the reader begin to work interpretively

with this information.

NEO PI-R INTERPRETATIONS AND

SELECT CASE PROFILES

Now that definitions of the 30 NEO PI-R facets has been provided,
this section presents three case profiles for interpretation. This provides
the reader an opportunity to apply these definitions to actual scores and
begin to develop a "feel" for how individuals come to be represented by
these scales.
In evaluating a NEO PI-R profile, I find it useful to proceed domain
by domain. I first evaluate a person's standing on the overall domain, say
Neuroticism, to gain an overview of the kinds of global dynamics that
characterize the person. Then I evaluate the facet scales for that domain in
order to isolate those aspects of the larger domain that are the most defin-
ing. Keep in mind that although the facet scales for a domain are all re-
lated, they were designed to be as nonredundant as possible (see Costa &
McCrae, 1995a). This can result in a high degree of interscale scatter within
each personality dimension. It is not uncommon for an individual to score
high (or low) on a domain, but be low (or high) on several of the facet
scales. Facet analysis can be very revealing and provide a more intimate
understanding of a client (Miller, 1991).
In evaluating the elevation of scores on each scale, T-scores between
45 and 55 are considered average or normative. There is usually no real in-
terpretation of such scores, because the individual can be equally likely to
exhibit behaviors characteristic of the high and low poles. T-scores below
45 are considered low and scores below 35 are seen as being very low. Con-
versely, T-scores above 55 are seen as being high and those above 65, very
high. When scores move into these regions, they carry interpretive value;
a respondent begins to reflect more conSistently the characteristics defin-
ing that end of the pole. When T-scores for a facet or domain go above 80
or below 20, this may indicate the presence of real deficits in an individ-
ual's ability to function in his or her environment. Therefore, one needs to
consider the real possibility of the presence of some pathological process.
The 35 scales of the NEO PI-R present a tremendous amount of infor-
mation to process. This is both good and bad. It is good because of the
vastness of insights revealed by the scales. There are many personological
implications for scores' from each scale, presenting a rather fine-grained
analysis of the respondent. The down side is really not that bad. So much
information means that it will take time to develop a working knowledge
 INTERPRETING THE NEO PI-R 93

of all the facet and domain scales and what they represent about a client.
Leaming the 30 facet names can be its own challenge. With practice, how-
ever, one can quickly master this task and become quite fluent in the lan-
guage of the NEO PI-R. Aside from facilitating interpretation, the
personality qualities represented by the NEO PI-R will also provide a lan-
guage for conceptualizing and evaluating other personality constructs and
scales. With this in mind, we turn now to interpreting some case profiles.
The cases presented are of actual individuals that have been culled
from several sources. Some are of actual clinical clients; others are from
data files accumulated by the author over several years. Every effort was
made to select indiYiduals with particularly interesting psychosocial his-
tories and profile interpretations.

CASE HISTORY: DEBBIE K.

The first profile is of a 17-year-old White female named Debbie K. At
the time of testing she was a senior in an all-girl parochial high school.
She was in the top of her class academically, having already received sev-
eral academic scholarships to many large, well-known universities. Over
the course of her high school experience, she had served in numerous
leadership positions, including class president. Her profile is presented in
Figure 3-1.
As can be seen, her overall Neuroticism score is very low, suggesting
a young lady who is emotionally stable and well adjusted. All her facet
scales for N are in the low to very low levels. She is portrayed here as being
calm, amiable, seldom sad, secure, self-controlled, and hardy. If she does
experience any negative affect, it may be around feelings of inadequacy
(N4), which is her highest N facet.
On Extraversion, she again scores in the low range, suggesting a more
formal, and reserved "approach to others. On the facet scales there is great
scatter. Debbie scores high on the E3 (Assertiveness) facet, suggesting that
she is confident and dominant. In a group of people she would prefer to be
the leader rather than a follower. She scores in the low range on E4 (Activ-
ity) and E5 (Excitement Seeking). This suggests that Debbie has a very
measured and unhurried personal tempo. There would appear to be little
pressure in her efforts at undertaking various tasks. Her low E5 score in-
dicates someone who is cautious and uninterested in thrills. Her other E
scores are in the average range, suggesting a more flexible orientation to-
ward others. E1 (Warmth) indicates that Debbie can be talkative and out-
going, although there may be some sense of distance and aloofness
experienced by those who may meet her for the first time. Her E2 (Gregar-
iousness) score is on the average to high cusp, and would suggest a degree
 94 CHAPTER 3

J:
.2'
J:
~
Q.ll0·
:> :

J:
.121 ~ :
J:

Q)
g>
Q3: 50·

~ ~

~
..J
~
~

FIGURE 3-1. NED PI-R profile of Debbie K.

of sOCiability and a desire for social contacts, although there is room for
solitary pursuits as well. Her low average score on E6 (Positive Emotions)
indicates a person who appears somewhat serious and may express hap-
piness in a somewhafmuted fashion.
The Openness to Experience domain score is average, suggesting an
individual who has some inner needs for structure and direction as well as
a capacity to explore new avenues of activity. Underlying this person is
some degree of rigidity; there are limits to how much change or novelty
Debbie would tolerate. Again, however, there is quite a degree of scatter
among the facet scales for this domain.
She is high on 01 (Fantasy) and 05 (Ideas). Debbie has an active inner
life composed of imagination and fantasy. No doubt this inner world con-
tributes to and benefits from her fond desire for reading. The characters in
her books are brought to life by her capacity to imagine. Her high score on
05 (Ideas) suggests that Debbie is very much interested in intellectual pur-
suits, theoretical arguments and discussions, and is willing to consider
new and unconventional ideas. These characteristics certainly work well
with her academic abilities and successes.
However, Debbie scores very low on 04 (Actions) and 06 (Values). The
low Actions suggests that Debbie prefers a great deal of structure in her life.
 INTERPRETING THE NEO PI-R 95

There is a preference for the known and a dislike for change. As Debbie once
told me, "1 am pretty set in my ways." Her score on this scale indicates some
degree of personal rigidity. Her low score on 06 (Values) indicates that Deb-
bie also holds some very dear personal beliefs and values that are not open
for discussion, evaluation, or change. Low scores here reflect a dogmatic ori-
entation to a value network that may be difficult to penetrate or to change.
She is accepting of authority and conforms to the dictates of her beliefs.
Debbie's average score on 02 (Aesthetics) represents some enjoyment
and appreciation of the arts. This score reflects her interests in reading, al-
though this scale also reflects interests in other aspects of the arts, such as
theater, ballet, and opera, for which Debbie has far less interest. Her aver-
age score on 03 (Feelings) suggests an individual who has some receptiv-
ity to emotional information and may have some sensitivity to the feelings
of others. However, given her low scores on the Neuroticism facets, Deb-
bie may feel much more comfortable on a cognitive level than on an emo-
tional one. Yet, her score is on the average to low cusp, suggesting that
there may be some emotions that Debbie does not like to confront. Feelings
may be less important to her than ideas.
Concerning Agreeableness, Debbie scores again in the average range.
This indicates an individual who is generally warm and trusting, but has a
degree of skepticism that will not allow her to take all people and events at
face value. She scores high on the Al (Trust), A2 (Straightforwardness),
and A3 (Altruism) scales, suggesting an individual who presents herself to
the world in a frank and "up front" manner and believes that others do the
same. She has a desire to help others in distress and can be very generous
and considerate. Debbie scored Iowan AS (Modesty), indicating that she
has no problems in talking about her achievements and may have a feeling
of superiority because of them. This low score in conjunction with her low
score on Values (06) suggests that Debbie may come across to others as
being arrogant or superior. She may project, in word and deed, a strong
sense of righteousness that others may find off-putting.
She scored in the average range on A4 (Compliance) and A6 (Tender-
Mindedness). A4 indicates that although Debbie can be deferential and
compliant, she can also be competitive and confrontational; when pushed,
she may push back. Her score on A6 indicates that she may not always be
moved by emotional appeals from others. Although her A3 score says she
is helpful, such assistance may not always be forthcoming, especially if she
believes that the person may have brought the problem on him- or herself.
Finally, her Conscientiousness score is in the very high range, sug-
gesting that Debbie is a very organized woman with a strong sense of per-
sonal organization and competency. She sets very high standards for
herself and strives hard to reach them. All of her facet scores for this do-
 96 CHAPTER 3

main are in the high to very high range indicating that Conscientiousness
is a broad-based definer of Debbie's personality. Cl (Competence) is very
high, indicating that Debbie believes herself to be very capable and com-
petent, and that this sense of self-efficacy generalizes to any context she
may find herself in. C2 (Order) shows her to be very organized and me-
thodical. C3 (Dutifulness) shows her to be dependable and organized; she
will follow through on her commitments. C4 (Achievement Striving)
shows that Debbie sets very high standards for herself and will work very
hard to attain them. C5 (Self-Discipline) shows her to be focused in her ef-
forts and not easily distracted from her goals. Finally, C6 (Deliberation)
shows that Debbie is a thoughtful and careful young woman. She is not
one to make spontaneous decisions; rather, she will weigh her options
carefully and make measured responses.
These very high Conscientiousness facet scores, coupled with the low
facets on Openness (e.g., Feelings and Actions) and the low AS (Modesty)
score may also suggest someone who has a high need for control over her
environment and possibly the people in it. Given her low Neuroticism
scores, this need for control is not defensive in nature (Le., an attempt to
protect herself against perceived emotional vulnerabilities). Rather, it stems
from this young woman's clear sense of direction and purpose in life.
The overall impression that emerges from this profile is one of a
young lady who is very ambitious and competent. Debbie has set some
very high goals for herself (such as getting a full scholarship to college)
and will do what it takes to reach them. What also emerges is a sense that
Debbie may be perceived as somewhat aloof and distant from others. Al-
though active in extracurricular activities at school, these endeavors may
serve the purpose of obtaining recognition and success more than as a
means of making friends and socializing. Her low Openness facets show
Debbie to have some personal rigidity in terms of her belief and value sys-
tems which may serve to isolate her from others who may not conform to
these personal ideals. Further energizing this rigidity is her high Dutiful-
ness (C3) scores, which show her to have a committed sense of right and
wrong. Whether she is involved with others is dependent on whether or
not they conform to her values and goals; her average score on E shows
that she can be comfortable with others or be alone. Her high Assertive-
ness (E3) and low average Compliance (A4) show her ability to be a leader
and manage others in ways that will take her closer to her goals.

CASE HISTORY: JOE W.

The second case profile is for Joe W., a 45-year-old White male who is
being seen in treatment for being a compulsive gambler. His profile is pre-
sented in Figure 3-2. At first glance, the profile seems to be rather "flat."
 INTERPRETING THE NED PI-R 97

N E o A c
N E 0 A C

.<: .
,!ill eo -
:I: .

.
II
I!
~" :
:
.
.
« :

FIGURE 3-2. NED PI-R profile of Joe W.

However, this belies the many interesting features that are present in this
case.
The overall N score shows an individual who experiences an average
amount of negative affect in his life. He is as blue and distressed as most of
us. An examination of his facet scales shows Joe to be low on Nl (Anxiety)
and N2 (Hostility), suggesting that he is a rather calm and easygoing person.
This low level of anxiety perhaps allows him to gamble without having to
worry about the future. His high score is on N5 (Impulsiveness), indicating
that Joe is frequently bothered by thoughts of doing things that he feels may
not be appropriate. He experiences strong urges or temptations. His average
score on N6 (Vulnerability) suggests that he can cope with these feelings rel-
atively well, although there may be times when he feels overwhelmed.
A point of interest here. According to the NED PI-R manual (Costa &
McCrae, 1992c), high scores on NS suggest that an individual will give in
to their temptations. I have found from my own usage of the NED PI-R
that individuals high on Impulsiveness are indeed bothered by many
urges, but whether they capitulate to them is a function of their scores on
Conscientiousness, particularly the CS (Self-Discipline) and C6 (Delibera-
tion) facets. If an individual's scores are low on these facets, then it is likely
that the person will give in to the urges. If scores are high, then the indi-
vidual is unlikely to give in.
 98 CHAPTER 3

The domain score for Extraversion is in the high range, indicating that
Joe is, overall, a sociable and engaging individual. He likes the company of
others and is very spirited and engaging. E1 (Warmth) is in the average
range, suggesting that Joe is discriminating in his approach to others; he
can be formal and aloof or friendly and approaching, depending on the sit-
uation. A more consistent description is provided by the remaining E
facets, which are all in the high range. Joe is certainly gregarious, wanting
to be with others. When in the presence of others, he enjoys being the
leader or center of attention. He wants others to look to him for direction.
He is high on E4 (Activity), so Joe is energetic and active. He has a fast per-
sonal temp<>to his life, always liking to be busy. High on E5 (Excitement-
Seeking), he enjoys risks and thrills. He wants to be where the action is,
preferably in the center of it. Finally, the high score on E6 (Positive Emo-
tions) shows Joe to have a high degree of elan and spiritedness. He exudes
much personal energy, and others may be attracted to his very upbeat and
positive outlooks. People may consider him to be charismatic.
His average score on the Openness domain shows Joe to have some
flexibility but also to have areas that need more structure. He is low on the
04 (Activity) and 06 (Values) facets. This, like Debbie, portrays an indi-
vidual who is rigid and dogmatic. He has a clear structure that he likes to
follow and a value system that clearly outlines what he needs to do. Joe
scored high on 03 (Feelings), suggesting that he is open to his own feelings
and those of others. He realizes the value of emotions in his life and is com-
fortable working with a wide range of them. Joe's average scores on 01
(Fantasy), 02 (Aesthetics), and 05 (Ideas) suggest that Joe does not have a
developed inner world and may prefer to keep focused on tasks and peo-
ple around him. His average Ideas score indicates that he may concern
himself less with "the big picture" and more with the bottom line. He
would appear direct and pragmatic.
Scores on Agreeableness are in the average range, suggesting that al-
though caring and concerned about the welfare of others, there is a real-
ization that people may not always be what they appear. As such, there is
some caution in joe's evaluation of others. Of note is his low score on A6
(Tender-Mindedness), reflecting a more hard-hearted, intolerant attitude
toward others. Joe is not moved by emotional appeals from others, tending
to interpret situational distress in more rational, cognitive ways. It would
be of interest to evaluate this low A6 with his low 06 score. Perhaps this in-
sensitivity may work out of his own value network. Given his high 03
(Feelings) score, it can be concluded that this coldness does not originate
from some avoidance of emotions or fear of them.
Finally, scores on Conscientiousness are also average, indicating that
Joe has some degree of personal organization and reliability that enables
 INTERPRETING THE NEO PI-R 99

him to work successfully at goals and to aspire toward socially approved

ideals. However, his Conscientiousness score is not so high as to indicate
a narrow focusing on success that may filter out other endeavors (as noted
with Debbie). He works hard, but may not bring his work home with him.
His low Achievement (C4) score, coupled with his low Anxiety (Nl) score
suggests that he is unlikely to worry that the time and money spent on
gambling will hurt or take away from his career status. Of note in this pro-
file is the high C6 (Deliberation) score, which shows Joe to be thoughtful
and cautious. Although he is high on this facet, his low C5 (Self-Discipline)
suggests that Joe may not have the discipline and motivation to follow
through on those tasks he knows he should. He can be easily distracted be-
cause he may lack the motivation to accomplish some of his goals.
Overall, there are some interesting points to be noted. First, as stated
previously, Joe is a compulsive gambler. Compulsive behaviors, whether
gambling, drug usage, or sexual, are all noted by scores on three facets:
high on N5 (Impulsiveness) and E5 (Excitement-Seeking) and low on C5
(Self-Discipline) and/or C6 (Deliberation). Individuals high on N5, E5,
and low on C5 and/or C6 all present some type of compulsive behavior.
This pattern of scores represents an individual who craves excitement and
thrills, is plagued by recurring thoughts to give in to those urges and who
lacks the personal discipline necessary to control the urges.
The average scores on Neuroticism and Conscientiousness, as well as
the high score on 03 (Feelings) indicates a good therapeutic prognosis for
Joe. Unlike most psychotherapy patients, Joe does not have a lot of nega-
tive affect that can clutter and undermine therapeutic interventions. His
scores on ConsCientiousness indicate that Joe would work at therapy and
follow through on the therapy interventions. His openness to feelings pro-
vides a doorway into his inner world; Joe is aware of his feelings and com-
fortable in discussing them. His levels of Extraversion make Joe a suitable
candidate for group therapy. He would enjoy working with others and
would certainly blend in well.
It is interesting to note that Joe is employed as a police officer. This
would explain his lower scores on Tender-Mindedness (A6), Actions (04),
and Values (06). As a police officer he has a clear set of values that guide
his behavior. His clear sense of right and wrong, respect for authority, and
appreciation for structure are all coincident with a career in law enforce-
ment, where the legal system provides a clear sense of structure and orga-
nization. The lower A6 makes adaptive sense in this circumstance. Joe is a
man who encounters a wide range of individuals in his daily work, most
of whom may have broken the law. His low Tender-Mindedness may pro-
vide a healthy skepticism and degree of objectivity in evaluating the cir-
cumstances surrounding the events in which he is involved in.
 100 CHAPTER 3

CASE HISTORY: FRANK AND JUDY V.

In the beginning of this chapter I talked about the usefulness of self-
report data. Although much effort has gone into (and continues to be in-
vested in) developing scales to evaluate the validity of self-report
information, the data to date suggest that self-reports are relatively free
from deliberate distortions. However, there are limitations to self-report
data. Even in good circumstances, there is always some level of distortion
present because no one has a perfect view of him- or herself. That is why
the use of observer ratings is encouraged. Observers can provide a very
different perspective on a client; at a minimum, an observer rating indi-
cates how the client is perceived by others and the kinds of impressions
that the client makes on others. Certainly, observer ratings have their own
limitations, but a comparison of a rating with a self-report can provide a
broader assessment context for evaluating a client. Points of convergence
between the two sets of assessments can provide greater confidence in the
accuracy of the self-report. Areas of divergence may indicate blind spots or
a lack of insight on the part of the client in terms of how they are perceived
by others in their environment. Chapter 5 provides more details for ob-
server data, particularly with couples.
The current case example involves a self-report by Frank V, a 35-year-
old White male employed in the family business which is sales-based. He
has a college degree. He is married to Judy V., a 38-year-old White female
who is employed in the mental health field. She provides an observer re-
port. They had been married for five years at the time of the assessment.
This is the first marriage for him and the second for her. They have two
children from the current union, and she brings two additional children
from her past marriage. The two profiles are presented together in Figure
3-3 in order to facilitate comparison. For purposes of illustration, I inter-
pret the self-report first, and then evaluate the wife's responses in relation
to the self-report.
As can be seen in Figure 3-3, Frank scores high on the Neuroticism do-
main. This indicates an mdividual who is experiencing some degree of
negative affect in his life. Worried and anxious, he may be unable to man-
age the stresses of his life. Individuals who score high on Neuroticism ex-
perience many somatic complaints and have a tendency to burn out when
their jobs become stressful. In evaluating Frank's facet scores for this do-
main, he scores high on Nl (Anxiety), N2 (Hostility), and N4 (Self-Con-
sciousness). Frank worries a lot about his life and finds himself easily
frustrated by events. His high N4 score indicates that Frank has low self-
esteem and may feel inadequate and inferior to others. The high score on
Hostility indicates Frank's tendency to become easily frustrated and upset.
 INTERPRETING THE NEO PI-R 101

FIGURE 3-3. NEO PI-R profile of Frank V. along with his spouse's rating.

Determining whether this score represents a potential for Frank to strike

out at others requires an examination of his Agreeableness and Conscien-
tiousness scores. Low Agreeableness and low Conscientiousness would
argue for the potential of Frank to act out his anger and aggress on others.
However, in this case, his average level of Agreeableness and high level of
Conscientiousness would suggest a good amount of impulse control.
The average score on N6 (Vulnerability) suggests that despite the pres-
ence of ongoing distress, Frank claims to be able to manage it appropriately.
Although he does admit to having his "bad" days, overall his capacity to
cope is commensurate with most other people's. His average score on N3
(Depression) and N5 (Impulsiveness) suggest that these qualities are less
defining of his overall distress. It should be pointed out here again that high
scores on the Neuroticism facets are not, in and of themselves, indicative of
any type of psychopathology. An individual can score high on the Depres-
sion facet and not have a diagnosable condition. All these affects are within
the normal range of functioning; they represent areas of distress and dis-
comfort, not areas of psychosocial impairment and dysfunction.
Frank's score on Extraversion is in the high average range, suggesting
that he enjoys the company of others. An examination of the facets shows
 102 CHAPTER 3

Frank high on E1 (Warmth) and E2 (Gregariousness), indicating that he is

a very personable individual who enjoys the company of others. Interper-
sonal activities are very much valued by Frank and reflect an adaptive
aspect of his vocational endeavor: sales. Frank scores low on E3 (As-
sertiveness) and E6 (Positive Emotions), indicating that among people he
is not very assertive or "pushy"; rather, he will tend to follow the lead of
others. His low E6 indicates a lack of spiritedness about him; he does not
radiate energy or ebullience. Instead, he has a more serious demeanor.
His Openness domain score is in the low range, suggesting a lack of
permeability between his inner and outer worlds. The low Openness sug-
gests a need for structure and clarity as well as a desire to follow routine.
Frank scores low on 01 (Fantasy), showing a very limited inner world; he
is more concerned with what is going on outside of him than on respond-
ing to inner issues. His low 05 (Ideas) score shows Frank to be a very "bot-
tom-line" type of person, not concerned with large theoretical issues but
focusing on more immediate, practical concerns. He scores high on 02
(Aesthetics), indicating an interest in and appreciation of art and beauty.
His high score on 03 (Feelings) indicates an openness to a wide range of
feelings. High 03 indicates an empathic orientation as well. Frank's aver-
age score on 06 (Values) indicates that there are some values that are im-
portant to him and are not available for debate or modification. However,
other values are more flexible and amenable to change and development.
On the Agreeableness domain Frank scores average, suggesting a
somewhat positive orientation toward others that may be conditional. He
likes others, but believes that some people may not be as trustworthy as
others. An examination of his facet scales shows Frank to be very high on
A3 (Altruism). He sees himself as being considerate and generous, cer-
tainly very willing to help others in need. He also scored in the very low
range on A4 (Compliance). This suggests that Frank is willing to compete
with others rather than cooperate and is capable of expressing anger when
necessary. Scores on this scale indicate a desire to do things one's own way
rather than acquiesce to the wishes of others. All the other facet scores on
this domain are in the average range, suggesting that Frank is as trusting
and straightforward as most people, and has a certain skepticism about the
motives of others.
Frank rates himself high on overall Conscientiousness, suggesting
that he sees himself as an organized, reliable, and hard-working individ-
ual who can be counted on to follow through on his commitments. An ex-
amination of his facet scores supports this broad-based interpretation. He
is high on C1 (Competence), indicating a general sense of personal self-
efficacy. The high score on C3 (Dutifulness) reflects that Frank strictly ad-
heres to his own ethical principles and can be counted on to fulfill all his
 INTERPRETING THE NED PI-R 103

moral obligations. High C5 (Self-Discipline) and C6 (Deliberation) scores

reflect an individual who is premeditative and focused. He carefully pre-
pares his course of action and will find the motivation to pursue it until
completion. His average score on C4 (Achievement Striving) shows that
Frank has some degree of personal aspiration and will strive for his career
goals. However, he has the capacity to step back from these job-related
goals and indulge in more leisurely pursuits. In short, it is not likely that
Frank would become a "workaholic"; he can set limits on the demands of
his job so that they do not interfere with other aspects of his life, such as
family. His average C2 (Order) score reflects some degree of personal or-
ganization.
From a larger, profile perspective, there are some points of interpre-
tive interest in this self-report. First, low scores on E3 (Assertiveness) and
A4 (Compliance) indicate possible passive-aggressive qualities. Here is an
individual who, in a group, prefers more the follower role than the leader
role. Yet, the low Compliance score indicates a preference to do things his
own way. Thus, Frank may find himself caught in many conflicting situa-
tions, wanting to follow along with the group, but feeling that they may
not be responding to his needs and desires to move in another direction.
His high score on N2 (Hostility) shows that Frank may frequently wind up
frustrated and angry. Although his A4 (Compliance) score shows him
likely to express anger directly, high C5 (Self-Discipline) and C6 (Deliber-
ation) scores suggest that any expression of anger may be very controlled
and not fully convey the depth of frustration he may be experiencing.
Another point of interest is the apparent contrast between the N4
(Self-Consciousness) and Cl (Competence) scores. One may question how
a person can feel both inadequate and efficacious at the same time. The
presence of this type of profile pattern underscores the reality that these di-
mensions are indeed independent of one another. They are capturing very
distinct types of dispositions. In this case, the high N4 suggests that Frank
has some internal feelings of inadequacy. These feelings may make him
appear defensive and he may shy away from some challenges because the
real potential for failure may exacerbate his self-consciousness. The high
Cl score suggests that within certain limited circumstances, Frank has de-
veloped a good sense of competence and ability at performing defined
tasks. High scores on both scales suggests that Frank may not have the
ability to generalize readily his developed competencies from one circum-
stance to another. Each new performance situation may find Frank anx-
ious, worried, and insecure. He may need much reassurance and support
in order to succeed. Once he does reach a level of success, feelings of self-
efficacy will emerge and he will feel increasingly secure. However, move-
ment to another performance situation will begin this process again.
104 CHAPTER 3

Frank's high scores on £1 (Warmth) and £2 (Gregariousness) indicate

a person who likes being with people, and others are attracted to him be-
cause of his easy demeanor and approachability. Because of his high scores
on Nl (Anxiety) and N2 (Hostility), he may become dependent on others
in order to find solace for his distressing feelings. He may look to others to
provide solutions to his problems or to take the lead in addressing his is-
sues. Yet, given his low £3 (Assertiveness) and high N4 (Self-Conscious-
ness), he may not be sufficiently direct or forthcoming with others about
his own needs and inadequacies. Thus, they miss his indirect and oblique
attempts at sharing and he may wind up becoming frustrated with the re-
lationship. As-a result, there may be much turnover in Frank's relation-
ships with others.
In evaluating the wife's (Judy's) ratings of Frank, it is clear that there
is much convergence between the two profiles. Thus, one can be confident
in the accuracy of the information provided by Frank. However, there are
some areas of disagreement that provide useful interpretive material. Keep
in mind that disagreements do not represent areas where Frank may be de-
liberately lying or distorting, or mean that the self-report is invalid. Rather,
areas of disagreement may indicate that the respondent has blind spots in
self-evaluation, or is not aware of the kinds of impressions he or she gen-
erates in others (Chapter 5 discusses the fact that when dealing with a mar-
ned couple such distortions may reflect areas of conflict or dissatisfaction
that the rater may be experiencing). Of course, the context of the rating
needs to be appreciated as well. A rater may encounter the target only in a
specific environment, or see the subject from a particular perspective. In
such circumstances, more confidence should be given to areas of agree-
ment than disagreement. However, a spouse's report can provide infor-
mation that may not be available from other types of raters, such as friends
or acquaintances. In this case, disagreements may be as revealing as areas
of convergence.
For this couple, the one salient difference between the two reports is
the magnitude of the ratings. This is most pronounced on the Neuroticism
and Conscientiousness domains. Yes, Frank's wife sees him as experiencing
negative affect, but much more than he admitted to himself. Yes, the wife
sees Frank as ambitious, organized, and reliable, but not to the same degree
as Frank perceives. Overall, Frank sees himself as rather closed, but not as
impermeable and dogmatic as Judy sees him. From this, one may hypoth-
esize that Frank may tend to minimize his weaknesses and maximize his
strengths. Given his high score on N4 (Self-Consciousness), this may be a
real possibility. However, it is equally possible that his wife tends to exag-
gerate Frank's weaknesses and play down his strengths. Observers do not
have a "lock" on the truth and should not be seen as the ultimate criterion.
 INTERPRETING THE NEO PI-R 105

There are some specific areas of disagreement that are noteworthy.

First, the difference on N6 (Vulnerability) indicates that Judy does not be-
lieve that Frank is coping as well with his negative affects as he indicates.
Frank believes that he is able to cope with his difficulties as well as the av-
erage person. Judy believes that Frank is frequently overwhelmed by these
stressors. In fact, Judy believes that Frank is more depressed (N4) than he
admits. The difference on the E3 (Assertiveness) facet suggests that Judy
sees Frank as much more passive and nonassuming. On the Openness do-
main, Judy rates Frank much lower on 02 (Aesthetics) and 05 (Ideas).
Judy sees Frank as not at all concerned about artistic issues and very much
concerned with th~ immediate implications of events. Frank is seen as hav-
ing a very focused, narrow, and unimaginative orientation toward the
world. On Agreeableness, Judy rates Frank much lower on Al (Trust), sug-
gesting there are real issues around intimacy and closeness with others. Fi-
nally, on Conscientiousness, Frank is rated lower on all of the facets except
C6 (Deliberation). Most notable are the C2 (Order) and C3 (Dutifulness)
facets. Here Judy believes Frank to be less organized and reliable than he
contends.

PROVIDING FEEDBACK
The NEO PI-R provides a wealth of information about an individual.
The information that is available is intended for interpretation by a pro-
fessional trained in psychological testing. Thus, providing test takers with
their interpretive reports would be inappropriate. However, there are
times when it may be useful therapeutically to share the results of testing
with a client. Feedback from the NEO PI-R may help provide a paradigm
for clients to understand their presenting problems and issues. Sometimes
clients suffer from distress that they are unable to articulate. The five-fac-
tor model can provide a framework for thinking about these personal is-
sues and can facilitate personal insight. 5ecOlld, the five-factor model can
be used as a common language for both therapist and client in discussing
therapeutic issues. The NEO PI-R may help frame the clinical dialogue in
ways that the patient can comprehend and that are meaningful for the
therapist.
To make such a disclosure would require a method of presentation
that, while accurate, would be nonthreatening to the client and under-
standable. The NEO PI-R comes with two methods for sharing results
with the client. The first is "Your NEO Summary," a standardized one-
page form that provides a general overview of the client's standings on the
domains. A reproduction of this form is presented in Figure 3-4. Each of
106 CHAPTER 3

Your

Summary Paul T. COlta, Jr., Ph.D., aDd Robert R. McCrae, Ph.D.

The NEO inventory measures five broad or adjustment. It does, however, give you some idea about
domains, or dimensions, of personality. The responses what makes you unique in your ways of thinking, feeling,
that you gave to the statements about your thoughts, feel· and interacting with others.
ings, and goals can be compared with those of other adults This summary is intended to give you a general
to give a description of your personality. idea of how your personality might be described. It is not
For each of the five domains, descriptions are a detailed report. If you completed the inventory again, you
given below for diffi:rent ranges of scores. The descriptions might score somewhat diffi:rently. For most individuals,
that are cMcked provide descriptions ofyou, based on your however, personality traits tend to be very stable in adult·
responses to the inventory items. hood. Unless you experience major life changes or make
The NEO inventory measures differences among deliberate elfurts to change yourself, this summary should
normal individuals. It is not a test of intelligence or ability, apply to you throughout your adult life.
and it is not intended to diagnose problems of mental health

Compared with the responses of other people, yow- respoDses sugest that you caD be described as:

0 Sensitive, emotional, and 0 Generally calm and able to 0 Secure, hardy, and generally
prone to experience feelings deal with stress, but you relaxed even under stressful
that are upsetting. sometimes experience conditions.
feelings of guilt, anger, or
sadness.

0 Extraverted, outgOing, 0 Moderate in activity and 0 Introverted, reserved. and

active, and high.spirited. enthusiasm. \bu enjoy the serious. \bu prefer to be
\bu prefer to be around company of others but you alone or with a few close
people most of the time. also value privacy. friends.

0 Open to new experiences. 0 Practical but willing to 0 Down-to-earth, practical,

\bu have broad interests and consider new ways of doing rraditioll8.l, and preny much
are very imaginative. things. \bu seek a balance ser in your ways.
between the old and the new.

0 Compassionate, good. 0 Generally warm, trusting, 0 Hardheaded, skeptical,

narured, and easer [0 and agreeable, bur you can proud, and competitive. \bu
cooperate and avoid con1lict. sometimes be stubborn and tend to express your anger
competitive. directly.

0 Cooscientious and well· 0 Dependable and moderately 0 Easygoing, oot very well·
organized. \bu have high wen·organized. \bu organized, and sometimes
sundards and always srrive generally hive clear goals careless. \bu prefer not to
to achieve your goals. bur are able to set your work make plans.
aside.

FIGURE 3-4. An example of "Your NEO Summary"

the five rows corresponds to one of the major dimensions. When an indi-
vidual scores in the high range for a domain (e.g., T-score from 55 and
above), one simply checks the first box. Average-range T-scores (e.g.,
45-55) are represented in the second box. Finally, the interpretation for low
range T-scores (e.g., 0 to 45) are provided in the third column. These de-
 INTERPRETING THE NEO PI-R 107

scriptions are clear and easily understood by individuals not familiar with
psychological terminology. Further, the presentation is nonthreatening, es-
pecially for the more sensitive dimensions of high Neuroticism.
Another way that information can be shared with clients is through the
computer scoring program. This flexible program provides many advan-
tages for clinical use. First, individuals can complete the NEO PI-R at the
computer itself. Information can then automatically be scored and profiles
generated. If this is not possible, then responses can be manually entered.
The computer program provides a number of interpretive options, includ-
ing a section on clinical hypotheses. But for our current purposes, the com-
puter scoring program produces a three-page summary report that can be
given to the respondent. An example of this report is provided in the Ap-
pendix, where the results for Debbie K. are given. As can be seen, this report
provides more information than the "Your NEO Summary" form. It can be
generated easily when the computer scoring program is used.
Of course, for some clients feedback may be inappropriate, and it is
the test user's ultimate responsibility to determine the suitability of the in-
formation for the client's consumption. However, this information can
help reinforce the therapeutic alliance between client and therapist. For the
therapist, the wide range of information provided by the NEO PI-R can
help foster greater understanding and empathy for the client.
Particularly in the short-term therapy situation, NEO scores can be
helpful in facilitating understanding by providing a wider personological
context for evaluating the client's presenting problems. For the client, NEO
scores can provide an opportunity for new insight and personal under-
standing. By being able to more succinctly voice personal issues, the client
may come to feel better understood. Finally, the language of the five-factor
model can provide a common bridge for client and therapist to use in dis-
cussing the issues that emerge over the course of treatment.

WHAT IF MY NEO PI-R PROFILE DOES NOT

MATCH THE SELF-CONCEPT?
Whenever a test is given, there is always the possibility that the pro-
file that emerges is inconsistent with the respondent's self-concept. There
may be areas of agreement, but there may also be points of difference that
seem very inconsistent with the person's apparent personality. When such
disagreements arise, there are several points to consider.
There is always the possibility that the test is in error. Remember, error
is everywhere! No personality instrument is infallible and there is always
the chance that the test failed to capture the person accurately. If it seems
 108 CHAPTER 3

likely that the test is incorrect, then you should consider accessing a dif-
ferent source of information about the client, such as an observer rating.
Observer ratings provide a different perspective on the client that may be
revealing. But before concluding that the test is in error, consider the fol-
lowing two possibilities.

ERROR IN TEST COMPLETION

First, there could have been an error made in completing the test, not
so much that the respondent was distorting or "lying," but that the indi-
vidual may have made clerical errors in filling out the forms. For example,
responses on the NEO PI-R are arranged from strongly disagree to
strongly agree. It is possible that a respondent may have reversed that or-
dering accidentally. Instead of indicating agreement, the respondent may
have circled the disagreement response. One way of detecting this prob-
lem is to refer to the three simple "validity checks" at the end of the self-
report answer sheet. Individuals are asked if they have completed each of
the items correctly. Someone who answers "Disagree" or "Strongly Dis-
agree" to this item may have made a mistake in responding.
Another possibility is that the person may have inadvertently shifted
answers on the answer sheet, so that the response to question 15 was
placed in the response space for question 16. Therefore, all remaining an-
swers are moved down one. This would serve to skew the entire scoring
process.
However, if such clerical errors can be dismissed, then an examination
of the item content with the respondent may be in order. It may be that the
client filled in the wrong response to several items. However, if this can be
ruled out, then test score variation from one's self-concept may be an indi-
cation that the test has captured something new about the respondent.

RECONTEXTUALIZING THE CLIENT'S MOTIVATION

Once clerical errors can be ruled out, it may be informative to sit down
with the respondent and review their responses on the scale(s) that seem
in error. It may be that the client does not have a good understanding of
what a particular facet scale represents. Keep in mind that the five-factor
model has very specific meanings for each of its measured constructs. The
usage of these labels may not necessarily correspond to the way the client
is using language. For example, I use the NEO PI-R in a number of reli-
gious and pastoral contexts. Frequently I give the instrument to clergy
members who usually score average to low on the Values (06) facet. Fre-
quently they are taken back by this score, believing that this means they
 INTERPRETING THE NEO PI-R 109

have few, if any, values; that they are amoral or immoral. I am frequently
told that they certainly do have values! What they may not understand at
first is that the Values facet does not speak to whether one is moral or not.
Rather, it examines the degree to which an individual has made some fun-
damental commitments to a worldview that he or she is not willing to
compromise. Certainly clergy have made some decisions about how the
world operates, and they feel strongly about those values. Thus the low to
average score on Values makes more sense for this group when it is prop-
erly understood.
Educating the client in the language of the five-factor model is very
important. It can not only clear up any misinterpretations of what scores
represent, but it also provides a clear and concise language for discussing
his or her own personality and the kinds of motivations that influence his
or her behavior. Both client and therapist can use the same language for
communicating.
Another advantage to reviewing the items is that it allows you to dis-
cern the context the client was using in responding to the items. At times,
clients will say that they were evaluating their behavior in terms of their
work personae, or how they are at home. Or they may have understood
the item differently from how it was intended. In a study looking at dis-
agreements between spouses on their NEO PI-R ratings, McCrae, Costa,
Stone, and Fagan (1996) found that the context and referent behaviors used
for interpreting the items were the most frequent reasons for lack of agree-
ment.
A final reason for discrepancy may be that the respondent is doing
things for reasons different than they think. It has been my experience that
when you can rule out all clerical and interpretive reasons for a discrep-
ancy, the most likely explanation is that the individual may not be truly
aware of his or her motivations. The systematic evaluation of a person's
behavior that is afforded by a standardized personality instrument may
provide a different interpretation. For example, a client does very well in
college, always in the top 10% of her class, always excelling in each class,
always wanting to do more and learn more. She may believe that she is
highly achievement motivated, and as such, high on Conscientiousness.
Although Conscientiousness is related to academic achievement, there are
many motivational reasons for excelling in school. If this woman scores
Iowan Conscientiousness, then it may be that her success can be attributed
to other sources of motivation, such as her high level of Openness to Ideas
(05) and Openness to Actions (04). Such individuals are very theoretically
oriented, like new and different adventures, and enjoy new experiences
and new knowledge. So, her success may be due less to her ambitiousness
than to her innate curiosity and enjoyment of novelty. Or she may be high
110 CHAPTER 3

on Compliance (A4) and does well in school because of the high expecta-
tions her parents have set for her.
Although certain behaviors may be thought of as prototypical of par-
ticular dispositions (e.g., Buss & Craik 1980, 1983), the reality is that any
given behavior can express any number of traits, and a given trait can be
expressed through any number of behaviors. We should never take be-
havior at face value. The merit of a personality assessment instrument is
that it can provide insights into the individual that are unobtainable from
other sources. If a test told us only what an individual can verify, then
there would be no reason to give a test. It is exactly this type of discrepancy
that lies at the- heart of evaluation-the opportunity to gain an enhanced
understanding of the individual.
 INTERPRETING THE NEO PI-R 111

APPENDIX NEO PI-R feedback report for Debbie K.

-YOUR NEO PI-R SUMMARY-

Results For DebbieK

Age 17
Sex Female
Test Form S
Test Date 03/28/96
Test Administrator Ralph L. Piedmont, Ph.D.

The Revised NEO Personality Inventory measures five broad domains or factors of per-
sonality, and six more specific traits or factors within each domain. The responses that you
gave to the statements about your thoughts, feelings, and goals can be compared to those of
other college-age respondents to give a description of your personality.
The NEO Personality Inventory measures differences in personality traits. It is not a test
of intelligence or ability, and is not intended to diagnose psychiatric disorders. It does, how-
ever, give you some idea about what makes you unique in your ways of thinking, feeling, and
interacting with others. This summary is intended to give you a general idea of how your per-
sonality might be described. If you completed the inventory again, you might score some-
what differently, and other people might have different views of what you are like.
If you have any questions or concerns about this summary, please feel free to discuss
them with the professional who administered the inventory.

A Description of Your Personality

The N Domain
Traits in the N domain reflect different ways of reacting emotionally to distressing cir-
cumstances. Low scorers are resilient, rarely experiencing negative emotions; high scorers
often have strong emotional reactions. Overall, your responses suggest that you are very low
on this factor. Specifically, you are calm, relaxed, and generally free of worry. You seldom feel
frustrated, irritable, and angry at others and you rarely experience lasting feelings of sadness
or depression. Embarrassment or shyness when dealing with people, especially strangers is
not a problem for you. You report being good at controlling your impulses and desires and
you are able to cope well with stress.

TheEDomain
The E Domain measures traits related to energy and enthusiasm, especially when deal-
ing with people. Low scorers are serious and introverted; high scorers are outgoing extro-
verts. Your total score puts you in the low range on this factor. You are average in your level
of warmth toward others, but you sometimes enjoy large and noisy crowds or parties. You are
as assertive as most women when the circumstances require. You have a low level of energy
and prefer a slow and steady pace. Excitement, stimulation, and thrills have little appeal to
you and you are less prone to experience feelings of joy and happiness than most women.

The 0 Domain
The facets of this domain measure responses to various kinds of experience. Low scorers
are down-to-earth and conventional; they prefer the familiar and the tried-and-true. High
scorers are imaginative and open-minded. You score in the average range. Your responses
 112 CHAPTER 3

suggest that you are somewhat open. You have a vivid imagination and an active fantasy life.
You are like most people in your appreciation of beauty in music, art, poetry, and nature, but
your feelings and emotional reactions are muted and unimportant to you. You seldom enjoy
new and different activities and have a low need for variety in your life. You are interested
in intellectual challenges and in unusual ideas and perspectives, but you are conservative in
your social, political, and moral beliefs.

The A Domain
This domain is concerned with styles of interpersonal interaction. Low scorers are hard-
headed and competitive; high scores compassionate and cooperative. Across the six facets in
this domain, you describe yourself as being generally average on this domain. In particular,
you have moderal! trust in others, but are not gullible, recognizing that people can some-
times be deceptive. You are very candid and sincere and would find it difficult to deceive or
manipulate others, and you are reasonably considerate of others and responsive to requests
for help. You can be very competitive and are ready to fight for your views if necessary. You
are quite proud of yourself and your accomplishments, and happy to take credit for them.
Compared to other people, you are average in your concern for those in need, and your social
and political attitudes balance compassion with realism.

TheCDomain
Traits in this domain describe differences in motivation and persistence. Low scorers are
easygoing and not inclined to make plans or schedules. High scorers are conscientious and
well-organized. Compared to other college-aged respondents, your score falls in the very
high range on this factor. You are rational, prudent, practical, resourceful, and well-prepared.
You are very neat, punctual, and well-organized, and you are highly conscientious, adhering
strictly to your ethical principles. You have a high aspiration level and strive for excellence
in whatever you do. You are determined, perSistent, and able to force yourself to do what is
necessary. You are cautious and deliberate and think carefully before acting.
 CHAPTER 4

PROFILE ANALYSIS USING

THENEOPI-R

ORGANIZING NEO PI-R INFORMATION

GENERAL

The five domain scores and 30 facet scales found in the NEO PI-R provide
a tremendous amount of information about a client that may appear over-
whelming to the beginning NEO PI-R user. Keeping the names of the
scales at one's fingertips can be difficult, and remembering the meaning of
each scale even more challenging. It is always good practice to use this ter-
minology as frequently as possible, not just when interpreting a profile.
The language of the NEO PI-R is useful for conceptualizing the motiva-
tions of people, and you should try to use these labels for describing the
behaviors of others. The more you use the terminology, the more adept
you will become at both remembering and understanding the constructs.
No doubt when a scale provides a great deal of information, it is help-
ful to organize that information into meaningful, interpretive chunks. Such
aggregation appreciates the naturally occurring covariance among scales,
and enables a broader interpretation of the individual. Rather than focus-
ing on specific traits and qualities, the test interpreter can discern larger
trends and patterns in the respondent's life. Such "chunking" is referred to
as either profile or typological analysis.
A typology is the flip side of the taxonomic coin (Bailey, 1994). A tax-
onomy enumerates the fundamental elements that describe some set of ob-
jects, while a typology identifies constellations of these elements that
co-occur. For example, we can talk about an individual's desire to com-

113
 114 CHAPTER 4

pete, high activity level, and proneness to anger; each construct tells us
something about the person and begins to provide a sketching of their per-
sonality. Or we can see each of these traits as being part of some larger psy-
chological organization that, as a whole, may reflect patterns of behavior
and life outcomes that may not be discernible from an examination of each
of the individual traits-say, the Type A personality (Friedman & Rosen-
man, 1959). This well-known typology represents an aggregation of traits
that naturally occur and are easily identified. Most important, though, are
the behavioral and health consequences associated with this pattern of be-
havior. These outcomes are not readily inferred from the individual traits
that comprise this typology.
The typological approach is not uncommon in assessment. Perhaps
the best-known measure of type is the Myers-Briggs Type Indicator
(Myers & McCaulley, 1985). Here responses to questions are used to find a
person's standing on four separate dimensions, which are then used to
classify the individual into one of 16 different "types." Each type provides
information about the person that goes beyond what each of the four
scales reflect. The MMPI is another example. The information provided
from the 10 clinical scales is frequently reported in terms of a "two-point"
or "three-point" code, which reflects the highest two or three scales. These
codes have been developed to provide an expanded clinical picture of the
respondent (Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989;
Graham, 1990).
Similar research has been done with the five-factor model. Lorr and
Strack (1993) cluster analyzed the NEO PI responses of a group of college
students (cluster analysis is a multivariate procedure that identifies homo-
geneous subgroups of individuals based on similarities across a set of at-
tributes, in this case NEO PI scores. For an expanded discussion of this
technique see Aldenderfer & Blashfield, 1984; L. Kaufman & Rousseeuw,
1990). Lorr and Strack found six orthogonal sets of profiles. Each cluster
contained a different combination of the five domains. For example, Clus-
ter 1 contained emotionally stable, extraverted, and agreeable individuals.
Cluster 2 included high neuroticism, low agreeableness, and high openness
scores. The presence of such cluster types sets the stage for identifying NEO
PI-based typologies; each may have their own psychosocial sequelae.
Hofstee, de Raad, and Goldberg (1992) provided a different approach to
organizing and presenting five-factor data. They combined the five
dimensions into their 10 nonredundant pairs (e.g., Neuroticism and Extra-
version, Openness and Agreeableness, Extraversion and Conscientiousness)
and constructed them as circumplexes. Acircumplex is a circular ordering of
traits around two independent, bipolar dimensions. This type of organiza-
tion permits an evaluation of how the two dimensions blend together per-
 PROFILE ANALYSIS 115

sonologically. For example, we can discuss the meaning of high Extraversion

or high Agreeableness; each has its own established construct validity, and
high scores on these domains represent clear psychological dynamics. How-
ever, what does being high on both Extraversion and Agreeableness mean?
What kind of person reflects high levels of each simultaneously?
Using self- and observer reports, Hofstee and colleagues (1992) pro-
vided descriptive labels for the entire circumference of the circumplexes.
They referred to this as the Abridged Big Five Dimensional Circumplex
(AB5C). In some ways, the AB5C model is the five-factor model equivalent
of the two-point codes found with the MMPI. The value of the AB5C is that
it provides a methodology for aggregating information from the five-fac-
tor model into brief typological-like sketches, which may have their own
psychological significance. A portion of this information infuses the inter-
pretive presentations in later sections. Readers are encouraged to obtain a
copy of the descriptors for each of the 10 circumplexes for use with their
interpretation of NEO PI-R profiles.
In the discussion that follows, the interpretive strategy presented is
based on the circumplex approach of Hofstee and colleagues (1992), but not
all 10 models are used. Rather, 5 particular two-dimensional models have
been selected because of their theoretical and practical importance in ap-
plied contexts. As we saw in Chapter 2, research has been done linking the
five-factor model to other theoretical and empirical personality models (see
Table 2-3), and the results of these studies have indicated that some per-
sonality systems can be efficiently located within the two-factor space de-
fined by various five-factor domains. The five particular two-dimensional
models that are presented in the following sections are Interpersonal Func-
tioning, Emotional Well-Being, Competitive Capacities, Character, and Psy-
chotherapeutic Treatment Response.

PROFILING NED PI-R SCORES

The interpretive models presented here are not circumplexes in the
true sense of the term. As noted earlier, a circumplex is a circular ordering
of traits around two independent dimensions. A summarization of such a
circumplex is presented here: a simple 2 x 2 matrix that contains the per-
sonological descriptions of that area of the circumplex that blends the two
factors together. Imagine a circle bisected by two axes, one running north
to south, the other east to west (these axes represent the two independent
dimensions contained in the circumplex and form a 90-degree angle to
each other). The ends of these two axes represent the poles of one of the
personality domains. The descriptive labels found here have already been
presented in Chapter 2 in describing each of the domain scores. However,
116 CHAPTER 4

the interpretive descriptions presented here are taken from the part of the
circle that is located 45 degrees from these two axes.
The two personality domains that are involved in these matrices are
listed. An individual is considered high on a domain if his or her T-score
is above 55, and considered low if the T-score is below 45. Of course, the
higher (or lower) the scores, the more clearly an individual can be assigned
a category, and the more characteristic the description is of the client's per-
sonality.
Individuals with scores in the mid-range on one or both of the do-
mains are more problematic. If an individual has a single T-score in the av-
erage range (e:g., T-score between 45 and 55), classification is still possible
although the description may be less accurate. If the respondent's other
score is in the high range, then consider the average score to be low. Con-
versely, if the extreme score is in the low range, then the average score
should be considered high. For individuals whose scores are both in the av-
erage range, no real classification is possible for the particular model. For
such respondents, it may be more useful to place emphasis on the facets
for each of the domains.

Interpersonal Functioning
Perhaps one of the most influential modem theories of personality is
the Interpersonal Circumplex (Leary, 1957). Interest in interpersonal rela-
tions has a long history (see Wiggins, 1996, for a brief history), and current
researchers have done much to extend this conception of personality to a
wide range of psychological phenomena, including abnormal behavior,
problems of living, and human social interaction (Horowitz, 1996; Lorr,
1996; Pincus, 1994; Pincus & Wiggins, 1990; Wiggins, 1980).
The circular ordering of interpersonal behaviors is anchored by the
constructs of Dominance versus Submission and Love versus Hate. All of
our interpersonal strivings are a combination of these two motivations.
Given our social nature, these qualities cut across much of human en-
deavor. Wiggins and Trapnell (1996) showed persuasively the utility of
these two dimensions across a wide range of perspectives, including evo-
lutionary psychology, sociology, and anthropology. They argued that
Dominance and Love are expressions of the higher-order constructs of
Agency and Communion (Bakan, 1966).
Given the value of this model, it is not surprising that the two con-
structs that define this circumplex also define the first two factors of the
five-factor model (in factor analytic studies the order in which these di-
mensions emerge is Extraversion, Agreeableness, Conscientiousness, Neu-
roticism, and Openness). McCrae and Costa (1989c) jointly analyzed
PROFILE ANALYSIS 117

measures of interpersonal behavior and the NEO PI and found that the
domains of Extraversion and Agreeableness corresponded well to Status
and Love, respectively. Domain scores from the NEO PI-R for Extraversion
and Agreeableness can be used to evaluate the interpersonal style of a re-
spondent. Definitions of the quadrants defined by these two domains are
presented in Table 4-1.
The four "types" presented here speak for themselves. They describe
very clear styles of interacting. Each has its strengths and weaknesses; no
one style should be considered superior. Again, the more extreme a re-
spondent's scores are on the Extraversion and Agreeableness domains, the
better these brief snapshots define the person's style. Finally, this table, like
the others to be presented, provides only broad descriptions of the re-
spondent. If more specific information about interpersonal style is needed,
then one should rely on inspection of the facet scales for these two do-
mains. In any event, these tables are best considered as hypotheses about
a client's interpersonal style that are in need of substantiation through the

TABLE 4-1. Interpersonal Styles Matrix

EXTRAVERSION
Low High
These individuals may experience These individuals are domi-
difficulties in expressing affection nant and self-assured.
toward others and may appear They tend to be assertive,
detached and unconcemed. They forceful, firm, persistent,
may tend to hold grudges. Such and self-confident. In a
Low individuals may enjoy more solitary group they tend toward
fJl pursuits and pride themselves on the leadership positions:
fJl
their self-reliance. Some descriptors They certainly enjoy direct-
t.Ll
for this category are aloof, skeptical, ing and managing others.
Z
t.Ll
unfriendly, joyless, cynical, and
-:J impersonal.
co
« These individuals tend to be unas- These individuals are warm
t.Ll suming and self-effacing. In a and agreeable type people.
t.Ll group setting, they may feel They are accommodating,
PI: uncomfortable with leadership kind, charitable, and
C,) roles. They are easily involved in sympathetic. They enjoy
« High worthy projects, but their support being with others and
is given without much fanfare. interact with them in a
Descriptors for this category include very nonthreatening,
humble, modest, naive, lenient, and compassionate manner.
obliging. They are usually accom-
modating of requests made of them
for assistance.
118 CHAPTER 4

use of additional, relevant material (e.g., observer reports, responses from

other instruments, clinical evaluation).

Emotional Well-Being
Psychological or emotional well-being is an important aspect of psy-
chological functioning that researchers have been interested in examining
for decades (Andrews & Withey, 1976; Cantril, 1965). Many important ques-
tions revolve around this issue: What makes people happy? How does ill-
ness and disease impact levels of well-being? Do good things improve our
emotional well-being, while traumatic events reduce it? Does well-being
decrease over the life span? All of these questions speak to central issues of
concern for all of us. Research in this area has provided several important
insights into how our sense of well-being emerges. First, subjective well-
being is very different from our objective life situations (Costa & McCrae,
1984). One would think that poorer health, less money, loss of sexual vigor
and physical attractiveness would all contribute to making our lives less
"happy," but this is not substantiated in the literature. Increasing age is not
associated with lower levels of well-being (Costa, McCrae, & Zonderman,
1987; Larson, 1978). What the research literature does show is that levels of
well-being seem to be quite stable over the adult life span (Costa, McCrae,
& Zonderman, 1987), and that life's ups and downs, although impacting us
in the short run, have little to do with our long-term levels of happiness
(Costa, McCrae, & Norris, 1981; McCrae & Costa, 1988).
Given the stability of our levels of well-being, it is not surprising that
this life view is strongly linked to personality. Research has shown that
subjective well-being is a function of levels of Neuroticism and Extraver-
sion (Costa & McCrae, 1980a; McCrae & Costa, 1983a). These two dimen-
sions play the major role in how we perceive the world we live in and cope
with the stressors we face. In short, Neuroticism reflects the amount of
negative affect we experience in life, while Extraversion is linked to levels
of positive affect. Note that positive and negative feelings are not opposite
ends of the same continuum. Rather, they define independent affective ex-
periences (see Watson & Clark, 1992). Although at any given moment one
or the other affect can dominate our feelings, over the course of life we can
certainly experience strong negative and positive emotions simultane-
ously.
The domains of Extraversion and Neuroticism also influence our ex-
perience of stress and our ability to cope with it (Costa & McCrae, 1980b,
1987, 1989b). Specifically, individuals high on Neuroticism tend to expe-
rience more somatic complaints although they do not experience higher
levels of disease (e.g., S. Cohen et ai., 1995; Costa & McCrae, 1987; Zon-
 PROFILE ANALYSIS 119

TABLE 4-2. Emotional Well-Being Matrix

NEUROTICISM
Low High
These are individuals who may These individuals have a low
not experience much negative sense of well-being. They can be
or positive affect. They may easily distressed and over-
appear always to be on an even whelmed by the pressures
keel, appearing unflappable of the situation. Life may be
Low and perhaps emotionally de- perceived as subjectively
tached. Descriptors for this difficult and they may feel
Z
category include tranquil, unprepared for the pressures

-
0
CIl
0::
placid, and unexcitable. These
individuals may not respond
to situations with much
of life. Descriptors include
self-critical, insecure, nega-
tivistic, fearful, self-pitying,
w emotionality. nervous, and fretful.
>
<: These individuals have a strong These individuals experience a
0:: sense of wen-being. They are wide range of affects. There
E-
hardy and adaptive, look are high levels of BOTH posi-
>< forward to what life has to tive and negative affect. Life
w
offer. Adjective descriptors may be experienced as a series
High for this category are coura- of emotional ups and downs.
geous, strong, assured, con- Sometimes their outgoingness
fident, hearty, buoyant, may mask their own inner pain.
uninhibited, and bold. Descriptors for this category
include explosive, extravagant,
excitable, and volatile.

derman, Costa, & McCrae, 1989). Individuals who are high on Extraver-
sion appear to be quite hardy and happy and are able to cope well with
stress. The combination of these two domains and their psychological sig-
nificance are presented in Table 4-2.
Aclient's standing on these two domains can provide insights into the
general happiness of the respondent, his or her levels of subjective well-
being, and his or her ability to cope with stress. Clinical clients tend to
have very high levels of Neuroticism. This makes sense; individuals who
are experiencing a great level of psychological distress tend to seek out
counseling for relief of symptoms. Interestingly, individuals high on both
Neuroticism and Extraversion may not appear in need of treatment; the
positive emotions characteristic of the extraverted individual may mask
his or her need for treatment. Nonetheless, therapists with clients high on
Neuroticism need to keep in mind that their efforts with such clients may
impact only current stress levels. More intense, long-term therapy may be
needed to bring about a more broad-based impact on the client's overall
120 CHAPTER 4

sense of well-being and to improve his or her ability to manage stressful

events more effectively.

Competitive Orientation
With the publication of their seminal work on achievement motiva-
tion, McClelland, Atkinson, Clark, and Lowell (1953) provided a new per-
spective on individuals and their aspirations. The construct of need for
achievement was found to have great heuristic and empirical value for un-
derstanding why some individuals and societies succeed and others fail
(Bendig, 1958; Edwards & Waters, 1983; H. B. Gough & Hall, 1975; Pied-
mont, 1988b; Piedmont & Weinstein, 1994; Schmeck & Grove, 1979; Steers,
1975). Need for achievement is clearly located on the Conscientiousness
domain, where personal organization and reliability reflect a capacity of
the individual to delay immediate gratification so that larger, long-range
goals can be met (Digman, 1989: Digman & Takemoto-Chock, 1981).
However, with time, research began to show that achievement moti-
vation alone was insufficient to explain or anticipate competitive perfor-
mance, especially for women (e.g., Homer, 1968, 1972; see Tresemer, 1976,
for a review). New constructs were added in order to enhance predictabil-
ity, for example, Fear of Success, Fear of Failure, and Test Anxiety. These
variables addressed the ability of the individual to manage the inevitable
stressors that any competitive situation involves and the sense of personal
confidence and esteem necessary to support any failure that is encoun-
tered. This ability to manage threats to self-confidence and negative affec-
tive arousal is the domain of personality named Neuroticism. Both of these
constructs were seen as essential for understanding competitive perfor-
mance (Piedmont, 1988a, 1995). The Neuroticism by Conscientiousness
matrix is presented in Table 4-3.
Adding the domain of Neuroticism (specifically, the facet scale N4:
Self-Consciousness) to predicting competitive outcomes provides a new
window for understanding how individuals approach and interpret
achievement situations. It helps to explain why some individuals who are
capable and have high aspirations fail, while others who are capable suc-
ceed. In the former situation there is the individual who is high on both
Neuroticism (N4) and Conscientiousness. These individuals may aspire in
an attempt to compensate for actual or perceived personal weaknesses.
Their competitive endeavors are characterized as being very intense and
focused. Although ambitious, their inability to manage their own feelings
of inadequacy may undermine their ability to reach long-range goals, or to
function well under circumstances of close observation.
In the latter situation are those who are low on Neuroticism (N4) and
high on Conscientiousness (particularly the facet C4: Achievement Striv-
 PROFILE ANALYSIS 121

TABLE 4-3. Competitive Performance Matrix

NEUROTICISM (N4: Self- Consciousness)
Low High
These are individuals who may These are individuals who actively
appear to have rejected con- avoid competitive situations be-
ventional or socially approved cause the potential for failing
definitions of success. These in- may exacerbate internal feelings
dividuals are not competitive or of inadequacy and poor self-
ambitious, and are quite com- esteem. These individuals are
cJ) Low fortable following their more likely to experience fear of suc-
cJ)
"live for today" philosophy. cess, fear of failure, and test
~
They prefer to follow their own anxiety in response to achieve-
Z inner needs and may eschew no- ment-related situations or tasks.
cJ)
tions of materialism. Descriptors Descriptors include impulsive,
::J include informal, noncompeti- frivolous, self-indulgent, hypo-

-
0
E-
tive, easy-going, and relaxed.

These are the prototypical achiev-

critical, inconsistent, and fidgety.

These are individuals who are am-

Z

-
~

U
cJ)
ers: Emotionally stable, capable
with a heightened sense of com-
petence and a drive to succeed.
bitious and competitive, but
worry about the outcome of their
efforts. They set high standards
They set high standards for for themselves because they may
Z themselves and have the per- believe that reaching them will
0 High sonal organization to attain be an outward sign that they are
U those goals. They may appear to not as inadequate as they feel on
be overly dispassionate. Some the inside. Obstacles to obtaining
descriptive adjectives for this their goals are responded to with
category are: robotic, objective, anger and hostility. Adjective de-
poised, concise, mechanical, me- scriptors include particular, com-
thodical, and self-diSciplined. petitive, sore losers, intense, and
highly focused.

ing). These are the prototypical achievers-those who set goals and work
hard to attain high, socially meaningful standards of success. The effort
these individuals bring to achievement situations can oftentimes com-
pensate for weaknesses in ability. Their lower levels of Neuroticism (N4)
indicate that they have high levels of self-esteem and can cope well with
whatever setbacks they may encounter. Thus, failure may not be per-
ceived as much of a discouragement, but rather as a barometer of how far
they have progressed. They persist until they are able to overcome their
challenges.
Scores on these two dimensions can be useful in assessing clients with
job selection issues. Challenging environments attract those who have
high aspiration levels. To succeed in such a context necessitates a tem-
perament that can withstand the many stressors that arise, whether they
are time constraints, task demands, or managing the inevitable failures
122 CHAPTER 4

that occur. Therefore addressing self-esteem issues, or working to improve

this quality, would have tremendous benefits to high Conscientious peo-
ple: It would help them remove obstacles to their own success. The more
general domain of Neuroticism has important implications for a wide
range of work-related outcomes and therefore needs to be considered in
any performance context. Individuals high on Neuroticism tend to bum
out (Piedmont, 1993), are rated poorly by supervisors on various aspects of
their job performance (Piedmont & Weinstein, 1994), have a low level of re-
liability (Hogan & Hogan, 1989), and will tend to perform poorly on em-
ployer-given integrity tests (Mount & Barrick, 1995; Ones, 1993).

Character
Perhaps one of the more established terms in personality is the con-
cept of "character." In one of the first textbooks on personality (Stagner,
1937), character is described as a subset of the entire personality that re-
lates to moral or ethical activity. Allport (1937) captured this idea of ethical
admirability in his review of the concept: "character is the aspect of per-
sonality that engenders stability and dependability, that is responsible for
sustained effort in the face of obstacles, or works for remote ends rather
than those that are nearer in time but of less worth .... Character enters the
situation only when this personal effort is judged from the standpoint of some
code" (p. 51; italics in original).
There are two components of this definition. The first is the effort to
work toward long-range, socially valued and respected goals that have
higher "payoffs," to delay gratification, and to overcome immediate ob-
stacles. This quality certainly describes the personality domain of Consci-
entiousness. This dimension reflects an individual's personal organization
and reliability, his or her capacity to delay gratification and aspire toward
socially valued goals. Individuals high on Conscientiousness can be
counted on to follow through on their obligations and commitments.
The second element centers on the notion of social evaluation of ef-
forts. The struggle and the means by which the individual pursues the goal
are marked by ethical and moral features. Such qualities may include a re-
spect for others and a belief in higher principles of justice (either of a reli-
gious or patriotic nature). Such personal features usually constitute what
is referred to as a philosophy of life. The five-factor domain relevant to
these qualities is Agreeableness, which impacts our social attitudes and
values (Costa, McCrae, & Dye, 1991).
One may have thought that Openness would be the relevant domain;
after all, it has a facet scale for Values (06). But an important distinction
needs to be drawn here. Openness refers to our inner permeability, the de-
PROFILE ANALYSIS 123

gree to which outer events impact our inner lives and the ability of our inner
dynamics to find expression in our outer behavior. Individuals high on
Openness have value systems that are available for evaluation and modifi-
cation; the inner world is always being "updated" as new information be-
comes available. Closed individuals have a system that is more rigid and
fixed; the commitment to tradition and respect for authority restricts oppor-

TABLE 4-4 Character Matrix

AGREEABLENESS
Low High
These are individuals who are self- These individuals project a nurtur-
seeking and controlling. They ing orientation toward the
are interested in gratifying their world; they will tend to see the
own needs in the here and now. best in others and the potential
They seek hedonistic fulfillment they may offer. Although they
and may see others as objects to may involve themselves in so-
satisfy these physical needs. cial justice causes, they may lack
They may engage in subterfuge the stamina and discipline to
Low to gain their ends. They may be follow through on these in-
slow to recognize and respond volvements. Their efforts may
CIl to the needs of others. Descrip- be casual and appeare seIf-in-
CIl tors include eogtistical, impolite, dulgent. The flower children of
"'-l inconsiderate, thoughtless, rash, the 1960s may provide the pro-
Z disrespectful, self-absorbed, and totype here. Descriptive adjec-
CIl
rebellious. tives include tolerant, accepting,
::J
SOCially aware, sensuous, indul-

-
0
f-
These are individuals who have
gent, and superficial.

This category represents individu-

Z

-
"'-l set high standards for them- als who are very concerned and
selves and are ambitious. These involved in the plight of others.
U individuals may have intemal- They are responsive to the
CIl
ized positive values and are needs of others, and will re-
Z willing to fight for what they be- spond in helpful ways. When
0 lieve. They can be very deter- moved to action, this group
U High mined and focused on reaching may not respond in confronta-
their goals. There is much pas- tive ways. Rather, they are will-
sion in their efforts and they ing and able to "work through
may be confrontative and direct the system" in order to reach
when resistance is encountered their goals. They have the per-
in reaching their goals. Such in- sistence and diScipline to see
dividuals may love humankind, their efforts bear fruit. Adjec-
but distrust specific humans. tives that describe this category
Descriptive adjectives for this include moral, respectful, rever-
category include strict, deliber- ent, polite, considerate, sincere,
ate, stem, and rigid. and understanding.
124 CHAPTER 4

tunities for change. However, regardless of the amount of flexibility, the

Openness domain does not speak to the kinds of values one possesses.
Agreeableness, on the other hand, reflects a style or philosophy of life.
Agreeableness reflects one's orientation toward others, an evaluation of how
one comes to perceive the motives, intentions, and goals of others. Agreeable
people tend to see the best in others and wish to reach out to them. People
with low Agreeableness tend to see others as being disingenuous and self-
oriented, and therefore view others with suspicion and distrust.
The combination of the Agreeableness and Conscientiousness domains
is presented in Table 4-4. As can be seen, quite a wide range of character
styles is represented. Low Agreeableness and Low Conscientiousness rep-
resents a selfish, manipulative style that tries to find immediate gratification
for personal needs. Others are valued only in terms of what they can offer
the individual. This combination is frequently found with the Antisocial
and Passive-Aggressive Personality disorders (see Table 2-6). On the other
hand, there is the high Agreeable-high Conscientious individual, who may
be described as the antithesis of the psychopath. There is a strength of pur-
pose and a compassion of action in these types of individuals.

Psychotherapeutic Treatment Response

The real value of clinical assessment is that it provides information
about a client that the therapist can use for determining intervention
strategies. Although the nature of the problem primarily dictates the type
of intervention, to the extent to which there is room for choice (i.e., direc-
tive versus nondirective, group versus individual treatment, etc.), person-
ality traits of the individual have important implications for both
treatment selection and outcome (Costa & McCrae, 1992a, 1992c). Miller
(1991) provided an interesting analysis of the five-factor dimensions and
how clients evidence these qualities in the therapy session. This study pro-
vides the basis for the information presented in this section.
Although Miller (1991) noted that all five domains have something
important to tell about a client's response to psychotherapy, he noted that
Extraversion and Openness were most useful in terms of selecting treat-
ment method. Those high on Extraversion may appear eager for therapy
because they may value the interpersonal opportunities it presents. High
Extraversion individuals have energy and enthusiasm as well as an abil-
ity to translate mental events into language. Openness, on the other hand,
considers the patient's ability to consider novel solutions to their current
problems. Individuals high on Openness can appreciate learning new
things about themselves, can accept a wide range of their emotional expe-
riences, and can respond well to the active imagination techniques that are
part of many therapies. Low Openness individuals may not respond well
 PROFILE ANALYSIS 125

TABLE 4-5. Psychotherapy Treatment Matrix

OPENNESS TO EXPERIENCE
Low High
These are individuals who may be These are individuals who have a
reticent in talking with others wide, developed inner world.
about inner feelings and emo- They are curious about feelings
tions. They may not be respon- and emotions, and may spend
sive to a wide range of feelings, time naturally inquiring into
most of which remain diffuse their own feelings and motives.
and nonspecific. Descriptive ad- Their low Extraversion suggests
jectives for this category include that they may be somewhat fo-
Low passive, bland, apathetic, unin- cused on their own inner expe-
quisitive, inarticulate, and riences. Adjective descriptors
unimaginative. Such individuals for this category are inner-di-
may respond well to a direct, rected, contemplative, intro-
nonemotional approach, such as spective, and meditative. These
Z a cognitive-behavioral regimen. individuals may value therapies

-
0
tTl
~
that focus on inner dynamics,
such as Freudian or Jungian
analysis, Gestalt therapies, or
hypnotherapy.
t.t.l
> These are individuals who enjoy These individuals also have a de-
~
~
the company of others, but may veloped inner world, and are
E- not be comfortable with express- focused on ideas, feelings, and
>< ing inner feelings, or difficult emotions. However, they also
t.t.l emotional topics, such as anger bring a strong need for social-
and sex. There may be a mora lis- ization. They like talking about
tic or self-righteous quality about these feelings with others. They
these individuals. Adjective de- are into sharing and working
High scriptors include verbose and with others. Descriptive adjec-
pompous. Such individuals tives include Inquisitive, elo-
may benefit from a more goal- quent, worldly, witty, dramatic,
directed, problem-focused ap- expressive, spontaneous, and
proach to treatment. Group adventurous. Group environ-
therapy may be useful, but more ments that are oriented toward
as a support network than any personal revelation such as en-
type of encounter group. counter groups or even psy-
chodramas may provide an
ideal therapeutic medium.

to efforts at exploring their emotions, and may want "therapy to be a reas-

suring, practical experience" (p. 426).
The matrix of these two domains is presented in Table 4-5. As can be
seen, different combinations of these qualities may respond to vastly dif-
ferent kinds of therapeutic modalities, if available. The closed introvert, for
example, may need a very direct, functional therapy that does not focus on
126 CHAPTER 4

emotions, ideas, or novelty. Conversely, the open extravert may enjoy a

wide range of imaginative, creative therapies that combine interpersonal
contact with fantasy-based exploration (e.g., psychodrama or Imago ther-
apy). Assessing clients on these five personality dimensions may help in-
crease the impact of therapy by matching clients to therapies that they are
prepared to accept. In closing, it is sad to note that there are not that many
treatments available to match with patient characteristics. However, our
understanding of personality can give form and direction to efforts at de-
veloping new interventions that may be differentially sensitive to these
personality issues.

Caveats and Conclusions

The matrices just described are provided to facilitate interpretation of
the NEO PI-R by enabling one to evaluate various facet scales and domains
in relation to one another rather than in isolation. This more configural
process has much to commend it. However, these are only a few of the pos-
sible matrices that can be created. In a new adaptation of the NEO PI-R, the
NEO-4, six different configural "styles" are identified and their interpretive
Significance highlighted. With increased experience and proficiency in
using the NEO PI-R, users are encouraged to identify new interpretive pat-
terns that are of Significance in particular assessment situations.
The heuristic value of these matrices is found in their ability to outline
the personological implications of NEO PI-R scores. They certainly pro-
vide a broader interpretive context to the scores. However, their ultimate
benefit is for the NEO PI-R user who can integrate the important psycho-
logical aspects of their own unique assessment context into these frame-
works. As they stand now, these matrices provide broad descriptions; it is
up to you, the user, to provide the nuances and subtleties. In no instance,
though, should these descriptions fill in for, or supersede, good clinical in-
sight and judgment.
Finally, keep in mind that each of these matrices is, by definition, an in-
complete description of personality. There are five major dimensions of per-
sonality, and any complete portrayal needs to include information from
each of these areas. The interpretations offered for each matrix can certainly
be meaningfully expanded by including information from the other do-
mains. For example, with regard to the Treatment matrix, certainly Extra-
version and Openness are important, but Neuroticism also plays an
important role. As Miller (1991) noted, Neuroticism was an important pre-
dictor of treatment outcome; those with lower levels of negative affect
responded better to any intervention than those with higher levels. Consci-
entiousness will also have much to contribute to deciding on the appropri-
 PROFILE ANALYSIS 127

ate treatment. Those high on Conscientiousness will certainly work harder

toward their treatment goals, and can be relied on to follow through on any
assignments they are given. Thus, these matrixes are useful for highlighting
important psychological issues, but do not forget that they do not provide
all relevant information. Any interpretation of the NEO PI-R must include
all five domains as well as an evaluation of the facet scales.
The following sections present a variety of case histories for interpre-
tation. The reader is encouraged to consult the definitions presented in
Chapter 3 as you follow along with the personality descriptions provided.
Eventually, the reader should be able to form interpretations from the pro-
files directly without having to refer to the definitions presented in Chap-
ter 3. Every effort is made to apply each of the matrices to the case,
although such applications are not always relevant (e.g., using the Psy-
chotherapy Treatment matrix with an individual who is not in therapy). It
is important to keep in mind that the matrix descriptors are more charac-
teristic of clients with extreme scores on the relevant dimensions than for
those with scores nearer the mean.

SELECTED CASE HISTORIES

In this section, several case histories are presented along with their
NEO PI-R results. Before turning to the text interpretation provided, I en-
courage you to try to evaluate the profile. In this way, you can apply and
stretch your own interpretive skills and then compare your results with
mine. Of course, you may find insights into these cases that are not dis-
cussed in the interpretations; the evaluations provided here should not be
seen as exhaustive. Rather, every effort was made to capture only the most
salient features. In making the interpretations, we start with a considera-
tion of scores in each domain, always moving from the domain score to the
facets for Neuroticism, Extraversion, Openness, Agreeableness, and Con-
scientiousness, in that order. Then we proceed to interpreting the individ-
ual's scores on each of the five matries noted earlier.

CASE HISTORY: ROBERT R.

Robert R. is ci 35-year-old White, married academic. He has a doctor-
ate in the sciences and enjoys teaching and research in his area of expertise.
His NEO PI-R results are presented in Figure 4-1.
As can be seen, his overall level of Neuroticism is average, with par-
ticularly low scores on N5 (Impulsiveness) and N6 (Vulnerability). In gen-
eral, he experiences some negative affect but no more or less than most
128 CHAPTER 4

FIGURE 4-1. NEO PI-R profile of Robert R.

people. His low score on N6 shows that he copes very well with whatever
stressors he encounters, both from within himself and in his environment.
His low N6 score shows that he can tolerate frustration well.
Levels of Extraversion are just in the low range, suggesting that Robert
prefers small social circles and may prize his solitude. His high scores on E3
(Assertiveness) and E4 (Activity) show that he exhibits much energy, having
a quick personal tempo and being in the forefront of any group activities: He
likes to lead and has the energy to invest in leadership. He scores lower on
the affiliative aspects of this domain: E1 (Warmth) and E2 (Gregariousness).
He may appear to others to be formal and distant, and he prefers to do
things either alone or with small groups. He would not consider himself to
be a "people person," although he may do well in social circles and be seen
as quite surgent and charming. However, Robert would find such social en-
deavors to be difficult or energy expending activities. After much interper-
sonal contact he may seek quiet time in order to "recharge" himself.
Robert scores quite high on Openness overall, although there is some
important scatter on the facets for this domain. High scores on 01 (Fan-
tasy),03 (Feelings), and 05 (Ideas) show an active inner world, both emo-
tionally and intellectually. The high 05 is not uncommon among
 PROFILE ANALYSIS 129

individuals with advanced degrees: These individuals enjoy dealing and

working with ideas and intellectual interests. The high 01 score may pro-
vide additional resonance for his attraction for ideas. Robert has a well-de-
veloped inner world, which may provide a medium for creative discourse.
The noticeably lower score on 04 (Actions) suggests that his Openness is
less expressed in the outer world of actions than in his inner world of ideas
and fantasies. The facets of 02 (Aesthetics), 04 (Actions), and 06 (Values)
focus on the outer world of the person, while the facets of 01 (Fantasy),03
(Feelings), and 05 (Ideas) focus on inner events. Notice that Robert's
scores are much higher on the latter set than the former. The average score
on 06 (Values) suggests that Robert, although tolerant of diverse value
systems and willing to reexamine some of his own values, has a core set of
beliefs that are "nonnegotiable."
Robert's Agreeableness domain score is in the very low range, and all
the related facets are uniformly in this range. Therefore, Robert may be per-
ceived as skeptical of the intentions and motivations of others, egocentric,
and ready to fight for his own interests. Skeptical and critical: these qualities
may nicely suit someone who is a researcher by trade. However, being hard-
nosed and suspicious may restrict the number of close meaningful relation-
ships Robert has with others. Coupled with his low Warmth (E1) and
Gregariousness (E2), Robert may find himself very much alone, which is
where he may want to be. His low Compliance (A4), coupled with high As-
sertiveness (£3), suggests a high amount of dominance and independence,
and he may use bullying tactics to get into that dominant position. Robert
may not take well to being supervised by others, especially if such oversight
involves the need for making changes (see 04). Robert may be confronta-
tional and argumentative when provoked, as well as guileful and manipu-
lative. These characteristics may work well for an individual researcher, but
if thrust into a leadership position that requires a more collaborative ap-
proach, this reliance on aggression and dominance may become a liability.
Finally, the overall level of Conscientiousness is very high and all the
facets are also in the high to very high range. Such scores portray Robert as
being very ambitious, strong-willed and determined. He has a high degree
of competence and goal aspiration. He is well organized and capable of
following through on his commitments. This high degree of focus and di-
rection in conjunction with a strong personal tempo (E4) suggests an indi-
vidual who can project considerable personal energy and strength. This
persona may not have the cheerful optimism of someone who is high on
Positive Emotions (E6), a quality that others may find attractive and infec-
tious. Rather, Robert exhibits a strong personal presence that more reticent
and retiring individuals may find abrasive, and more insecure individuals
may experience as intimidating.
130 CHAPTER 4

The most salient aspect of Robert's profile is his low Agreeableness

and high Conscientiousness. This combination reflects an individual who
has a general wariness and suspicion of others as well as strong ambitions
to reach his own goals. In an individual context these traits may work well,
especially given his low sociability needs. However, if he were to move
into a more interdependent, nurturing environment, he may be found
quite abrasive.
Overall, Robert R. is a very energetic and ambitious individual who
has no doubt developed clear goals for himself which he pursues in a de-
termined and vigorous manner. He has many qualities that make him well
suited for a research career. He certainly prefers more solitary pursuits and
enjoys thinking abstractly. He is a natural skeptic, never accepting any-
thing at face value. However, he may be perceived as overpowering and
abrasive by others, not by intention, but because of the tremendous
amount of energy he possesses. However, when provoked, Robert can be
confrontational and may overreact to perceived threats. He may not al-
ways capture the interpersonal subtleties of the situation. Once injured,
Robert may not seek opportunities for reconciliation.
Within the Interpersonal Matrix presented in Table 4-1, it is clear that
Robert is in the Low Agreeableness/Low Extraversion vector. Robert has a
rough, curmudgeon-like interpersonal style that may make establishing
meaningful interpersonal relationships difficult. He may appear brusque
and uncaring, finding it difficult to express much tenderness and compas-
sion. Concerning the Well-Being matrix of Table 4-2, Robert falls into the
low Extraversion, high Neuroticism vector. Although Robert may experi-
ence many successes in his life, there is an underlying sense of lowered
well-being. Life may be seen as a struggle, filled with hurdles to be cleared.
Although energetic and confident, there may be times when Robert feels
that events have worn him down. Keep in mind that Robert's scores on
Neuroticism and Extraversion are mostly in the average range, so that his
experiences are not as negative as those of someone who has more extreme
scores on these dimensions. Very rarely would one expect him to be over-
whelmed by events and self-pitying. Rather, Robert may experience life as
a struggle, but one that he feels he can still win.
Concerning his Competitive Matrix status, he is high on Conscien-
tiousness and in the average to high range on Neuroticism. This pattern is
characteristic of an individual who is very competitive and ambitious. At
times Robert may take his competitive strivings personally, so that a given
setback may be experienced as evidence of personal failure or inadequacy.
In general, though, Robert is a strong achiever, capable of setting high goals
and reaching them. He can become very absorbed in his work, and may lose
sight of his interpersonal commitments. On the Character Matrix, Robert
 PROFILE ANALYSIS 131

falls into the high Conscientiousness-low Agreeableness vector. Here we see

an individual who has set high, socially valued goals for himself, and who
may pursue them in a guileful manner. Given the drive and focus of
Robert's life, as well as the importance he has attached to his own goals,
reaching them becomes important. He is willing to do what it takes to reach
these ideals even if it means cutting some comers along the way. Finally, if
Robert were a candidate for psychotherapy, his high levels of Openness and
low levels of Extraversion would make him an appropriate candidate for
some type of insight-oriented therapy. He would certainly enjoy the experi-
ence of sounding his inner world. The inner-directed, introspective nature of
these types of therapy are very consistent with his temperament.

CASE HISTORY: ANGELA W.

Angela W. is a 42-year-old, White, married female who is diagnosed
with a Borderline Personality Disorder. She presents with many problems,
including suicidal ideations. She has repeatedly injured herself in a number
of ways, including cutting and burning herself. She has been on drugs for
10 years and abused alcohol for at least 5 years. She is also a compulsive
spender, having spent her husband's credit cards to their limits several
times. Even when everything is confiscated, she continually opens new
credit card accounts and then proceeds to " max them out." She reports
being a victim of child sexual abuse. Her profile is presented in Figure 4-2.
Evaluating her NEO PI-R profile shows much variability over the do-
mains and facets. Most noticeably, her Neuroticism domain score is off the
chart. She is experiencing an extremely high level of negative affect which,
as can be seen from her facet scores, spans the entire spectrum of this di-
mension. She is highly anxious, depressed, hostile, with strong feelings of
inadequacy, she is impulsive, and does not cope well with any types of
stress. Such individuals find themselves unable to deal with any type of
emotional difficulty.
With such a high level of distress, I consider it important to evaluate
scores on Conscientiousness, which serves as a type of "control" dimen-
sion. Conscientiousness assesses the ability of the individual to manage
and organize his or her worlds, despite the presence of any emotional dis-
tress or dysphoria. Angela's very low levels of Conscientiousness suggest
that she will seek immediate gratification and succorance for her needs.
Her behavior will appear erratic, undirected, and impulsive. When she is
in pain, she seeks any outlet for relief, whether it is spending money, at-
tempting to hurt or kill herself, or abusing controlled substances. For An-
gela, there needs to be relief, now. It is not uncommon to find this pattern
of high Neuroticism and low Conscientiousness among people diagnosed
 132 CHAPTER 4

FIGURE 4-2. NEO PI-R profile of Angela W.

with Borderline Personality Disorder. But note the extreme level of func-
tioning on these two dimensions; facet scores on these two domains are all
in either the "very high" or "very low" range.
Overall her levels of Extraversion are quite low, suggesting that An-
gela prefers to be alone. She has few social contacts and may feel isolated
and detached from others. She has no one to tum to for help, and therefore
seeks more solitary pursuits for self-gratification (e.g., going on spending
sprees). Note that the very low levels found with the facet scales reflect a
defect in her ability to establish social relationships and bonds. Scores in
this extreme high or low range are indicative of an aberrant level of func-
tioning, especially when all the facets in the domain are extreme. Such ex-
tremity reflects more than just a strong tendency to express a particular
trait; there are real defects in functioning.
Concerning Openness, she again scores in the low range. Interestingly,
she scores high on 01 (Fantasy), suggesting that she has an active inner
life. This inner world may serve as an escape for her from the pain she ex-
periences. It may be here that she is able to strike back at those whom she
feels have injured her in the past by acting out her anger on her mind's
stage. Her low scores on 04 (Actions), 05 (Ideas), and 06 (Values) shows
 PROFILE ANALYSIS 133

her to possess a high degree of rigidity. She prefers structure and familiar-
ity. She is concrete in her thinking with a strong set of values which she ad-
heres to strongly.
Agreeableness is in the low average range, but this score may be due to
the very high AS (Modesty) facet score. She is certainly not trusting of oth-
ers (A1: Trust) and does not think warmly of others (A6: Tender-Minded-
ness). She is also just in the low range on A4 (Compliance), indicating that
she likes to do things her own way. The combination of low Compliance
and low Assertiveness (E3) presents the possibility of some type of passive-
aggressive behavior. Individuals with this combination like to follow their
own lead and have things done their way. However, being low on As-
sertiveness means that the individual may not have the capacity to speak
up for herself in the presence of others. Because the unassertive person likes
to take a back seat to others, when the group moves in a direction that the
individual does not like, there is little recourse but to resist in noncon-
frontational ways.
Angela's high Modesty (AS) score suggests that she may present her-
self in a nonassuming manner. Given her high scores on Neuroticism, such
modesty may be used to protect her already impaired sense of self-esteem
and value. Her modesty may help to keep others from looking critically at
her and protect her from any additional psychological insults.
Angela's Conscientiousness was discussed previously. Such very low
scores over the entire domain suggest a high degree of impulsiveness and
self-seeking behavior. Individuals low on this domain jump at any oppor-
tunity that is convenient and holds the promise of gratification.
In examining Angela's-scores in the context of the matrices, many ad-
ditional insights emerge. Concerning the Interpersonal Matrix, she would
certain fall in the low Extraversion quadrants, but given her average level
of Agreeableness I would assign her to the "high" category. In many ways
this category captures her behavior. She is certainly passive and shy. She
may present as being timid and bashful. In a group situation she would
prefer to have a very peripheral role. Although outwardly she may acqui-
esce to requests made of her, her low Compliance indicates that if she does
not want to move in that direction, she will resist in nonconfrontational
ways. The likelihood is that Angela would oppose trying new experiences
given her low scores on the Openness facets (e.g., 04, 05, and 06). This
pattern of results shows that Angela has a highly structured world with
clear preferences. She is not likely to modify them or experiment with
other options. Unless one approaches her through her narrow confines, it
is unlikely that she would be cooperative.
On the Emotional Well-Being Matrix she can unambiguously be clas-
sified in the high Neuroticism and low Extraversion category. These are
134 CHAPTER 4

individuals with a very poor sense of well-being. Easily overwhelmed and

distressed, these individuals may feel unprepared to manage the pressures
of life. Angela certainly is negativistic, insecure, and self-critical. She expe-
riences a profound sense of emptiness that is reflected in her many suici-
dal attempts. Her extreme scores on these domains, and across their
respective facets, suggests a very dynamic profile in need of close clinical
scrutiny.
In the clinical setting Neuroticism and Conscientiousness are key do-
mains for evaluating the potential of an individual to act out their issues.
Angela's high Neuroticism and low Conscientiousness signal a person
quite likely to strike out unpredictably. Individuals with this pattern are
very self-indulgent and impulsive. As I noted earlier, Conscientiousness
serves as a control mechanism, evidencing the ability of a person to man-
age and direct their inner turmoil. For me, in many ways Conscientious-
ness is very similar to the Blocks' notion of Ego-Control (}. H. Block &
Block, 1980). High levels of Conscientiousness reflects someone who is
ego-controlled, that is, an individual who is constrained and inhibited,
manifests needs relatively indirectly, and can delay gratification. Low Con-
scientiousness corresponds to the ego-undercontroller and reflects some-
one who can "manifest needs and impulses relatively directly in to
behavior ... to readily manifest feelings and emotional fluctuations, ... to
be distractible ... and to live life on an ad hoc, impromptu basis" (p. 44).
When levels of Neuroticism are high, it is important to consider levels
of Conscientiousness. When this latter construct is high, then one can have
confidence that the respondent is not likely to act out inappropriately. This
person would have internal mechanisms of control capable of keeping
impulses internalized. Low levels of Conscientiousness should raise the
concern about impulsive thoughts becoming expressed. If there are self-in-
jurious ideas, then one must seriously consider the likelihood that the in-
dividual will act on them. If levels of Agreeableness are also low,
physically assaultive behavior is possible. The extremeness of Angela's
scores on these domains strongly indicates the likelihood of her striking
out when under stress, towards either herself or others.
Evaluating Angela on the Character Matrix requires some subtlety.
Clearly she is low on Conscientiousness, but her Agreeableness domain
score is low average. It is likely that her average score is a result of the very
high A5 (Modesty) facet score. All her other scores are in the low average
to very low range. Because of this, I would place her in the low Agreeable-
ness category. Her placement in this category certainly provides a useful
description of her behavior: very self-centered and careless. Such individ-
uals see things only in terms of what it means to them. There are some
antisocial-type features in this category. This is typified in Angela's ongo-
 PROFILE ANALYSIS 135

ing deceit and manipulativeness in acquiring credit cards and then consis-
tently exceeding her limits.
Finally, on the Treatment Outcome Matrix, Angela falls squarely into
the low Openness, low Extraversion category. She may respond well to a
direct, nonemotional approach. However, given her emotional dysphoria
and self-centered approach, she will certainly be a therapeutic challenge.
In measuring therapeutic success, one may need to use a very modest mea-
sure. At one point in her treatment, she came"in actively hallucinating and
delusional. She went into a psychiatric hospital, but treatment there was
not working well. She was transferred to a substance abuse program for
about 3 weeks. She nearly died during withdrawal. When she came out of
this program she was reasonably stable for about 1 to 2 weeks, but then re-
verted to her initial behavior.
Another interesting point about this profile centers on the substance
abuse. As noted in the previous chapter with Joe W., compulsive substance
abusers usually score high on NS, ES, and low on either CS or C6, or both.
This pattern was not found for Angela; she was high on N5 and low on
both C5 and C6. Her ES score was very low (although it was much higher
than her other facet scores for this domain). That she does not conform to
the typical substance abuse pattern casts some interesting light on the role
drugs have played in Angela's life.
The NS, ES, C5/C6 pattern represents an individual who is impulsive
and uses drugs as a thrill-seeking adventure. The drugs provide a
"charge" for the person, a source of stimulation that offers a distraction
from the more mundane qualities of the person's world as well as a source
of pleasure and sensual gratification. Angela's personality pattern lacked
the excitement-seeking aspect of this profile. Rather than a distraction or
source of adventure, drugs may have been used as a tool for self-medica-
tion, an attempt to soothe and anesthetize the high levels of emotional pain
she felt inside. The drugs may also have served as an immediate outlet for
her self-destructive impulses. Despite heavy drug use and addiction, An-
gela may not be like the typical drug abuser.

CASE HISTORY: BRIAN M.

Brian M. is a 40-year-old White male, father of three young children.
He works in a white-collar management position and does not have much
work ambition. Presenting symptoms are chronic depression with extreme
pessimistic thinking. He is preoccupied with dying although he is not sui-
cidal. He stays alive for his children. From an Axis II framework he was
given the diagnosis of Schizotypal Personality Disorder by this therapist.
Such individuals are characterized as having odd or eccentric speech
 136 CHAPTER 4

and/ or beliefs. As we saw in Chapter 2, Table 2-6, such individuals tend to

have high scores on Neuroticism and low scores on Extraversion and Ag-
greeableness. We want to see whether this pattern is present with Brian.
Completely enraged with his family of origin, he sees his parents as
very horrible people, imbued with evil. In reality, they are very average
people. According to his spouse he interacts with them very well, although
he maintains these very bizarre thoughts and fantasies about them. He has
been on a variety of antidepressants, but does not respond well to them.
He took the NEO PI-R toward the end of treatment when he was being
seen for marital counseling. His profile is presented in Figure 4-3.
As can be seen, Brian scores very high on Neuroticism, with all scales
elevated except N5 (Impulsiveness). He clearly experiences a high level of
negative affect, induding anxiety, anger, depression, and low self-esteem.
He does not cope well with these feelings, easily feeling overwhelmed by
both environmental stress and the strong negative affects he experiences.
His overall Extraversion score is very low, suggesting a very restricted
interpersonal circle. He is reticent, desiring to avoid group activities, and
when in the presence of others would prefer to have a low profile, being
very passive. He has a very slow personal tempo; he likes to take the
course of events at a leisurely pace.

N E o A c
N E 0 A C

~
.2'
J:
~
~ 7D:

FIGURE 4-3. NEO PI-R profile of Brian M.

 PROFILE ANALYSIS 137

His overall Openness score is low, suggesting a closed, emotionally

rigid individual. His very low 03 (Feelings) score shows that he has very
restricted affect. There are many types of emotions that he does not like to
deal with or express. This accounts for the noticeable degree of flat affect
that characterizes him. Not very empathic, he is unable to resonate with
the feelings of others. The low 04 (Actions), 05 (Ideas), and 06 (Values)
suggests a very rigid individual who prefers much structure and routine.
He is oriented to the bottom line, and tends to focus on a limited number
of topics. Curiously, his 01 (Fantasy) score is in the high range. This sug-
gests a well-developed and active inner life. This is consistent with the
idiosyncratic and bizarre thoughts he maintains about his family of origin.
All of the fantasies that he has about them are certainly well developed
and reflect this capacity to elaborate and enhance daily experiences.
Brian's overall Agreeableness score is in the average range, although
there is a high level of intertest scatter on the facets. Modesty (AS) is his
highest score. He presents himself as very nonassuming and humble; he
appears very nonthreatening. However, he is very Iowan Trust (Al), Ten-
der-Mindedness (A6), and Altruism (A2). This pattern portrays an indi-
vidual who is very suspicious of others and their motives. He does not care
much for people in general and would not exert himself to reach out to
help if called on. People may be seen as very threatening and hurtful.
Finally, Brian's overall Conscientiousness is low, suggesting an indi-
vidual with a lackadaisical attitude, who may seek immediate gratification
for his needs, and who may have a tendency to act out. His lowest facet
scale is C4 (Achievement). He certainly is not a competitive individual,
does not set high standards for himself, and may not have a sense of di-
rection. This is consistent with his very unconcerned attitudes about work
and career. He really does not care much for work and is not driven to suc-
ceed there. He is low in Competence (C1) and Order (C2), suggesting that
he feels worthless and unable to successfully meet the challenges he may
encounter. He is not fastidious or well organized. His low Dutifulness (C3)
score suggests that he does not follow through on his commitments and
obligations. Interestingly, he scores high on Deliberation (C6). Brian cer-
tainly will not act without thinking things through; he tries to be careful.
Coupled with his low Impulsiveness (N5) score, one can be confident that
Brian has some control over his impulses. Although it cannot be ruled out
that at some point Brian may act against his parents or himself, such an
outburst will not be a situationally provoked event. Rather, Brian will
spend some time thinking through his actions.
Overall, the profile here matches the kinds of symptoms and issues
Brian presents. The bizarre fantasies about his family and the idiosyncratic
ideation he presents when discussing them may be working out of his high
 138 CHAPTER 4

Fantasy (01) score. The flat affect or anhedonia is certainly characteristic of

someone low on Feelings (03). The pattern of high Neuroticism and low
Extraversion is indicative of someone who is experiencing much distress
and dissatisfaction in his or her life. There is little positive feeling or sense of
well-being. Interpersonally, Brian has a Scrooge-like attitude toward people.
He is untrusting and uncaring, perhaps perceiving others as potential
threats. His high Modesty (AS) score coupled with his low facet scores on
Extraversion would suggest that he presents himself in a very nonthreaten-
ing manner. He tends to keep to himself and takes a back seat in group ac-
tivities. His nonassuming demeanor belies the churning emotions that exist
below the surface.
From a Character perspective, Brian takes a selfish orientation toward
life, concerned more with his own issues than with those who surround
him. His low 06 (Values) facet score suggests that Brian has a very clear
ideological framework from which he operates, although his low C3 (Du-
tifulness) score suggests that he may not be consistent in how he applies
these values. Still, it would be important to try to understand his value net-
work and its related imperatives. This would be important in understand-
ing how he orients himself to the world.
From a therapeutic perspective, Brian presents numerous challenges.
He is very concrete and emotionally limited. Although his fantasy world is
rich, Brian may not be able to fully express the emotions that play them-
selves out in his mind's eye. A therapist would need to take a very focused
and directed approach to Brian. His high levels of Neuroticism certainly
complicate whatever treatment is employed. There is tremendous dyspho-
ria indicated, and this may not improve greatly over the course of therapy.
It may be possible to tactically manage issues with Brian's family of origin
and his current marital issues. However, on a strategic level, there may al-
ways be a strong sense of inadequacy, anger, depression, and distrust.
His detached interpersonal style is another factor that undermines im-
provement. Brian's preference to form few relationships deprives him of
important opportunities to draw on coping resources. The presence of oth-
ers can help provide an additional sense of support and succorance for
Brian's negative feelings and hostile attitudes toward his family. His re-
luctance to make such contacts isolates him and encourages his further re-
treat into his overpersonalized inner world.
It is interesting to note that the overall profile matches what one
would expect for a Schizotypal Disorder. He scores very high on the Neu-
roticism facets, low on all the Extraversion facets, and low on all but one of
the Agreeableness facets. The high Modesty (AS) score is interesting here
because it is so distinct from the other facets on this domain and it also
serves to give him an overall average level on the domain. In evaluating
 PROFILE ANALYSIS 139

the high AS score, one should turn to the N4 (Self-Consciousness) score,

which is in the high range. His desire not to speak out about himself may
be due to his perceived negative self-image. Thus his modesty may be a
consequence of his belief that he has nothing of value to share with others.
The low Trust (AI) and Tender-Mindedness (A6) scores are certainly char-
acteristic of the Schizo typal, who may fear others and seek punishment for
those who may be threateaning.
The high Fantasy (01) and low Ideas (05) scores are also relevant here
to the Axis II issues. Certainly Brian has an active inner world where is-
sues, people, and experiences are all elaborated on. These internal explo-
rations may lead to the development of bizarre ideas and feelings, but the
internal rigidity characterized by the low scores on the other facets sug-
gests that such ideas may be difficult to remove.

CASE HISTORY: BARBARA W.

Barbara is a 35-year-old married woman who is very depressed and
resentful. She came into therapy to deal with sexual harassment in the
workplace. She is very unhappy with her marriage and her life. She is em-
ployed as an office worker, and is very well organized and successful in
this position. She is appreciated by her superiors and well thought of by
her colleagues. She took the NEO PI-R prior to revealing the sexual issues
at work. Her response to therapy (which was cognitive-behavioral) has
been successful, and she is able to manage her issues of despair and feel-
ings of sexual abuse. She is now moving toward dealing with more long-
term issues concerning her ongoing feelings of emptiness. She is very
duty-driven and lives her life by obligation. She needs to learn how to take
time for herself. When she was very depressed, the only thing that kept her
alive was her responsibility for her children. Marital therapy is effective in
helping to obtain more emotional support from her husband. Her profile
is presented in Figure 4-4.
As can be seen, Barbara has an extremely high level of Neuroticism,
suggesting a tremendous amount of negative affect covering a broad range
of dysphoric feelings. All of her facets are in the high to very high range,
indicating that her negative feelings are indeed pervasive. She is depressed
and anxious, easily angered, possesses low self-esteem and certainly is
being bothered by recurring negative thoughts. Most important, though, is
the fact that her Vulnerability score is off the chart; she certainly feels that
she cannot manage the stresses that she confronts. She is easily over-
whelmed by the pressures of her life as well as her inner turmoil.
On Extraversion Barbara scores very low, indicating a solitary orien-
tation. She prefers to keep to herself, having few friends and social con-
140 CHAPTER 4

FIGURE 4-4. NEO PI-R profile of Barbara W.

tacts. She projects a stoical image to others, having few positive feelings or
emotions. Interestingly, Barbara scores very high on E4 (Activity), indicat-
ing a rapid personal tempo. She likes to stay on the move doing things.
The combination of very high Neuroticism and very low Extraversion in-
dicates an individual with a rather poor sense of well-being. She feels in-
secure and vulnerable; she finds little support and comfort from her life.
No doubt there is a strong feeling of emptiness and despair.
Scores on Openness are also low overall, suggesting a degree of per-
sonal rigidity. Barbara scores low on Fantasy (01), Actions (04), and Val-
ues (06). This indicates an individual who very much desires structure
and form. She has a clear sense of right and wrong, possibly appearing to
others as conventional and conservative. There is little imagination, but
more of a down-to-earth, black-and-white orientation. It is interesting to
note her high score on Activity (E4). She likes to stay busy, but the low
score on Actions (04) indicates that she invests that energy in a very cir-
cumscribed routine. Such a restricted range of activity may help bind the
high levels of negative affect that she experiences.
Barbara scores in the high range on Aesthetics (02) and Feelings (03).
The latter score indicates that Barbara is open to a wide range of feelings,
 PROFILE ANALYSIS 141

and would suggest that she experiences her negative affect with a greater
level of sensitivity than someone who is low on this facet. The Aesthetics
score suggests a further ability to be emotionally moved by artistic experi-
ences. The imagination and creativity suggested by a high score on this
facet stand in contrast to her low scores on Values and Actions. This com-
bination of scores suggests that Barbara, although appearing formal and
staid, can be moved by certain imagery, perhaps of a religious nature.
Barbara's score on Agreeableness is in the high average range, indi-
cating that she does feel some compassion for others, but these feelings of
care are certainly guarded. She scores low on Trust (AI) and Tender-Mind-
edness (A6), indicating a suspiciousness of the motives of others. She be-
lieves that the pain others may experience is likely to be a product of their
own misplaced efforts. She scores high on Straightforwardness (A2), indi-
cating that Barbara tends to be frank and sincere toward others. It is inter-
esting that her Assertiveness (E3) score is so low. This suggests that
Barbara is not proactive in sharing her thoughts. She is rather shy and ret-
icent in social situations; one may have to directly query her to find out
what her feelings are. Once asked, she would be forthcoming, but an invi-
tation needs to be offered.
The combination of high scores on Straightforwardness (A2) and Al-
truism (A3) and the low score on Assertiveness (E3) may mean that Bar-
bara is vulnerable to exploitation. She may wish to reach out to others less
out of a sense of social responsibility than out of compassion. She may feel
compelled to reach out altruistically to others, even though she knows that
they may take advantage. Perhaps her high levels of Neuroticism may pre-
vent her from appropriately establishing boundaries to her helping be-
havior. She may be seeking her own personal sense of solace from helping
others, and the strength of her needs for reassurance and comfort may
override her natural suspiciousness.
Barbara also scores high on Modesty (AS), indicating that she is mod-
est and unassuming. Again, the picture that is emerging of Barbara is of a
woman who is quiet and reticent. She prefers to stay in the background
and not attract any attention to herself. If offended, she may not respond
directly. Instead, she may wait for some invitation to speak. Then she may
voice her issues. In short, one should not mistake her outward quiet and
calm for assent. There may be many issues percolating within her that are
not readily apparent. This is evidenced in the therapeutic situation where
the most pressing issue for her was the apparent sexual discrimination she
experienced at her job. These events distressed her greatly, yet it was some
time into the treatment process before she brought it up. Some of this reti-
cence may be a function of her low Trust; she needs to feel safe and secure
before she shares her intimate feelings.
142 CHAPTER 4

Her low average score on Compliance (A4) suggests that Barbara

would prefer to do things her own way. Given her strong commitment to
her value system and her high need for structure and routine, it may be
very important that events proceed in ways that are consistent with her ex-
pectations. A relatively low Compliance score in conjunction with a low
Assertiveness score may indicate passive-aggressive tendencies as well.
The strong need to follow her routine coupled with an equally strong need
to shrink from social leadership roles creates a conflict for Barbara when
events begin to move in directions that are inconsistent with her values.
Rather than initiating some response, she is prone to acquiesce, not want-
ing to create a "situation." She may therefore try to sabotage the process in
an oblique manner.
Barbara scored high on Altruism (A3), indicating an active concern for
the welfare of others. This is interesting given her very low scores on Trust
(AI) and Tender-Mindedness (A6). This may be indicative of Barbara's
commitment to "duty," an acute awareness of social responsibilities that are
defined by her value system. She helps others not out of a feeling of deep
compassion or empathic concern (even though she does score high on 03-
Openness to Feelings). Rather, she has a sense of obligation that needs to be
fulfilled. Her assistance is more perfunctory than compassionate.
On Conscientiousness Barbara scores extremely high, suggesting
some serious psychological issues. She has a high degree of personal re-
sponsibility and organization. She sets very high standards for herself and
strives hard to attain them. Although she may experience a number of im-
pulsive ideas, her high level of Conscientiousness argues against her act-
ing on them without due consideration of their consequences. She has a
high degree of personal competence and is well organized. Interestingly,
her lowest score on this domain is CS, Self-Discipline. She believes that at
times she is distracted from her obligations and indulges herself. This may
be a source of guilt in her life, given her high levels of Depression, Impul-
siveness, and Vulnerability, as well as her strong attachment to her value
system. Given her high score on Achievement Striving (C4), she no doubt
sets very high, perhaps unrealistically high, standards for herself to attain.
She may not feel that she is able to reach these goals all the time and one
may want to evaluate to what degree she may be self-punishing.
Her high score on Dutifulness (C3) fits well into the developing picture
of this client. Individuals high on this scale follow through on their commit-
ments. This scale represents the "superego" function; individuals scrupu-
lously adhere to their moral principles. The combination of low 06 (Values),
low 04 (Actions), high A3 (Altruism), high C3 (Dutifulness) and high C4
(Achievement) all portray a moralistic, rigid, determined individual who as-
pires to very high, perhaps unrealistic, standards. There may be an unre-
 PROFILE ANALYSIS 143

lenting quality to her endeavors, given the high E4 (Activity) and high C4
(Achievement Striving) scores. As noted in the description of her case, she
is very duty-bound, leaving little or no room for her own needs and desires.
She was able to thwart her suicidal issues by thinking of her children. No
doubt, she was reminded of what a "good mother" should be like.
An interesting combination in this profile is the high score on N4 (Self-
Consciousness) and the high score on C1 (Competence). The former indi-
cates a low level of self-esteem and feelings of inferiority. Yet she scores
high on Competence, suggesting a good sense of self-efficacy. Such a com-
bination of scores usually suggests an individual who has found a partic-
ular niche in which they feel very comfortable and competent, often
receiving much recognition for his or her performance. However, these in-
dividuals may be very reluctant to move from this environment to another,
new one. Increased responsibilities, new job reporting structures, and so
on, can all induce a high degree of anxiety and feelings of inadequacy. Ex-
pertise in one situation is not perceived as generalizable to new situations.
These individuals feel very uncomfortable with the change, but over time
will recapture their feelings of competence in this new niche, until the next
time they have to move (the Impostor Phenomenon is a related construct;
see Clance, 1985; Clance & Imes, 1978).
Another point of interest in this profile is the combination of high Self-
Consciousness and high Achievement. This is a very powerful combina-
tion of constructs. High achievement is associated with individuals setting
higher personal standards for themselves. They aspire to reach goals that
are socially valued. High achievers can become quite absorbed in their ef-
forts and ambitions and find success to be quite satisfying and rewarding.
However, a corresponding high score on Self-Consciousness adds another
dimension to these aspirations. Because of the underlying feelings of in-
adequacy, the competitive strivings become an attempt by the individual
to find redemption for their perceived inadequacies. The very high stan-
dards such individuals set for themselves are in compensation for the
weaknesses they feel.
As was noted in the "Competitive Performance" section, individuals
with high N4 and high C4 perceive success as bringing relief from their
inner pangs of deficiency, but this respite is only temporary. Soon the need
for new successes rises and the quest to achieve begins again. In this
process, failure becomes an awful experience to be avoided because it pro-
vides some external confirmation of the weaknesses kept hidden within.
This is similar to Horney's (1950) notions of the "search for glory" (p. 23),
the neurotic's attempt to reach the comprehensive neurotic solution-that
fantasied image that, when attained, will solve all the problems. But this
image of success is unattainable. So the individual high on these two facets
144 CHAPTER 4

pursues his or her goals with the affective intensity of a desperate individ-
ual. Barbara may be looking for validation and approval through her ef-
forts to live up to high, socially commendable criteria of success. Being
"the best" is more than just personal development; it is an effort to find
personal justification. Her successes provide an emotional high point, her
failures an affective nadir.
From a therapeutic standpoint, treatment needs to consider that such
high levels of affective dysphoria may not be entirely ameliorated. There
may always be high levels of negative affect, whether depression, anxiety,
self-consciousness, or anger. These feelings are characteristic of Barbara,
and treatment should focus on techniques that can help her deal better
with her internal distress. The high levels of Conscientiousness are an asset
for the therapist. Barbara will work hard toward her therapeutic goals and
can be counted on to follow through with the treatments on her own time.
Her lower levels of Openness suggest that a more focused, behavioral path
may be most effective. Although she is open to experiencing her emotions,
whether she feels comfortable speaking about them may be a function of
her values and level of trust and comfort with the therapist. Marital ther-
apy may be ideal for Barbara because it gives her a much needed interper-
sonal context with an individual with whom she should have some level
of trust and comfort.
H should be pointed out that despite the presence of so much affective
and personal distress, Barbara is able to function in her environment. Al-
though one may be tempted to see much more impairment in such ex-
treme scores, it is important to remember that the NEO PI-R is a measure
of normal personality dynamics. High scores on these domains are not di-
rect indicators of psychopathological functioning. Although most of the
cases presented here are clinically oriented and therefore such high scores
are not surprising, it must be remembered that one can have high scores on
these domains and not have any type of discernible pathology.

CASE HISTORY: BEVERLY N.

Beverly N. is a 38-year-old woman seen in treatment for social phobia.
She has been diagnosed by her therapists as having a severe Avoidant Per-
sonality Disorder. When she first entered treatment, her avoidance behav-
iors were so severe she would vomit before going to her part-time job
several times each week. She was terrified of facing the public, even
though she was a receptionist. Beverly dropped out of college because she
was unable to stand up and give presentations in class. She is a religious
individual who is very conscientious. She is a good wife and mother. Be-
havioral therapy and cognitive work was effective for the avoidance be-
havior. Her NEO PI-R profile is presented in Figure 4-5.
 PROFILE ANALYSIS 145

N E 0 A C
N E 0 A

f
:I:
as'
·m
""
:I:
:- ~

.f
:
.
iil
CD

." i

if
<
i

FIGURE 4-5. NEO PI-R profile of Beverly N.

As can be seen, Beverly scores very high on Neuroticism, indicating a

large amount of negative affect. The facet scores indicate that this negativ-
ity is centered around issues of Anxiety, Depression, and Self-Conscious-
ness. Interestingly, she scores in the average range on Vulnerability,
indicating some capacity to cope with the stressors she encounters. Thus,
although suffering from a severe social phobia (high N4: Self-Conscious-
ness), she is able to manage these profound feelings sufficiently to obtain
and keep her part-time job.
On Extraversion Beverly scores very low, suggesting a very detached,
socially isolated world. She prefers to have very few interpersonal con-
tacts, and would not be considered very "approachable" (low scores on El,
Warmth and E6, Positive Emotions). She is very passive, although she has
an average level of personal energy (E4, Activity). Like Angela W., Beverly
does not experience much personal well-being; there is not much per-
ceived quality of life.
Overall Openness is in the high range, suggesting a degree of personal
flexibility and originality. Beverly is particularly high on Aesthetics, easily
moved by the arts. She is also high on Ideas (05), suggesting that she likes
to look at the "big picture"; she is theoretically oriented and broad based
in her thinking. Her average score on 06 (Values) suggests that Beverly
 146 CHAPTER 4

has some values that are important to her and toward which she is firmly
committed. However, there are other aspects of her belief system that are
much more flexible and open to evaluation and change. She is low on 04
(Actions), indicating a desire for structure and clarity. She prefers to follow
an established routine and does not do well in unstructured situations. She
is also low on 01 (Fantasy), which indicates that Beverly prefers to keep
her mind on the tasks at hand.
She scores in the very high range on Agreeableness, indicating a very
prosocial, compassionate orientation toward others. She is very trusting
of others, tending to see the best in them. She is direct and honest in re-
vealing information about herself; she tends to tell others how she feels.
However, like Angela, her low Assertiveness score indicates that she may
not be very proactive in making her ideas or wishes known to others. She
would prefer to wait to be asked rather than to take the lead and initiate
a conversation. She is very Altruistic, ready to help others when called
upon. The key term here is "called upon." She may not initiate a response.
High on Modesty and Tender-Mindedness, Beverly is self-effacing and
easily moved by emotional appeals. She is friendly, soft-hearted, and gen-
tle, never overbearing or arrogant. She scores in the average range on
Compliance, suggesting that at times she likes to have things done her
own way. Given her low score on Assertiveness, she may resort to
passive-aggressive strategies.
The domain score on Conscientiousness is in the very high range,
indicating an individual who is very ambitious, organized, reliable, and
self-controlled. She is high on all facets except Achievement (C4). She
has a good sense of Competence (Cl), although given her N4 score,
these feelings may be experienced within a certain well-defined niche.
She may have difficulty in generalizing these feelings to other situa-
tions. She is Orderly (C2) and Dutiful (C3); she is efficient and orga-
nized, and can follow through on commitments that she makes. Scores
on the C5 (Self-Discipline) and C6 (Deliberation) scales, along with an
average N6 score, show that Beverly is not likely to act out impulsively
on any of the negative feelings she experiences. She has enough per-
sonal control and discipline to manage these dysphoric feelings in ap-
propriate ways.
It is interesting to note that Beverly's avoidant personality exists de-
spite her very high Agreeableness score. Although she is afraid of being
with others, she sees them as basically good. Her fears of interpersonal
contact may stem from very deep feelings of inadequacy and real fear of
social ridicule. She may feel truly conflicted, liking people and possibly
wanting to be with them, but afraid that she may not live up to their stan-
dards or may falter interpersonally. The high level of Neuroticism is con-
PROFILE ANALYSIS 147

sistent with the Avoidant Personality Disorder diagnosis and not with the
Schizoid Disorder.
Therapeutically, Beverly's high Openness and Conscientiousness
scores augur well for treatment. She is capable of thinking in new ways
about herself and will work hard at reaching the treatment goals. Given
her scores on the Agreeableness facets, it should be relatively easy to es-
tablish a therapeutic alliance with her. Her relatively low N6 score is an ad-
ditional resource that Angela W. does not have. Beverly already has
developed an important sense of being able to cope with the difficulties
that confront her. Despite her fear about being with others, she was still
able to take on, and perform well, a pOSition as, of all things, a receptionist.
This job certainly puts her in a position to meet and work with others, a
natural outlet for someone high on Agreeableness. Yet she is able to man-
age the terror this position provokes in her. This is an important founda-
tion on which therapy can build.

CASE HISTORY. TOM S.

Tom S. is a 34-year-old, born-again Christian fundamentalist. He has
a severe gambling disorder which led to the breakup of one relationship.
He is working as a salesperson. He presents as a very likable person, a very
good conversationalist. He came to therapy because he is engaged to be
married and his fiancee was curious to know how trustworthy he was. He
was abstinent from gambling for about 60 days before coming to therapy.
He was in treatment 4 weeks before taking the NEO PI-R. His results are
presented in Figure 4-6.
Tom's overall domain score on Neuroticism is in the high range,
showing a broad range of negative affect. He claims to be anxious, easily
frustrated, depressed, having low self-esteem, and distressed by thoughts
about doing things that he knows he should not. The only Neuroticism
facet in the average range is N6 (Vulnerability), suggesting that Tom can
manage his distress to some degree.
His domain score on Extraversion is in the average range, indicating
that Tom can do well in groups, although he appreciates his privacy. His
lowest facet score is on E3 (Assertiveness), indicating a desire to let others
provide initiative in group situations. He scores in the high range on the E5
(Excitement-Seeking) and E6 (Positive Emotions) facets. He exudes a cheer-
ful and optimistic persona; others may perceive him as being humorous,
enthusiastic, and jolly. People may seek out his company to enjoy the posi-
tive energy he radiates. His high score on Excitement-Seeking indicates an
enjoyment of thrill-seeking activities and being where lithe action is." He
may be perceived as being charming and spunky. Some may even consider
148 CHAPTER 4

FIGURE 4-6. NEO PI-R profile of Torn S.

him to have some degree of charisma. However, despite this personal" en-
ergy," Tom may lack a good sense of personal well-being. His levels of neg-
ative affect may overwhelm any positive feelings he may have.
Openness to Experience is in the average range, suggesting some de-
gree of personal flexibility, creativity, and spontaneity, although there are
aspects that call for some need for structure and conventionality. His high-
est scores are on the facets 02 (Aesthetics) and 03 (Feelings). The Aesthet-
ics facet relates to one's interest in art and beauty. Tom certainly can be
emotionally and physically moved by theater, poetry, and music. His high
Feelings score indicates an individual who is open to a wide range of emo-
tions within himself, and is capable of responding empathically to the af-
fective needs of others.
It is interesting to note that although he is a born-again Christian fun-
damentalist, he does not have the low 06 (Values) score that one may ex-
pect from someone who adheres to such a conservative ideology. Tom's
average level score on this facet suggests that the type of clarity and struc-
ture he wants in his religious values does not generalize to other aspects of
his life (e.g., political views). In some areas he may be very tolerant of di-
vergent viewpoints, and even enjoy some philosophical debates on those
 PROFILE ANALYSIS 149

topics. However, there are issues, possibly religious, where he needs to

have a strong sense of tradition.
Concerning Agreeableness, Tom scores overall, and on all of the
facets, in the average range. Such a pattern of scores indicates an individ-
ual who has a basically "pro-person" attitude toward others, but who may
not be uncritically accepting. He may initially approach others in a
friendly, accepting way, but may need some period of evaluation before
becoming fully welcoming.
Finally, overall Conscientiousness is in the low range, indicating that
Tom may not always be as reliable as he needs to be. He enjoys his free
time, may feel encumbered by commitments and responsibilities, and may
not always follow through on them (low C3, Dutifulness). There is also a
degree of distractibility (low C5, Self-Discipline); other, more enticing
events (especially those with an adventurous or thrill component-see his
E5 score) may lure him away from the "drudgery" of his work. The low
Conscientiousness also speaks to a degree of selfishness and a preference
for the immediate gratification of needs. He also experiences an impaired
sense of self-efficacy; therefore he may give up on tasks if he finds them too
difficult or demanding.
Tom's profile evidences the classic impulsive triad-high N5, high E5,
and low C5 and/or C6. This is consistent with his gambling problem,
which is certainly of an impulsive nature. He likes the "action" and excite-
ment that gambling brings. Even though he has been "clean" for more
than 60 days, his profile continues to manifest this triad, suggesting that he
may be likely to recidivate. Whatever treatment he has received for this
problem has not been effective in changing its underlying motivation.

CASE HISTORY; ERICA J.

Erica J. is a 38-year-old single female. She is the mother of a teenage
daughter and works part-time as a hair stylist. She is not presenting for
treatment. Interpersonally, Erica is an outgoing, engaging woman. She en-
joys the "club scene," going out frequently to meet new people. She has
never been married, although she was in a committed relationship for a 5-
year period. She found this relationship to be unfulfilling and eventually
broke it off. Since that time she has had a string of mostly superficial rela-
tionships and is hoping to meet that "special someone." Her sole source of
support is from her part-time stylist position, so money is always "tight."
She is able to make ends meet with the support of her mother. The results
of her NED PI-R are presented in the Appendix, which provides her com-
puterized report. This information was obtained from the computer scor-
ing software that is available for processing NEO PI-R responses.
 150 CHAPTER 4

The first page of the computer-scored document provides the basic

copyright information. Page 2 of this report contains a profile plot of the
client's scores. The third page of the report lists the T-scores for each of the
35 scales. These values have been standardized by gender and age. It
should be pointed out that the T-scores for the domain scores are based on
the validimax factors. (Keep in mind that T-scores have a mean of 50 and a
standard deviation of 10). These factors optimize the independence of the
domains and provide assessments of each domain based on all 35 facets.
The overall interpretation of Erica's NEO PI-R profile begins on page
4 of the report. The first step in interpretation focuses on domain scores. As
can be seen, she scores very low on Conscientiousness and high on Extra-
version. Scores on the remaining domains are in the average range. Global
interpretations of what these scores mean are provided.
On the fifth page of the report, a more detailed evaluation is given
based on facet scores. It should be pointed out that at times there may ap-
pear to be contradictions within the report between the overall and de-
tailed interpretations. The overall interpretations are based exclusively on
the overall domain score, without considering the facets. Thus, someone
may score low on say Conscientiousness and would be interpreted as hav-
ing low levels of order, competence, dutifulness, achievement orientation,
self-discipline, and responsibility-an across-the-board evaluation. How-
ever, it is possible that on some of those facet scales the individual scored
in the high range. Therefore, in the later section of the report, the printout
will say that the person is competent, or achievement oriented, or respon-
sible-the opposite of what was said earlier.
Although this may appear confusing, one has to keep in mind that
personality is hierarchically organized. As one moves from the broad to
the specific, more information becomes available and nuances can be de-
tected. High, or low, domain scores provide overall assessments of the re-
spondent, but do not argue for uniform facet scores; there can be
tremendous interfacet scatter within a domain. In Erica's case, we see ex-
amples of this. On Extraversion, she scores in the high range, suggesting
that she is very other-oriented, enjoys social contacts, and has a very up-
beat personal style and tempo. For the most part, her facet scales mirror
this pattern; all are in the high range except for Assertiveness (E4). Despite
such surgency, Erica tends to be passive in the presence of others.
Similar outcomes are found with Conscientiousness, where her do-
main score was in the very low range. Erica has low levels of Competence,
Achievement Orientation, Self-Discipline, and Responsibility. In contrast
to that, she evidences average levels of Order and Dutifulness. Although
she may not be competitive and focused, she has a sense of personal orga-
 PROFILE ANALYSIS 151

nization and can be counted on, to some degree, to follow through on her
obligations. In working with the interpretive report, keep in mind that the
process of evaluation is working from the broad to the specific. If there are
any apparent inconsistencies, rely on the facet interpretations to be more
accurate than the domain evaluations.
Some of the psychosocial implications of the scores are on page 7 of
the computer report. Included in this discussion are sections on Coping
and Defense styles, Somatic Complaints, Psychological Well-Being, and In-
terpersonal Characteristics, Needs and Motives, and Cognitive Style. All of
this information was taken from the extensive construct validity research
that has been done using the NEO PI-R. Also included in this report is a
section concerning Axis II clinical hypotheses. Although the NEO PI-R is a
measure of normal personality, as shown in Chapter 2, there are strong
linkages between these dimensions and abnormal personality functioning.
What these associations mean is that high (or low) scores on any of the
NEO PI-R scales do not indicate pathology, but individuals with various
characterological disorders have distinct NEO PI-R profiles.
When dealing with a clinical client, this section may be useful for di-
agnostic formulations. When dealing with a nonclinical client, this section
may help outline salient characterological styles of a nonpathological na-
ture. Erica's profile is similar to those characteristic of the Histrionic and
Dependent Personality Disorders. This does not mean that Erica can be di-
agnosed as having either or both of these disorders. Rather, it should be
interpreted that she has an interpersonal style that seeks attention and suc-
corance from others. The final section provides some treatment implica-
tions for this profile, how the individual is likely to appear in therapy, and
issues that are likely to emerge.
The last page of the report (not reproduced here) provides a statement
of the answers recorded for each item. This is useful in case there is a dis-
agreement between the stated interpretation and any perceptions, either
by the interpreter or the client. When such disagreements arise, the first
thing to do is check the accuracy of the responses. Either the client made a
mistake in recording his or her answer, or the person entering the data into
the computer made a mistake. Another strategy is to examine the response
summary at the bottom of the page. This tells the percentage of scores for
each response category. Someone who answers "Strongly Agree" or
"Agree" to all or most of the items (an acquiescence effect) can be easily
identified. The number of items not responded to is also presented. A high
percentage here may invalidate the profile (no more than 40 items, 17%,
should be missing). In the case of Erica, the response summary suggests a
valid protocol, with no items left blank.
152 CHAPTER 4

CASE HISTORY: SALLY INES

Our final case history for this section is for Sally Ines, a 36-year-old
married female. She has two young children and is employed full-time in
the banking industry. She has an MBA degree and is currently a mid-level
manager for a large financial institution. She is not seeking treatment for
any problems. Her profile is presented in Figure 4.7.
As can be seen, Sally has a low level of Neuroticism, suggesting a
calm, emotionally secure, and hardy disposition. If there is any type of
negative affect in her life, it is likely to be either Anxiety (Nt) or Self-Con-
sciousness (N4). She is certainly not very Impulsive (N5) and copes very
well with the stressors she encounters in her world (low N6).
Overall she is average on Extraversion, suggesting a more flexible
need for interpersonal interaction. Her low Excitement-Seeking score (E5)
shows that she does not like thrills or risks. Sally scores in the high range
on both the Gregariousness (E2) and Assertiveness (E3) facets. This sug-
gests that Sally enjoys the company of others, and may prefer doing things
in groups. Her high E3 score shows that when with others she prefers to be
the one moving the group. She likes to be the leader rather than a follower.
The combination of low Neuroticism and relatively higher Extraversion

N E 0 A C

FIGURE 4-7. NEO PI-R profile of Sally Ines.

 PROFILE ANALYSIS 153

domain scores suggests that Sally feels a sense of personal well-being and
overall life satisfaction.
Interestingly, although Sally sees herself as being gregarious, she in-
dicates that she may have a more formal style in dealing with others (low
average Warmth score, El). People may not always find her to be easily ap-
proachable, and this may at times lead Sally to experience a sense of dis-
tance from others. She may enjoy being with people, but finds it difficult
that others may not immediately "warm up" to her.
On Openness Sally scores in the average range overall. This suggests
that although capable of being flexible and receptive to ideas, feelings, and
activities, Sally may evidence some degree of conventionality. An exami-
nation of her facets shows that the Aesthetics facet (02), is very low; that is
responsible for the average overall score. A score on 02 at this level sug-
gests an individual who is not concerned with art and beauty.
Sally scores high on 01 (Fantasy) and 04 (Actions). The former sug-
gests a developed inner world and an active fantasy life. This is consistent
with her strong interest in reading romance novels, which she feels pro-
vides her with a way of relieving tension and finding solace and enjoy-
ment. Her high Actions score indicates an ability to work in unstructured
situations and a desire for novelty and variety. As she likes to tell others at
her work, "Justifying doing something on the basis that it is the way we
have always done it is not acceptable."
Sally's score on 05 (Ideas) is in the low average range, suggesting a
bottom-line-oriented attitude. Although on one level this seems consistent
with someone who is employed as an accountant (and even adaptive),
usually individuals with advanced degrees score higher on this facet. Sally
may not be a "big-picture" thinker, but she is certainly concerned with the
nuts and bolts of any process, and focuses her energies on creating tangi-
ble results.
Sally's score on Agreeableness is in the low range, suggesting a more
suspicious, competitive orientation toward others. Sally can be skeptical
and critical of the intentions and motivations of others. Most of her facet
scales are also in the low to very low ranges. Al (Trust) is low, indicating
that Sally does not always see others as being honest or helpful. The low A2
(Straightforwardness) score indicates someone who is guarded in express-
ing her feelings. She may use flattery and craftiness to manipulate others. In
some ways this is an adaptive quality for a manager to have. Managers
need to be able to motivate individuals to reach established goals. Some
people may need a pat on the back, others may need to be patted a little
lower; some need cajoling, others confrontation. A good manager is able to
look past his or her own feelings about an individual and provide the kind
of feedback and incentives that will be useful to the other person.
154 CHAPTER 4

Sally scored low on the AS (Modesty) and A6 (Tender-Mindedness)

facets, indicating that she is quite able to talk about her accomplishments
and can promote herself when necessary. Her low A6 score indicates an in-
dividual who is not moved by emotional appeals and instead focuses on
rationality and logic. She scores in the average range on the Compliance
(A4) and Altruism (A3) facets. Sally may prefer to do things her own way,
and certainly can be interpersonally adept at developing support for her
agenda, but she can also take direction. Her A3 score shows that she can be
generous and willing to help when a request appeals to her rationality.
Overall, Sally's Agreeableness scores seem to be well suited for someone in
a management position.
Finally, Sally's overall level of Conscientiousness is in the low range,
indicating an individual who may be lackadaisical in moving toward
goals. Individuals low in Conscientiousness are not very ambitious, orga-
nized, or focused. Sally's lowest facet scale is Achievement Orientation
(C4), suggesting that she is not driven to succeed; she may not have her
eyes on attaining any type of upper management position, and may be
quite content with her current level. She prefers not to work excessive
overtime and tries not to bring any of her office work home with her. She
also scores low on Order (C2), indicating a tendency to be unmethodical.
Her highest facet scores are on the Competence (Cl) and Dutifulness (C3)
scales. Sally maintains an average sense of her value and worth, feeling
that she can accomplish those tasks that are assigned her. She can be
counted on to follow through on commitments that she makes, although
she may strive to do only what is minimally required. Levels of Self-Disci-
pline (CS) and Deliberation (C6) are in the average range and show Sally
to have some degree of personal organization and control.
Sally's configuration of scores on this domain show her not to be nec-
essarily sloppy and disorganized. She has other priorities and goals for
herself that may not necessarily be consistent with those of the competitive
business world, but may be concerned with her family life and other recre-
ational commitments. Nonetheless, she is very successful in her profes-
sional career. Mid-level management is the place in which she feels the
most comfortable. She can interact with others and manage them through
setting their goals, delegating tasks, and mobilizing others into coopera-
tive undertakings. At the end of the day, though, she can leave this all be-
hind and go home to devote her attentions to her family. Her Competitive
Style (see Table 4-3) is relaxed, informal, and easygoing.
Sally's Character Style (Table 4-4) reflects a more self-seeking and ma-
nipulative orientation. She enjoys getting other people to move toward
goals that she has selected. However, the descriptions in the low Agree-
ableness-low Conscientiousness category appear too extreme to describe
Sally. Her scores on the C3 (Dutifulness) and 06 (Values) facets indicate
 PROFILE ANALYSIS 155

someone who may be more conventional than the prototype suggests.

Again, this underscores the importance of looking at all the facets in mak-
ing any interpretation. Certainly Sally can manipulate, but her efforts may
not have the same brazen, self-indulgent qualities that are found with a so-
ciopathic-type individual.
The Interpersonal Style Matrix (Table 4-1) provides a more accurate
portrait of Sally. Her high Extraversion-low Agreeableness placement is
quite accurate. She is dominant with a degree of self-assurance; Sally is cer-
tainly assertive and firm. She gravitates toward leadership positions and is
quite successful in managing groups of people.
What I find of interest in Sally's profile is that it contains elements that
some may consider socially undesirable. For example, low Agreeableness
is usually not seen as a positive value. However, one needs always to keep
in mind that there are adaptive and maladaptive aspects to each of these
five domains. Being low in Agreeableness may not be an asset in someone
wanting to be in a religious ministry, but for someone in the corporate en-
vironment, this can be adaptive. In Sally's case, her lower levels of Agree-
ableness are probably instrumental in creating the level of success she
experiences in managing others. Leadership is inherently a manipulative
endeavor; one needs to persuade, threaten, encourage, force, and appease.
Low Agreeable individuals can assume not only a supportive role, but can
also be confrontive when the situation requires. They are also able to look
beyond the immediate needs of the individual toward larger goals.
Another interesting aspect of Sally's profile is her lower levels of Con-
scientiousness. Levels of Conscientiousness have been shown to be a strong
predictor of work success (Barrick & Mount, 1991; Piedmont & Weinstein,
1994). Despite such a low score, Sally has been quite successful in her career.
Again, there are many factors that predict job success, depending on the spe-
cific qualities of the position itself. Leadership ability is important for man-
agement, and Sally certainly has this quality. What can be anticipated from
Sally's profile, though, is that she most likely has reached the highest level of
management she is capable of attaining. Moving into upper management
may require the kinds of focus, dedication, and time commitment that Sally
may not be comfortable in making. Her aspiration levels do not take her in
such a competitive direction. Nonetheless, even her relatively modest
amounts of Conscientiousness can provide her with a degree of success.

CONCLUSIONS

The eight case histories described should provide the reader with
some exposure to the kinds of issues and concerns involved in making
NEO PI-R interpretations. There are, no doubt, many nuances involved in
making any interpretation, and only practice and experience can provide
156 CHAPTER 4

that level of expertise. However, there are a few issues that one needs to
keep in mind about making interpretations.
First, never assume pathology in the profile, especially when dealing
with clinical clients. The NED PI-R is a measure of normal personality; it
highlights personological issues that are characteristic of everyday levels
of functioning. Extreme scores are not necessarily indicators of pathology.
The only way to determine whether pathology is present is to employ an
instrument designed to measure such. It is important to develop an inter-
pretive skill that enables you to appreciate what various levels of a con-
struct represent personologically. How does a low Openness individual
appear and what type of environment do they gravitate toward? Being
able to appreciate how these constructs are expressed in everyday living
will provide you with a better interpretive sense of the instrument.
Second, remember that there are adaptive and maladaptive aspects of
each pole of the five factors. Do not immediately assume that it is inher-
ently better to be Agreeable or Conscientious. Personality is an adaptive
structure that individuals develop to help them manage their environ-
ments. So the question is never whether a personality is "good" or "bad";
rather the issue is whether someone with a certain disposition would func-
tion well in a new environment. Always appreciate the context within
which the respondent functions. Then, personality assessment can be use-
ful for highlighting potential issues the person may experience and iden-
tifying resources that can be drawn on.
Finally, always strive to incorporate multiple sources of information
in your assessment battery, and look for convergence over these diverse in-
struments. Never rely on a single scale or test to provide definitive an-
swers. Rather, psychological tests are designed to generate hypotheses
(about a client) that are in need of further experimentation. Evidence can
be gleaned from other instruments, clinical interviews, and observer re-
ports. Finding convergence for a hypothesis over diverse information
sources provides a high degree of confidence in its accuracy. There is no
substitution for good information and no test is perfect-error is every-
where!
 PROFILE ANALYSIS 157

APPENDIX NEO PI-R Computerized Report for Erica J.

- REVISED NEO PERSONALITY INVENTORY-

Interpretive Report
Developed By
Paul T. Costa, Jr., Ph.D.
Robert R. McCrae, Ph.D.
and PAR Staff

- CLIENT INFORMATION-
Results For Erica J.
Age 38
Sex Female
Test Form S
Test Date 07/31/96
Prepared For Ralph L. Piedmont, Ph.D.

The following report is based on research using normal adult samples and is intended
to provide information on the basic dimensions of personality. The interpretive information
contained in this report should be viewed as only one source of hypotheses about the indi-
vidual being evaluated. No decisions should be based solely on the information contained in
this report. This material should be integrated with all other sources of information in reach-
ing professional decisions about this individual. This report is confidential and intended for
use by qualified professionals only; it should not be released to the individual being evalu-
ated. "Your NEO PI-R Summary" provides a report in lay terms that may be appropriate for
feedback to the client.
 158 CHAPTER 4

Page 2

Factors NEO PI·~R Profile

--------- N N N N NNE E E E E E 0 0 0 0 0 0 A A A A A Ace C C C C
N E 0 A C 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6
>74----------------------------------------------------------------------->74
V
E
~
I

I
I I.
I

70- * -70
H I I

I I
~---+---*----------------------*--------------------------------------------+--
H :
I II
I I I
G60- * * -60
H I * * I
I I
I * * **
---+-*---------------------------------------------------------------------+--
* I
I
i
I
A I
~
RSO-
I * *
* * I
-50

iI * * * * • *. I
---+-----------------------------------------------------------------------+--
I I
L
040-
I I
-40
W * I
I
I * I
~--i---------:----------------------------------------:---------------:----i--
30- -30

~ II *1
W
<26-------------------------- _________________________ --------------------<26
:
N E 0 A C N N N N NNE E E E E E 000 0 0 0 A A A A A Ace C C C C
1 2 3 4 5 6 1 2 3 4 5 6 123 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6
 PROFILE ANALYSIS 159

Page 3

- NEO PI-R DATA TABLE -

Raw
Scale Score TScore Range
Factors
(N) Neuroticism 55 AVERAGE
(E) Extraversion 65 HIGH
(0) Openness 50 AVERAGE
(A) Agreeableness 47 AVERAGE
(C) Conscientiousness 32 VERY LOW
Neuroticism Facets
(N1) Anxiety 15 49 AVERAGE
(N2) Angry Hostility 16 57 HIGH
(N3) Depression 22 66 VERY HIGH
(N4) Self-Consciousness 13 46 AVERAGE
(N5) Impulsiveness 21 60 HIGH
(N6) Vulnerability 19 70 VERY HIGH
Extraversion Facets
(E1) Warmth 27 59 HIGH
(E2) Gregariousness 24 65 HIGH
(E3) Assertiveness 14 47 AVERAGE
(E4) Activity 21 57 HIGH
(E5) Excitement-Seeking 21 60 HIGH
(E6) Positive Emotions 25 59 HIGH
Openness Facets
(01) Fantasy 19 56 HIGH
(02) Aesthetics 17 47 AVERAGE
(03) Feelings 21 50 AVERAGE
(04) Actions 19 56 HIGH
(05) Ideas 21 56 HIGH
(06) Values 23 57 HIGH
Agreeableness Facets
(AI) Trust 22 51 AVERAGE
(A2) Straightforwardness 15 33 VERY LOW
(A3) Altruism 25 52 AVERAGE
(A4) Compliance 17 44 LOW
(AS) Modesty 22 56 HIGH
(A6) Tender-Mindedness 20 47 AVERAGE
Conscientiousness Facets
(Cl) Competence 17 36 LOW
(C2) Order 18 47 AVERAGE
(C3) Dutifulness 22 47 AVERAGE
(C4) Achievement Striving 12 31 VERY LOW
(C5) Self-Discipline 17 39 LOW
(C6) Deliberation 8 28 VERY LOW
 160 CHAPTER 4

Page 4

TM
Client Name : Erica J.NED PI-R
Test Date : 07/31/96 INTERPRETIVE REPDRT

--Validity Indices--
Validity indices (i.e., Aand C questions, total number of items missing, and response set)
are within normal limits.

--Basis of Interpretation--
This report compares the respondent to other adult women. It is based on self-reports of
the respondent.
At the broadest level, personality can be described in terms of five basic dimensions or
factors. NED PI-R domain scores provide good estimates of these five factors by summing
the six facets in each domain. Domain scores can be calculated easily by hand and are there-
fore used on the (hand-scored) Profile Form. More precise estimates of standing on the five
factors, however, are provided by factor scores, which are a weighted combination of scores
on all 30 facets (see Table 2 in the NED PI-R Professional Manual). Factor scores are best cal-
culated by computer.
Because factor scores have somewhat higher convergent and discriminant validity, they
are used as the basis of this report. In general, domain T scores and factor T scores are very
similar; occasionally, however, they differ. In these cases, the factor T score, which incorpo-
rates information from all 30 facets, is usually a more accurate description of the individual.
Factor scores are used to describe the individual at a global level, based on a composite
of facet scale scores. To the extent that there is wide scatter among facet scores within a do-
main, interpretation of that domain and factor becomes more complex. Interpretive statements
at the factor level may occasionally conflict with interpretive statements at the facet level. In
these cases, particular attention should be focused on the facet scales and their interpretations.

--Global Description of Personality: The Five Factors--

The most distinctive feature of this individual's personality is her standing on the factor
of Conscientiousness. Women who score in this range have little need for achievement,
putting personal interests or pleasure before business. They prefer not to make schedules, are
often late for meetings and appointments, and have difficulty in finishing tasks. Their work
is typically accomplished in a haphazard and disorganized fashion. They lack self-discipline,
prefer play to work, and may seem aimless in setting goals for their lives. They have a relaxed
attitude toward duties and obligations, and typically prefer not to make commitments. Raters
describe such people as careless, neglectful, unreliable, and negligent.
This person is high in Extraversion. Such people enjoy the company of others and the
stimulation of social interaction. They like parties and may be group leaders. They have a
fairly high level of energy and tend to be cheerful and optimistic. Those who know such peo-
ple would describe them as active and sociable.
Next, consider the individual's level of Neuroticism. Individuals scoring in this range are
average in terms of their emotional stability. They experience a normal amount of psychological
distress and have a typical balance of satisfactions and dissatisfactions with life. They are neither
high nor low in self-esteem. Their ability to deal with stress is as good as the average person's.
 PROFILE ANALYSIS 161

PageS

TM
Client Name : Erica J.NEO PI-R
Test Date : 07/31/96 INTERPRETIVE REPORT

This person is average in Agreeableness. People who score in this range are about as
good-natured as the average person. They can be sympathetic, but can also be firm. They are
trusting but not gullible, and ready to compete as well as to cooperate with others.
Finally, the individual scores in the average range in Openness. Average scorers like her
value both the new and the familiar, and have an average degree of sensitivity to inner feel-
ings. They are willing to consider new ideas on occasion, but they do not seek out novelty for
its own sake.

--Detailed Interpretation: Facets of N, E, 0, A, and C--

Each of the five factors encompasses a number of more specific traits, or facets. The NEO
PI-R measures six facets in each of the five factors. An examination of the facet scores pro-
vides a more detailed picture of the distinctive way that these factors are seen in this person.

Neuroticism
This individual is occasionally nervous or apprehensive, but no more so than the aver-
age individual. She often feels frustrated, irritable, and angry at others and she is prone to
feeling sad, lonely, and dejected. Embarrassment or shyness when dealing with people, es-
pecially strangers, is only occasionally a problem for her. She reports being poor at control-
ling her impulses and desires and she is unable to handle stress well.

Extraversion
This person is very warm and affectionate toward others and she usually enjoys large
and noisy crowds or parties. She is as assertive as most women when the circumstances re-
quire. The individual has a high level of energy and likes to keep active and busy. Excitement,
stimulation, and thrills have great appeal to her and she frequently experiences strong feel-
ings of happiness and joy.

Openness
In experiential style, this individual is somewhat open. She has a vivid imagination and
an active fantasy life. She is like most people in her appreciation of beauty in music, art, po-
etry, and nature, and her feelings and emotional reactions are normal in variety and intensity.
She enjoys new and different activities and has a high need for variety in her life. She is in-
terested in intellectual challenges and in unusual ideas and perspectives and she is generally
liberal in her social, political, and moral beliefs.

Agreeableness
This person has moderate trust in others, but is not gullible, recognizing that people can
sometimes be deceptive. She is willing at times to flatter or trick people into doing what she
wants, but she is reasonably considerate of others and responsive to requests for help. This
individual can be very competitive and is ready to fight for her views if necessary. She is
humble, unassuming, and uncomfortable talking about her achievements. Compared to other
 162 CHAPTER 4

Page 6

TM
Client Name: Erica J.NEO PI-R
Test Date : 07/31/96 INTERPRETIVE REPORT

people, she is average in her concern for those in need, and her social and political attitudes
balance compassion with realism.

Conscientiousness
This individual is sometimes inefficient or unprepared, and has not developed her skills
and talents fully. She is moderately neat, punctual, and well-organized, and she is reasonably
dependable and reliable in meeting her obligations. She has limited aspirations and might be
considered somewhat lackadaisical or lazy. She sometimes finds it difficult to make herself do
what she should, and tends to quit when tasks become too difficult. She is occasionally hasty
or impetuous and sometimes acts without considering all the consequences.

--Personality Correlates: Some Possible Implications--

Research has shown that the scales of the NEO PI-R are related to a wide variety of psy-
chosocial variables. These" correlates suggest possible implications of the personality profile,
because individuals who score high on a trait are also likely to score high on measures of the
trait's correlates.
The following information is intended to give a sense of how this individual might func-
tion in a number of areas. It is not, however, a substitute for direct measurement. If, for ex-
ample, there is a primary interest in medical complaints, an inventory of medical complaints
should be administered in addition to the NEO PI-R.

Coping and Defenses

In coping with the stresses of everyday life, this individual is not very likely to react with
ineffective responses, such as hostile reactions toward others, self-blame, or escapist fantasies.
She is likely to use both faith and humor in responding to threats, losses, and challenges. In
addition, she is somewhat more likely to use positive thinking and direct action in dealing
with problems.

Somatic Complaints
This person likely responds in a normal fashion to physical problems and illness. She is
prone neither to exaggerate nor to minimize physical symptoms and is fairly objective in as-
sessing the seriousness of any medical problems that she might have.

Psychological Well-Being
Although her mood and satisfaction with various aspects of her life will vary with the
circumstances, in the long run this individual is likely to experience the normal course of pos-
itive and negative feelings and generally be happy.

Cognitive Processes
This individual is likely to be about average in the complexity and differentiation of her
thoughts, values, and moral judgments as compared to others of her level of intelligence and
 PROFILE ANALYSIS 163

Page 7

TM
Client Name : Erica J.NEO PI-R
Test Date : 07/31/96 INTERPRETIVE REPORT

education. She would also probably score in the average range on measures of ego develop-
ment.

Inperpersonal Characteristics
Many theories propose a circular arrangement of interpersonal traits around the axes of
Love and Status. Within such systems, this person would likely be described as dominant, as-
sured, warm, loving, and especially gregarious and sociable. Her traits are associated with
high standing on the interpersonal dimensions of Love and Status.

Needs and Motives

Research in personality has identified a widely used list of psychological needs. Indi-
viduals differ in the degree to which these needs characterize their motivational structure.
The respondent is likely to show high levels of the following needs: abasement, affiliation, ag-
gression, change, dominance, nurturance, play, succorance (support and sympathy), and un-
derstanding (intellectual stimulation). The respondent is likely to show low levels of the
following needs: achievement, cognitive structure, and harm avoidance (avoiding danger).

--Clinical Hypotheses: Axis II Disorders and Treatment Implications-

The NEO PI-R is a measure of personality traits, not psychopathology symptoms, but it
is useful in clinical practice because personality profiles can suggest hypotheses about the
disorders to which patients are prone and their responses to various kinds of therapy. This
section of the NEO PI-R Interpretive Report is intended for use in clinical populations only.
The hypotheses it offers should be accepted only when they are supported by other corrobo-
rating evidence.
Psychiatric diagnoses occur in men and women with different frequencies, and diag-
noses are given according to uniform criteria. For that reason, information in this section of
the Interpretive Report is based on Combined Sex norms.
Since Same Sex Norms were used for the Interpretive Report, there may be some appar-
ent inconsistencies in score levels and interpretations.

Axis II Disorders
Personality traits are most directly relevant to the assessment of personality disorders
coded on Axis II of the DSM-IV. A patient may have a personality disorder in addition to an
Axis I disorder, and may meet criteria for more than one personality disorder. Certain diag-
noses are more common among individuals with particular personality profiles; this section
calls attention to diagnoses that are likely (or unlikely) to apply.
Borderline Personality Disorder. The most common personality disorder in clinical
practice is Borderline, and the mean NEO PI profile of a group of patients diagnosed as
having Borderline Personality Disorder provides a basis for evaluating the patient. Profile
agreement between the patient and this mean profile neither suggests nor rules out a diag-
nosis of Borderline Personality Disorder; it is comparable to agreement seen in normal in-
dividuals.
164 CHAPTER 4

PageS

TM
Client Name : Erica J.NEO PI-R
Test Date : 07/31/96 INTERPRETIVE REPORT

Other Personality Disorders. Personality disorders can be conceptually characterized by

a prototypic profile of NEO PI-R facets that are consistent with the definition of the disorder
and its associated features. The coefficient of profile agreement can be used to assess the over-
all similarity of the patient's personality to each of the nine other DSM-IV personality disor-
der prototypes.
The patient's scores on N2: Angry Hostility, N4: Self-Consciousness, N6: Vulnerability, E1:
Warmth, E2: Gregariousness, E4: Activity, ES: Excitement-Seeking, E6: Positive Emotions, 01:
Fantasy, 03: Feelings, 04: Actions, 05: Ideas, AI: Trust, A2: Straightforwardness, A3: Altruism,
C1: Competence, and C5: Self-Discipline suggest the possibility of a Histrionic Personality Dis-
order. Histrionic Personality Disorder is relatively common in clinical practice; the patient's co-
efficient of profile agreement is higher than 90% of subjects in the normative sample.
The patient's scores on N1: Anxiety, N3: Depression, N4: Self-Consciousness, N6: Vul-
nerability, E3: Assertiveness, A3: Altruism, A4: Compliance, AS: Modesty, and C4: Achieve-
ment Striving suggest the possibility of a Dependent Personality Disorder. Dependent
Personality Disorder is relatively common in clinical practice; the patient's coefficient of pro-
file agreement is higher than 90% of subjects in the normative sample.
It is unlikely that the patient has Paranoid Personality Disorder, Schizoid Personality Dis-
order, Schizotypal Personality Disorder, Avoidant Personality Disorder, or Obsessive-Com-
pulsive Personality Disorder because the patient's coefficients of profile agreement are lower
than 50% of the subjects in the normative sample.

Treatment Implications
Like most individuals in psychotherapy, this patient is high in Neuroticism. She is likely
to experience a variety of negative emotions and to be distressed by many problems, and
mood regulation may be an important treatment focus. Very high Neuroticism scores are as-
sociated with a poor prognosis and treatment goals should be appropriately modest.
Because she is extraverted, this patient finds it easy to talk about her problems, and en-
joys interacting with others. She is likely to respond well to forms of psychotherapy that em-
phasize verbal and social interactions, such as psychoanalysis and group therapy.
Because the patient is low in Conscientiousness, she may lack the determination to work
on the task of psychotherapy. She may be late for appointments and may have excuses for not
having completed homework assignments. Some evidence suggests that individuals low in
Conscientiousness have poorer treatment outcomes, and the therapist may need to make
extra efforts to motivate the patient and structure the process of psychotherapy.

--Stability of Profile--
Research suggests that the individual's personality profile is likely to be stable through-
out adulthood. Barring catastrophic stress, major illness, or therapeutic intervention, this de-
scription will probably serve as a fair guide even in old age.

END OF REPORT
 CHAPTERS

APPLICATIONS OF THE RATER

VERSION OF THE NEO PI-R

Self-reports are perhaps the most frequently employed method of assess-

ment. Doubtless this is because of their ease of administration, scoring,
and interpretation. However, as was noted in Chapter 3, there are issues
surrounding the use of self-report measures in applied settings. The issues
of concern center on whether individuals are able to be straightforward in
responding to the items, or, whether they distort their responses, either de-
liberately or unconsciously. Such concerns have motivated many to de-
velop "validity scales," indices designed to detect the presence of both
random (e.g., capricious response to items) and systematic (e.g., acquies-
cence) error. As noted in Chapter 3, much more energy needs to be in-
vested in establishing the ability of these validity scales to detect such
errors. Piedmont and McCrae (1998) noted that validity scales seem to pos-
sess little validity themselves. They argued that test users need to be sen-
sitive to the dynamics of any assessment situation and to present the
testing materials in such a way as to enhance candidness and cooperation
from the test respondent. But even in the best of cases, there is always
some possibility that distortion may be present. In such circumstances,
Piedmont and McCrae argued that one should include other, nonself-
report measures as part of the testing battery. Of particular relevance are
observer ratings.
Observer ratings provide a counterpoint to self-reports (e.g., McCrae,
1994a). Individuals providing ratings rarely have the same motivations to
distort their responses as the target him- or herself. Raters are not neces-
sarily committed to making the target appear "good" or socially desirable.
Further, raters can provide another perspective on the individual, such as

165
166 CHAPTERS

the type of social impressions he or she generates. Thus, observer ratings

provide fresh insights into a person that may not be obtainable from a self-
report and are an alternative source of information when a self-report may
be of questionable validity.
As noted in Chapter 2, the NEO PI-R is unusual among personality
measures in that it has its own established observer rating form (Form R).
Statements in this version are identical to Form S (the self-report) except
that they are phrased in the third person; one format is for individuals rat-
ing women, the other for rating men. Separate norms have been developed
for Form R, allowing one to directly compare a rating with a self-report.
Research has shown that among volunteer, nonclinical groups there is sub-
stantial cross-observer convergence: Scores on the self-reports agree very
well with observer ratings (see Table 2-5).
This chapter focuses on the utility of observer ratings, how they can
both complement and extend self-report data. Of particular interest is the
use of self- and observer-ratings with married couples. Systematic com-
parisons of self-reports with spouse ratings may provide insights into the
kinds of issues and problems a couple may be facing. Such an analysis out-
lines the underlying sources of personal motivation that may be fueling
conflict in the relationship. It also provides an opportunity to explore the
perceptions each member of the relationship has of the other and the re-
sulting expectations such perceptions may entail.

USING OBSERVER RATINGS

The NEO PI-R computer scoring program provides a very useful plat-
form for evaluating observer ratings and offers a number of advantages.
First, not only does the program carry the normative data for a rating, it
also allows one to directly compare a self-report with a corresponding rat-
ing within the same report. The program lists, side by side, results from the
two assessments. The program also highlights the location of significant
differences between any given pair of ratings. Second, it is possible to si-
multaneously plot the results of such tests on a graph. One can, therefore,
see areas of convergence and divergence between the two profiles. Finally,
the computer-generated report provides an empirical index of profile sim-
ilarity, allowing one to determine how congruent the two ratings are.
These values have also been normed, permitting one to assess the likeli-
hood of finding a particular level of agreement.
As noted earlier, ratings provide an independent source of informa-
tion about an individual that is not contaminated by response distortions
characteristic of a self-report. Therefore, when scores from a self-report
 RATER VERSION APPLICATIONS 167

converge with observer ratings, one can have confidence that the results
are not the product of any type of response distortion. Such agreement can
have important therapeutic consequences. For example, a college student
presents herself to the counseling center because she is experiencing er-
ratic school performance. The student rates herself high on Conscientious-
ness. Personality ratings provided by the parents also rate her high on this
domain. This agreement would alert the counselor that the problematic
academic performance represents a recent change rather than a long-
standing habit and would influence the direction of further assessment.
Areas of disagreement can also be interpretively valuable, depending
on the context. For example, a client in treatment may see herself as being
very low on Neuroticism, while her family members may see her as being
quite high. Such a disagreement can alert the therapist to a number of pos-
sibilities. The client may be unaware of the level of her emotional distress
or there could be significant family dysfunction. The client could be delib-
erately attempting to hide her issues from the therapist, or she may be in
denial. In either case, the therapist has important information that can be
employed in the therapeutic context.
Observer ratings can also be used to help validate a self-report that
may be questionable for any number of reasons. In Chapter 3 I brought up
the issue of validity scales in assessment. Validity scales are indices devel-
oped by test makers to determine whether some response pattern evi-
dences any type of distortion or bias. Such biases include, but are not
limited to, denial of problems, random responding, trying to create a pos-
itive image of self, trying to create a negative image of self, and acquies-
cence. Such biases tell us nothing about the individual and therefore a
profile contaminated by such distortions cannot be trusted. Numerous
scales have been developed to detect the presence of such errors.
All too often, however, high scores on these validity scales are used to
reject a test protocol. On the one hand, a test interpreter does not want to
make an erroneous conclusion about a respondent, especially when the
stakes are quite high (as they may be in a forensic or job selection environ-
ment). Yet, to reject the protocol out of hand because of a high score on one
or more validity indices may waste resources. Time, energy, and money
were invested in having the assessment completed and scored; there is a
real financial cost to discarding a protocol. Further, asking the individual
about who they are is the most direct and intuitively appealing approach
to doing assessment. To reject what a person has told us undermines many
aspects of the assessment process. For example, do we rely on a structured
interview? If the person distorted information on a assessment question-
naire, what confidence do we have that they will not distort responses in
the interview? After all, interviews are less reliable than self-report
168 CHAPTERS

measures. Do we tum to letters of recommendation? Well, the same issue

applies here. If the respondent is not forthcoming with us, either directly
in an interview or indirectly through the questionnaire, then what assur-
ances do we have that they are not being manipulative with certain others
whom they elected to write letters of support?
Observer ratings can be helpful in this type of situation. Of course, rat-
ings are not perfect indices; they, too, have their own sources of error and
bias. And, as previously noted, the individual selected to provide the rat-
ing may have an impact on the result. However, the value of ratings is
threefold. First, a rating enables an individual to do a more nuanced and
precise assessment of a target in a standardized format. Unlike a re-
commendation where a rater can provide broad generalizations within a
self-selected range of convenience, a questionnaire rating outlines the per-
sonological issues of interest. Second, the level of assessment requires a de-
gree of precision and depth that is not necessarily required or found in a
recommendation. Finally, the results of a personality rating are evaluated
normatively, making possible a broader interpretive perspective not avail-
able with either an interview or recommendation.
Thus, observer ratings provide a robust alternative to self-reports and
have advantages that may not be paralleled by other types of assessment.
Further, the more raters one has providing information, the more confi-
dence one can have in conclusions based on areas of convergence. An il-
lustrative example may help here. Piedmont and McCrae (1998) evaluated
the efficacy of a number of validity scales and were interested in deter-
mining whether scores on such indices really were able to identify invalid
protocols. The criterion for these analyses were observer ratings of the per-
son. If a validity scale was truly "valid," then self-report protocols that had
high scores on these indices (indicating the presence of some type of dis-
tortion) would not show good convergence with observer ratings (which
were based on the combined scores of two independent raters). If, how-
ever, good convergence was found, that would question the efficacy of the
validity scales. The results of our study provided strong evidence that va-
lidity scales (we evaluated 13 different indices) do not do a good job in
identifying invalid protocols. The profile in Figure 5-1 shows why.
The person represented here (Barbara Nettle, a fictitious name) is one
of the subjects from this study. In looking at each subject's scores on the 13
validity indices, this 20-year-old college student was found to have a high
score (indicating an invalid protocol) on 4 of them. That means, according
to 4 different measures of test validity, this profile should be considered in-
valid. No one in our sample had high scores on more than 4 of these indices.
In Figure 5-1 her self-reported profile (the solid line) is presented along with
a composite rater evaluation based on two observers (the dotted line). This
figure is the graph generated by the computer software scoring program.
RATER VERSION APPLICATIONS 169

NED PI-R Profile

t-t Barbarl Nettle •...• BJf

~ 80-
..

.
I

,
~

%
- "
,, ,,
.,
I. 70-
! 1\ ~ ~ H' ~ ) , ,
:

II •
I

.- /\
• V.
~, 1'1• j'H
W.~ /•1 '\.
I~

,.or
' 60- • I ., ,, "
~. I
•
• .,. • ..~J:',··1 I
, ' "

... • I
" :
,I.
0 ,, ,, Y-" • ',I ,."" "

r
~~ 50- , ,' ,
- ,
I ..• :.:
.,
•
3~ ""
• . .....
I,
, "
I I I

...
,
~ 40-
- •
3

,
~ 30-

~ 20-
-
II
UlOID 04IUl'fn'4 04 IUl 'fn '4 04IU'fn'4 04I'U'fn'4 04 IUl 'fn '4
Factors Z Z Z Z Z Z III III III III III III 0 0 0 0 0 0 C C C C a: a: Cl Cl DDDCl
FIGURE 5-1. NEO PI-R self-reported and composite observer rating for Barbara Nettle.

A visual inspection shows that the two profiles agree quite well. Rat-
ings on the Extraversion, Openness, and Agreeableness domains are all
similar. On the Neuroticism domain both she and the composite rating see
her as high, although there is a difference in magnitude. A noticeable dif-
ference is observed on the Conscientiousness domain, where she rates her-
self in the Very High range while the rating is in the high Average range.
Consistent with this visual similarity, the profile agreement index (which
is generated by the scoring software) between the two sets of ratings is .81,
suggesting an overall agreement that is very high in comparison with the
 170 CHAPTERS

level of agreement seen among research volunteer couples. A statistical

evaluation of the differences between the self and rating scores indicates a
significant difference on only two of the 35 scales: Impulsiveness (N5) and
Tender-Mindedness (A6). All other differences are nonsignificant.
Interpreting this profile provides much valuable information. Clearly
Barbara sees herself as being more emotionally distressed than do the in-
dividuals who provided the composite rating, particularly when it comes
to levels of Hostility (N2) and Impulsiveness (N5). Given her ability to
cope with distress (average score on Vulnerability, N6) her internal distress
may not spill out into her relationships with others; this may explain why
the ratings are lower on these two domains. Also, given her low score on
Straightforwardness (A2), Barbara may do well in concealing her negative
feelings from onlookers. Interestingly, the raters see Barbara as having
much more Trust (AI) and being more Tender-Minded (A6) than she sees
herself. Clearly, others see endearing capacities in Barbara that she may
not be able to recognize in herself, particularly given her high self-rated
score on Self-Consciousness (N4). She may feel inadequate and unworthy
as a person; therefore, she may undervalue her own abilities to reach out
and touch the lives of others. The high rating on Altruism (A3) shows that
Barbara is perceived as being much more caring and helpful than she rec-
ognizes. Others see her as greatly extending of herself; yet as reflected in
Barbara's high Self-Consciousness score, she, again, will tend to under-
value her gifts.
Barbara's slightly higher self-ratings on the Conscientiousness facets
may reflect her own efforts at compensating for these perceived inadequa-
cies. By having a strong aspiration to control and master her environment,
Barbara may believe that she can win for herself the personal value she
does not feel she currently has. Her raters do not see these aspirational
strivings as strongly, suggesting that Barbara may restrict her ambitions
for very select environments (e.g., doing well in school). These discrepan-
cies in scores may also reflect some of Barbara's unique interpretations of
situations; interpretations of which others may be unaware.
This profile is interesting because of the high degree of convergence
that is found although the protocol would have been considered invalid.
Relying exclusively on those validity indices would have led a test inter-
preter to reject this profile, even though it does indeed have much validity
to it. Certainly, the most immediate lesson here is never to rely exclusively
on validity indices to determine profile adequacy. Such measures should
never preempt good assessment practice, which should endeavor to in-
clude multiple sources of information (such as observer ratings). Throwing
away this protocol would have been wasteful: wasteful of Barbara's time
and effort and wasteful of the test materials that would have been unnec-
 RATER VERSION APPLICATIONS 171

essarily consumed. Also wasted would be the opportunity to help Barbara.

The observer rating shows that this self-report is indeed quite valid and
has some important insights into Barbara.
It is also interesting that we get additional interpretive mileage ob-
tained by including an observer report. It is comforting to see such strong
convergence between the two sets of ratings. It reinforces our belief in the
accuracy of our interpretations. However, within this context of agree-
ment, the areas of disagreement also become of interest. As we saw with
Barbara's profiles, there may be issues occurring within her that are not
available to those who are close to her. This may be due to her reluctance
to share some of her own inadequacies, because she herself is quite sensi-
tive to this issue. Therefore, she may not always be straightfonvard in ex-
pressing her feelings. In a therapeutic context one would want to carefully
explore her inner world. Her feelings of inadequacy may lead her to de-
value the gifts she has to offer others, but she may compensate for these
feelings by trying to achieve high goals for herself.
The value of observer ratings is that they enable us to look at a given
person from two different angles. The focus of these views is slightly dif-
ferent, providing greater interpretive clarity where the focus overlaps and
generating additional hypotheses in reconciling the differences. However,
disagreements between a self-report and a rating do not always cast doubt
on the accuracy of the self-report. Such disagreements may speak more to
the distorted perceptions of the rater than to any inaccuracies in the self-
report. The next section amplifies the issues that arise when using such
multimodal assessments within the marital context. Observer ratings can
be used to great advantage when dealing with couples and attempting to
evaluate and manage their issues and conflicts. Here, discrepancies be-
tween observer ratings and self-reports can be used to highlight the possi-
ble motivational bases of ongoing conflict in the relationship.

OBSERVER RATINGS WITH MARRIED COUPLES

Understanding couples seems to be a most appropriate application of
the NEO PI-R. The availability of a standardized and normed observer
version is well suited to the task of understanding the quality of a rela-
tionship as well as exploring those personological qualities that may be
operating to disrupt or undermine the level of satisfaction that each mem-
ber may be experiencing.
Researchers have long been interested in understanding committed
relationships (see Karney & Bradbury, 1995, for a review) and have found
that personal dispositions significantly impact the relationship in many
 172 CHAPTERS

ways. Levels of neuroticism have consistently been shown to correlate

negatively with the amount of marital satisfaction experienced by the cou-
ple (Kosek, 1996b; Lester, Haig, & Monello, 1989; Russell & Wells, 1994). In-
dividuals with high levels of neuroticism experienced less satisfaction
with the relationship, although neuroticism was not linked with decreas-
ing satisfaction over time (Karney & Bradbury, 1997).
Personal temperaments also have been linked to the kinds of prob-
lems individuals confront in their relationships. For example, Buss (l991a)
showed that submissive men (low Extraversion) complained about their
wives being condescending; low Conscientious males complained that
their wives were moody and self-centered; men low on Openness com-
plained that their wives were possessive, abusive, unfaithful, and physi-
cally self-absorbed. The pattern of personality correlates with complaints
was very different for women. Wives low on Agreeableness complained
that their husbands were possessive, neglecting, abusive, unfaithful, moody,
sexually withholding, and inconsiderate.
In a separate study, Buss (1992) showed that the five major dimen-
sions of personality were differentially linked to the kinds of manipulation
tactics individuals employed in their relationships. Interestingly, all five
personality domains were linked to a variety of tactics, and the patterns
were similar for both genders. For example, those scoring high on Agree-
ableness tended to use Pleasure Induction as a tactic of influence; those
low on Extraversion tended to use self-debasement tactics (e.g., allowing
themselves to be debased so that their spouses would do what they
wanted); those high on Openness and Conscientiousness would use rea-
son as a major tactic; those high on Neuroticism would use coercion and
monetary reward to get their way.
These two studies show that one's personality has a profound impact
on both the quality and tempo of one's relationships with intimate others.
Our personal dispositions bring with them tendencies to enact certain pre-
dictable behavior patterns in order to fulfill our basic needs. These actions,
which are so much a part of ourselves, may be found objectionable by
spouses. Buss (1991a) showed that both husbands and wives who were
low in Agreeableness performed actions that created upset in their
spouses, such as being condescending, unfaithful, moody, self-centered,
and abusive. Husbands and wives high on Neuroticism provoked upset in
their spouses through actions that were deemed possessive, dependent,
and jealous in nature.
It is obvious that what an individual brings temperamentally to a re-
lationship has a significant impact on levels of intimacy and satisfaction.
There is no doubt that someone high on Neuroticism will experience a
lower level of marital satisfaction. Such individuals find little succor in the
 RATER VERSION APPLICATIONS 173

world; they are unhappy at work, unhappy in play, and unhappy in love.
Further, such individuals also behave in ways that may provoke negative
reactions from spouses. But relationships are not static arenas in which a
single individual attempts to find personal gratification. Rather, they are
dynamic. There are two individuals working (or competing) to find satis-
faction. Rather than merely imposing our will on the environment, the
presence of an "other" represents a salient stimulus in our environment
that must be attended to, responded to, and accommodated.
One of the limitations of the research mentioned previously is its
reliance on self-report data. Self-reported behaviors are linked to self-
reported personality characteristics or self-reported problems of one indi-
vidual are correlated with the self-reported personality of the spouse. This
type of research, although important for outlining how individuals act in
a relationship, overlooks more of the dynamic aspects of the dyad. Specif-
ically, I am referring to the kinds of internal representations individuals
carry of their spouses, and how these images create expectations about the
partner and the relationship itself. In order to capture this dimension, one
needs to assess not only individuals, but their perceptions of their spouses,
something that can be done only through an observer rating.
Observer ratings provide an opportunity to access an entirely new
level of the relationship through comparisons of a person's ratings of his or
her spouse with the spouse's self-report. Systematically comparing the de-
gree of congruence between these two sets of images may provide a win-
dow into the degree of marital satisfaction experienced by the couple and
the motivational sources contributing to their level of harmony or conflict.

SELF-OTHER CONGRUENCE IN COUPLES

Early research on marital happiness provided two theories of marital

satisfaction. One theory, complementarity (Winch, 1958), suggested that
individuals looked for mates who balanced their own personal needs. For
example, a dominant male would search for, and be happy with, a wife
who was submissive. Another theory, homogamy, asserts that individuals
select as mates those who resemble themselves in temperament (Barry,
1970; Tharp, 1963). Research showed that the correlation between person-
ality self-reports was higher among happily married couples, although the
pattern of these results has been inconsistent. Some believe that happy
couples become more similar over time (e.g, Gruber-Baldini, Shaie, &
Willis, 1995), while other research shows that any similarity is a function of
initial assortment (e.g., Keller, Thiessen, & Young, 1996). Overall, recent re-
search shows that generally there is a low, positive relationship between
the self-rated personalities of spouses (e.g., Buss, 1984, 1991a).
174 CHAPTERS

Other research has begun to evaluate the kinds of perceptions indi-

viduals hold of their spouses and the degree to which they correspond to
spouses' self-perceptions. The driving hypothesis here is that the amount
of congruence between a self-report and the spouse's rating would be re-
vealing of the level of satisfaction experienced by the rater. This approach
goes by different names, such as insight analysis (Megargee, 1972) or criss-
cross testing (Taylor & Morrison, 1984). Whatever its label, the approach
provides a new method of evaluating couples by focusing on the images
each holds of the other and the role these expectations play in the unfold-
ing process of marital adjustment.
What I find appealing about this process is that it addresses the heart
of the relationship, which is how each partner perceives the motivations
underlying the behavior of his or her spouse. When individuals get mar-
ried, they have an inner image of the person they believe their spouse to be.
These internal images generate certain expectations about their spouse's
behavior both inside and outside the relationship. As the couple spends
more time together, these expectations, or, better yet, hypotheses, are tested
out. Behaviors are observed and the correspondence of these behaviors to
the putative personality are evaluated. A good "fit" leads to more satisfac-
tion in the relationship. After all, the spouse is acting in a way consistent
with the images and expectations the person has. However, when these be-
haviors are not consistent with the assumed personality of the spouse, then
dissatisfaction begins to set in. The spouse is acting in ways that are unex-
pected and unanticipated, leading to a general sense of discomfort. The re-
ality of the spouse's personality does not match the images that were
originally developed. This leads to a general decline in satisfaction.
Another reason that a lack of congruence would be related to dissat-
isfaction is that the individual misperceives the motivations of the spouse.
For example, John may believe that his wife scores low on Neuroticism;
she is an emotionally stable person who is not prone to emotional outburst
and prolonged experiences of negative affect. If, however, his wife is in-
deed high on Neuroticism, then John will misperceive her many com-
plaints, her nagging, and her yelling at him. Rather than seeing these
behaviors as a sign of her need for succorance and reassurance, he will in-
terpret these behaviors as attacks on him. He may see her as very antago-
nistic, leading him to withdraw from her or to retaliate in kind. This, in
tum, only exacerbates her feelings of loneliness and abandonment. Thus,
a negative cycle of conflict and disharmony is created because John does
not accurately perceive the needs of his wife.
In either scenario, the nature and degree of cross-observer congruence
between a rating and self-report can be a useful barometer of the levels of
marital satisfaction, and can outline the motivational dynamics that may
 RATER VERSION APPLICATIONS 175

be precipitating and maintaining conflict. It is important to note here that

disagreement between the rater and the self-report is diagnostic of the lack
of satisfaction in the rater. If the rater has misperceived the spouse, then he
or she will be unable to accurately anticipate the spouse's behavior or will
be led to make incorrect attributions of the spouse's motivations. In either
case, the individual has a tenuous grasp of the interpersonal context of the
relationship and this leads to conflict and unhappiness.
There is not a large literature that evaluates cross-observer dynamics.
Creamer and Campbell (1988) outlined a model of interpersonal percep-
tion in dyadic adjustment and found that the level of cross-observer agree-
ment was positively correlated with levels of adjustment, although the
relationship was stronger for women than men. Further, they noted that
self-reported agreement in personality also predicted greater satisfaction
(Le., the homogamy hypothesis). Ptacek and Dodge (1995) showed that the
correspondence between self-reported coping ability and spouse ratings of
coping ability was positively related to relationship satisfaction. These au-
thors noted that the perceived similarity in coping abilities was a stronger
predictor than those based on self-reported "actual" similarities. As Ptacek
and Dodge noted:
This suggests the possibility that one's perceptions may be more important
to relationship satisfaction than are "actual" coping similarities. Consistent
with this possibility are the arguments of other authors that negative affect
in relationships stems not from incompatibility but from misunderstand-
ings and faulty interpretations of each other's behaviors. (p. 82).

Thus, evaluating cross-observer agreement provides an opportunity

to evaluate the expectations individuals hold of their spouses' actual be-
haviors, and the interpretations made of those behaviors. This is why the
analysis is taken from the vantage point of the observer. As we saw in
Chapter 2, there is good cross-observer agreement in NEO PI-R scores
among normal adults. Therefore, the lack of agreement between a rating
and a self-report (particularly between individuals in a marital relation-
ship), speaks more to the distorted perceptions of the perceiver than any
inaccuracies in the self-report. This approach, I believe, leads to a more dy-
namic assessment of the relationship. It addresses the needs and expecta-
tions of both members of the dyad and the behavior patterns these forces
may be following in order to find satisfaction.
The value of the NEO PI-R in this context is threefold. First, the NEO
PI-R provides a useful language for talking about and describing person-
ality. Second, it provides a medium for couples to express their own ex-
pectations about each other. Finally, it provides for clinicians' insights into
the motivational forces that may be creating conflict and dissatisfaction for
176 CHAPTERS

the couple. These patterns may suggest intervention strategies that would
benefit the couple. Taylor and Morrison (1984) provide a good description
of the value of cross-observer analysis:
The test profiles can be effectively used to shift the focus from the imme-
diate complaints to an examination of the influence of the two personal-
ities; as well as to develop an understanding of the interpersonal
dynamics involved. The test results can help the couple objectify their
problems and focus more on the role played by their individual person-
alities and behavior in the overall situation. (p. 17)
The next section provides an overview of what is involved in per-
forming cross-observer analyses of this type, as well as several case histo-
ries that outline the interpretive value of this approach.

CROSS-OBSERVER AGREEMENT ANALYSES

Overview
To conduct a cross-observer agreement (COA) analysis, one needs to
have a couple complete the NEO PI-R for both themselves and their part-
ner. This generates four separate profiles for each couple. The NEO PI-R
works well in this scenario because it provides self-report and observer
booklets as well as separate norms. Scores can also be readily translated
into T-scores and the resulting profiles can be jointly plotted on the same
profile sheet. In addition, the computer scoring program has the capacity
to combine a self-report ~nd an observer rating into a single report, and
will statistically compare the two profiles as well as determine whether
any facets are significantly different. The computer scoring program can
also generate a simultaneous plotting of both profiles (as was demon-
strated in Figure 5-1).
Given the four profiles, which ones should be compared? Any combi-
nation would be of interest and value. One can directly compare the two
self-reports. This would provide information about how well the two in-
dividuals complement each other temperamentally. For example, if one in-
dividual is high on Assertiveness (E3) and low on Compliance (A4) and
the other is just the opposite, then this couple may find themselves a good
"fit." However, such comparisons could also indicate potential areas of
conflict. For example, if both individuals score high on Neuroticism, then
one could expect a great deal of emotional controversy and contentious-
ness in the relationship. Finally, a comparison of the two self-report scores
could be the foundation of a discussion of the kinds of life directions each
person is likely to follow and how the life trajectories of the spouse may fit
with the expectations or needs of the self-reporter.
 RATER VERSION APPLICATIONS 177

A similar type of analysis could be performed on the two observer rat-

ings. These scores reflect the perceptions each person has of the other. To
some extent, these observer ratings reflect expectations each holds of the
other. For example, he may see her as being passive while she may see him
as being dominant. Although there may be a complemetarity to these rat-
ings, one needs to determine whether they reflect "real" dispositions on
the part of the ratee. Because he sees her as passive, the man believes that
he must act in a dominant way. But there are two important questions to be
evaluated here. First, is the woman really passive? Second, does the man
really prefer to act in a dominant manner? Thus, comparing the two rat-
ings allows for couples to discuss the images they hold of each other and
the latent expectations that may coincide with these perceptions. They can
then determine whether they are trying to live up to these expectations or
acting in a more genuine manner, consistent with their own personal dis-
positions.
Finally, one can compare an individual's ratings of his or her spouse
with the partner's self-report. For our purposes, this is the focus of our
work. As we noted earlier, such a eOA (cross-observer agreement) profile
speaks to the needs and issues of the person doing the rating. Areas of dis-
agreement should be diagnostically revealing of the kinds of issues and
problems the rater is experiencing with the partner. Differences are deter-
mined by subtracting the self-report T-score from the observer rating T-
score (Le., Rating minus Self-Report). Differences of 15 points or more
should be considered statistically significant. Significant differences on the
domains indicate a general class of problems that the rater is experiencing.
For example, a higher rating on Extraversion in comparison to a self-report
may suggest that the rater finds the spouse to be too dominant, overly in-
volved with friends and acquaintances, overcontrolling, and taking too
many risks. A lower rating on Extraversion may suggest that the rater
finds the spouse to be unaffectionate, passive, unassertive, and not com-
municating in the relationship. Significant differences on the facet scales
may suggest more specific issues and areas of contention.
Overall, the more discrepancies between the two profiles, the more
dissatisfaction the rater is experiencing. Given the interpretive pedigree
of the NEO PI-R, it would be relatively straightforward to derive hy-
potheses for interpreting cross-observer differences for each domain and
facet scale. Because of the paucity of research in this area, clinicians will
need to rely on their understandings of the personological content of
each NEO PI-R scale to use fully eOA analysis with couples. However, I
have conducted some basic research that may help to provide a frame-
work for using the NEO PI-R in this manner. The next section outlines
some of these findings.
178 CHAPTERS

Cross-Observer Differences and Marital Problems

In order to evaluate the validity of COAanalysis it was first necessary
to have a criterion measure that could empirically document the interpre-
tive value of the process. Such a measure would need to have a number of
specific, behaviorally oriented items that would be relevant to interper-
sonal conflict in a relationship. These particular behaviors would also have
to be theoretically relevant to the personality dimensions of the five-factor
model. In this way it could be demonstrated that discrepancies between
observer ratings of the spouse and his or her self-report results on each of
the five personality factors could be linked with specific interpersonal is-
sues. To address this need, the Couples Critical Incidents Check List was cre-
ated (CCICL; Piedmont & Piedmont, 1996).
The CCICL was developed with specific emphasis on content validity.
To this end a number of clinicians who work with couples in crisis were
asked to generate a list of specific behaviors that have been raised as issues
of conflict among their clients. Further, we also developed a list of specific
behaviors that we believed would reflect the behavioral tendencies of
someone very high (or very low) on each of the five factors. This list was
reviewed by another group of clinicians to determine the appropriateness
of each behavioral item. Then the items were sorted into one of six cate-
gories. The first five, Emotional, Interpersonal, Flexibility, Cooperativeness,
and Personal Reliability were designed to parallel the five major personality
dimensions, respectively. Items included reflected potential problems that
someone either high or low on that personality dimension would be likely
to have. The final category, Relationship Context, contained behaviors that
were seen as being particularly relevant to understanding conflict and dis-
satisfaction but could not be classified unambiguously into one of the
other categories (e.g., physically abusive, sexual difficulties, gambler,
mocks me in front of others). Table 5-1 provides examples of items for each
category.
As can be seen, individuals high on Neuroticism are more likely to be
perceived by their spouse as being moody and whiny, while those low on
Neuroticism would be seen as exhibiting too much emotional control and
never showing any weaknesses. Those high on Extraversion would be
perceived by their spouses as being too dominant and too bossy, while
those low on Extraversion would be seen as aloof and passive. These cate-
gories reflect the "dark side" of each of these personality domains; the di-
mensions of the five-factor model have their adaptive and maladaptive
aspects. As we have noted, individuals with certain personality character-
istics tend to exhibit certain types of behaviors likely to be found upsetting
to a spouse (Buss, 1991a). From the perspective of COA analysis, it is hy-
 RATER VERSION APPLICATIONS 179

TABLE 5-1 Prototype Items for Each of the Six CCICLCategories

Couples Critical
CCICLItems
Incidents Check
List Domain High end of domain Low end of domain
Emotional is whiny, easily panics, never shows weakness, too calm,
obsessive, moody too much emotional control
Interpersonal too dominant, has too many loner, very passive, aloof, poor
friends, always wants to social skills
be boss
Flexibility dreamer, nonconformist, rigid, intolerant of diversity,
untraditional unimaginative, uncultured
Cooperativeness naive, gullible unable to set cynical, stubborn, selfish, con-
limits, easily manipulated ceited, arrogant, manipulative
Personal responsibility too regimented, miserly, lets sloppy / messy, lazy, unfaithful,
work interfere with family unorganized, self-centered
time
Unassigned items gambler, sexual difficulties, physically abusive, substance abuser,
mocks me in front of others
Note. Items from Couples Critical Incidents Check List, by R L. Piedmont and Rl. Piedmont, 1996, Baltimore:
Author. Copyright 1996 by R L. Piedmont and R.l. Piedmont. Reproduced with permission.

pothesized that rating a spouse higher (or lower) on a given personality

domain than the spouse's self-rating makes it likely that the rater would
identify problems noted in the high end (or low end) for that domain as
sources of dissatisfaction in the relationship.
The CCICL is a simple check list of 133 items that cover a wide range
of interpersonal issues. Individuals check off those behaviors that they
find problematic in their spouse. Individuals also rate, on a 7-point Likert
scale, the amount of dissatisfaction they are currently experiencing in their
relationship. In collaboration with a student, we evaluated the validity of
the CCICL and its utility for COA analysis (see Kosek, 1996a, for a more
complete listing of the study's details).
One hundred and seven married couples (average length of marriage
22 years; range 1 to 58 years) were given the NEO PI-R to complete for
themselves and to rate their spouse. They also completed the CCICL and
the Locke-Wallace Marital Adjustment Test (LWMAT; Locke & Wallace,
1959). The CCICL was scored so that high scores for each section repre-
sented the high end of the personality domain putatively assigned to that
category. (It should be noted that scoring for the CCICL is complex; scores
need to be adjusted statistically for the total numbers of adjectives checked
 180 CHAPTERS

and for gender.) The first order of business was to show that scores on the
CCICL have no correlation with self-rated personality. In other words,
how an individual rates his or her spouse should have nothing to do with
the individual's personality. There were no significant correlations be-
tween the CCICL and self-rated NEO PI-R scores.
The next step was to show that ratings on the CCICL are correlated
with both NEO PI-R observer ratings of the spouse and with the spouse's
self-reported NEO PI-R scores. Table 5-2 presents these results. These re-
sults provide some preliminary validity evidence to the CCICL. First, they
show that an individual's ratings on the CCICL (which indicate the types
of problems the person is having with the spouse) correlate with the per-
son's ratings of the spouse on the NEO PI-R. The pattern of correlations is
consistent with our hypotheses, in that the categories of the CCICL con-
verge with their appropriate dimensional relative on the NEO PI-R. For
example, individuals who are rated as evidencing many emotional prob-
lems on the CCICL are rated as being high on NEO PI-R Neuroticism. A
second interesting piece of validity evidence is that this pattern of conver-
gence between the CCICL and the NEO PI-R extends to the self-report
scores as well, indicating that the behavioral issues people are finding
problematic in their spouses correspond with temperamental dispositions
acknowledged by the spouses themselves. Coupled with the findings that
self-rated personality does not correlate with CCICL ratings, these data
show that the problems acknowledged by individuals are not figments of
their imaginations, or their own self-projections. Rather, they reflect real
problems that are linked to the motivational styles of their partners. Thus,
the CCICL can be considered a valid index for capturing interpersonal is-
sues that are linked to personality dispositions.
With this support in hand, we moved on to the next, and most cen-
tral, question: "Are discrepancies between an observer rating of person-
ality and the corresponding self-report indicative of the kinds of marital
difficulties that the rater is having with his or her spouse?" Although ear-
lier I recommended that subtracting the rating from the matching self-re-
port score was sufficient for determining the actual difference between
scales (15 T-score points was to be considered statistically significant),
such a procedure does not work very well in a research context. The prob-
lem is that difference scores are not as reliable as their constituent ele-
ments <1. Cohen & Cohen, 1983). For example, say that we wish to
perform a COA analysis on the Openness dimension. The reliabilities for
this domain in self-report and observer versions are .87 and .89, respec-
tively. These are quite high reliabilities for personality scales. Costa and
McCrae (1992c) report spouse-self agreement on this domain as .65. Sub-
tracting the self-report score from the rating would result in a difference
 RATER VERSION APPLICATIONS 181

TABLE 5-2 Correlations between CCICL Ratings and NED PI-R Self- and
Observer Ratings.
CCICL Category
Personal
Inter- Coopera- Respon-
NEO PI-R Domain Emotional personal Flexibility tiveness sibility
Husband's CCICL Ratings
Husband's rating of wife
Neuroticism .39' .05 .04 -.12 -.04
Extraversion -.15 .2Sh .18 .01 -.10
Openness -.08 .06 .41' .04 .11
Agreeableness -.3Qh -.13 .16 .4'7'" -.10
Conscientiousness -.22" -.11 -.20" -.04 .3'7'"
Wife's self-report
Neuroticisn .24h -.07 -.11 -.12 .12
Extraversion -.15 .24h .22" .04 -.14
Openness -.13 .14 .39' -.04 .04
Agreeableness -.10 .06 .22' .26b -.04
Conscientiousness -.03 .08 -.13 -.15 .17

Wife's CCICL Ratings

Wife's rating of husband
Neuroticism .35' -.03 -.04 -.07 -.04
Extraversion -.04 .39' .10 -.10 -.05
Openness .06 .12 .22" .04 .06
Agreeableness -.20' -.22" -.05 .42( .13
Conscientiousness -.08 -.18 -.12 -.13 .50'
Husband's self-report
Neuroticism .28" -.13 .02 -.03 -.15
Extraversion -.14 .2~ .05 -.01 -.03
Openness .01 .05 .16 .04 -.05
Agreeableness -.13 -.OS -.11 .22" .17
Conscientiousness -.01 .05 -.08 -.06 .40'
N= 107.
·p<.OS
"p < .01
'p < .001, two-tailed
Convergent correlations given in bold.
Note. Based on a reanalysis of data from Criss-Cross Ratillgs of the Big Five Persollality Dimmsiolls as all Illdex
of Marital Satisfaction, by K. B. Kosek, 1996, Baltimore: Unpublished doctoral dissertation, Loyola College.
Adapted with permission.

score that had a reliability of only .66. Quite a drop! Although for clinical
purposes such a simple arithmetic operation will suffice, for empirical
analysis a more efficient procedure is needed that still reflects the dy-
namic of interest.
182 CHAPTERS

We addressed this issue by creating a residualized rating score. This

was done by use of a simple multiple regression analysis. Here, the de-
pendent variable in the analysis was the observer rating score that was re-
gressed on the independent variable: the self-report score on the same
domain. From this analysis we saved the residual. The residual in this
analysis represents that part of the observer rating that has nothing in com-
mon with the self-report. This score represents qualities that the rater sees
in the target that are not present in the actual person (as determined by the
self-report). Therefore, this residualized index represents the degree of dis-
tortion or amount of misperception the person has of the motivations of
his or her partner.
This COA index should tell us two things about the person. First,
these values should be related to experienced levels of marital satisfaction.
Greater levels of distortion may reflect an inability of the person to either
accurately anticipate or to appropriately respond to his or her partner's
needs. For example, a wife may perceive her husband as being higher on
Neuroticism than he admits to being. Thus, the wife may find him to be
very whiny, obsessive, and moody. These behaviors upset her which may
lead her to dismiss his complaints and ignore his negative feelings. This, in
turn, further frustrates her husband who may feel ignored and emotion-
ally isolated, leading him to experience and express more negative feel-
ings. This pattern can create a negative cycle of conflict that exacerbates the
situation. Further, because she feels that his negative feelings are part of an
ongoing affectively distressed style, she may overlook legitimate, situa-
tionally induced stressors in his life and thus may fail to act either to miti-
gate their impact or to avoid their occurrence.
A second expectation-is that COAscores should correlate with CCICL
ratings. Thus, a man who sees his wife as more Conscientious than she ad-
mits to being, will certainly encounter is!iues with his wife surrounding
her sense of regimentation and desire for organization. In other words,
he may overinterpret the motivational significance of behaviors that
surround his wife's sense of organization and persistence. Correlations be-
tween the CCICL ratings and the residualized COA index provide vali-
dating support for the utility of eOA analysis by demonstrating that
differences between the two sources of ratings are linked to specific points
of contention in the relationship.
Tables 5-3 and 5-4 present the results of these analyses. First, as can be
seen in Table 5-3, correlations between the COA indices and scores on the
Locke-Wallace Marital Adjustment Scale show that the degree of distor-
tion on each of the five personality domains is significantly related to lev-
els of marital satisfaction. The pattern of these relations is similar for men
and women. These results are interesting because they also show that the
 RATER VERSION APPLICATIONS 183

direction of the distortion is very important. For example, individuals are

less satisfied with their relationship when they rate their spouse as being
higher on Neuroticism than the spouse does. However, just the opposite
pattern emerges on the other four domains. Individuals report greater mar-
ital satisfaction when they see their spouses as being higher on Extraver-
sion, Openness, Agreeableness, and Conscientiousness.
This suggests that some types of distortion may be beneficial. Seeing
one's spouse as outgoing, empathic, caring, and responsible seems to con-
tribute to an individual's sense of satisfaction with the relationship. How-
ever, an opposite pattern-seeing the spouse as socially isolated, rigid,
antagonistic, and self-centered-contributes to a greater level of dissatis-
faction.
Table 5-4 presents the correlations between the COAindex and scores
on the CCICL. These data provide evidence for the validity of the COA
index as a measure of marital conflict. As can be seen, for both men and
women there is a similar pattern of association: COA index scores con-
verge with their respective CCICL category. Thus, individuals who rate
their spouse higher on Neuroticism tend to experience conflict with their
spouse surrounding issues such as moodiness, jealousy, and immaturity.
Conversely, when the individual rates the spouse as being lower on Neu-
roticism, then this person is likely to have issues with the spouse sur-
rounding lack of emotional expression.
The bolded correlations (convergent validity coefficients) show that
the discrepancies in NEO PI-R ratings on each personality domain corre-
spond to marital issues outlined in the respective CCICL category. Notice
that these correlations are the largest values in each row and column (with
only one exception, found with the husband's COA rating for Extraver-

TABLE 5-3. Correlations between the Residualized COAIndex and Spouses'

Self-Ratings on the Locke-Wallace Marital Satisfaction Scale
Locke-
Wallace
NEO PI-R based COA Index
Satisfaction
Rating Neuroticism Extraversion Openness Agreeableness Conscientiousness
Husbands _.29b .2Sb .34' .39'
Wives -.23" .37" .44' .33'
N= 107
'p<.05
'p< .01
'p < .001, two-tailed.
Note. Based on a reanalysis of data from Criss-Cross Ratings of the Big Five Personality Dimensions as an Index
of Marital Satisfaction, by K. B. Kosek, 1996, Baltimore: Unpublished doctoral dissertation, Loyola College.
Adapted with permission.
184 CHAPTERS

TABLE 5-4. Correlations between CCICL Ratings and NEO PI-R Residualized
COAIndex.
CCICL Category
Inter- Coopera- Personal
NEO PI-R domain Emotional personal Flexibility tiveness Responsibility
Husband's CCICL rating
COA index of wife
Neuroticism .31' .13 .14 -.06 -.14
Extraversion -.07 .13 .06 -.03 -.13
Openness -.02 -.02 .24h .08 .11
Agreeableness -.29 h -.18 .07 .40' -.08
Conscientiousness -.23" -.17 -.15 .03 .33'

Wife's CCICL rating

COA index of husband
Neuroticism .25~ .02 -.07 -.07 .12
Extraversion .05 .27~ .09 .13 .04
Openness .06 .11 .16 .01 .10
Agreeableness -.17 -.2D" -.01 .37' .07
Conscientiousness -.08 -.23" -.10 -.11 .36'
N=107
'p<.05
'p <.01
'p < .001, two-tailed.
Convergent correlations given in bold.
Note. Based on a reanalysis of data from Criss-Cross Ratings of the Big Five Personality Dimensions as an Index
of Marital Satisfaction, by K. B. Kosek, 1996, Baltimore: Unpublished doctoral dissertation, Loyola College.
Adapted with permission.

sion). Further, few of the off-diagonal correlations are significant (only 4

out of 40). This provides a measure of discriminant validity to the COArat-
ings. These findings tell us that a rating diverging from the respective self-
report indicates not only that a person is having some degree of conflict
with the spouse, but that he or she is having a specific kind of issue with the
spouse. High COAscores on Conscientiousness indicate that the person is
having problems with the spouse's sense of personal responsibility, and
only those types of issues. High COA indices on Agreeableness indicate
difficulties surrounding the spouse's level of caring and nurturance. The
COAscore's being higher or lower than the self-report determines whether
the conflict is over the spouse being too regimented and accommodating
or too disorganized and antagonistic.
These data provide an empirical foundation for the convergent and
discriminant validity of COAindices as a means of identifying specific as-
 RATER VERSION APPLICATIONS 185

peets of marital distress and conflict. eOA indices can be useful for deter-
mining the motivational bases of the conflict. These results show that cer-
tain kinds of distortions are particularly associated with lowered levels of
marital satisfaction (e.g., when the person rates the spouse as being higher
in Neuroticism). But all eOAscores, whether they are higher or lower than
the self-report rating, are associated with particular kinds of issues the per-
son is having with the spouse. These issues become more disruptive as the
overall level of marital satisfaction decreases.
Thus, eOAanalysis is a dynamic approach to understanding couples
because it relies less on actual behaviors of the spouse and more on the
subjective impact these behaviors have on the individual. It is these inter-
pretations of the spouse's behavior that playa major role in how people re-
spond in their relationships (Gottman, 1979; D. K. Snyder, 1982). As D. K.
Snyder (1982) pointed out, "a calm discussion to one may represent veiled
hostility to the other" (p. 193). By capturing these subjective interpreta-
tions, eOAcan allow the therapist to plan interventions that are meaning-
ful to the couple. These interventions can take several forms. For example,
in the preceding quotation it is clear that an individual perceives the low
Neuroticism of his or her spouse as low Agreeableness. A therapist may
wish to engage in some cognitive restructuring to help the person gain a
different interpretation of the spouse's behavior, one more consistent with
its underlying motivations. Or, in the case of an individual who perceives
her spouse higher on a dimension that he himself already scores high on, a
therapist may want to help this husband find better ways of expressing his
motivations in this area, ones that are less upsetting to his mate. In either
situation, the information from the eOA analysis is useful for giving cou-
ples a language for talking about issues that they are eager to discuss. This
language can help them to understand better their spouses' needs and can
promote rapport. A successful outcome in marital therapy would be re-
flected by greater cross-observer agreement at the end of treatment than at
the beginning.
The next section presents some case examples of using eOA analyses
to help better understand the dynamics of the relationship. Although there
are three different comparisons that can be made from performing a eOA
analysis (e.g., self versus self, rating versus rating, and self versus rating),
only the self-reports versus the observer rating information are presented.
This provides an opportunity to become familiar with doing eOA analy-
ses. This is also an opportunity to apply the interpretive skills learned in
the previous chapters for evaluating the self-report results. The reader is
also encouraged to explore the interpretive and clinical value of the other
two types of analyses.
 186 CHAPTERS

CROSS-OBSERVER AGREEMENT ANALYSIS: SELECTED CASES

In performing COA analyses, there are two important steps. The first
is to understand the person being rated. Who do they present themselves
as being? What are their distinguishing personality characteristics and
how would these qualities manifest themselves in an intimate relation-
ship? The second step is to examine the points of disagreement between
the self-report and the spouse's rating. On those dimensions where there is
disagreement, it is important to keep in mind that the person providing
the rating has an exaggerated view of those personality qualities. He or she
sees the spouse as particularly evidencing the negative aspects of these
qualities. The caveat here is that the rater is experiencing some level of dis-
satisfaction with the spouse and the greater the dissatisfaction, the more
dysfunctionality can be inferred from the lack of convergence in the rating.
However, as an aside, given the pattern of correlations presented in Table
5-3, it would be interesting to explore what divergence would represent in
a rater who is experiencing satisfaction in the relationship. Perhaps diver-
gence there may reflect sources of satisfaction being experienced.
It should also be kept in mind that distortions in perception are not
always bidirectional. It is possible that only one partner in a relationship is
experiencing distress, and that person's COAshould be low. Yet, the part-
ner may have a higher level of marital satisfaction and evidence better
cross-observer congruence. Thus, COAcan be useful in identifying which
individual is having a problem with what type of issues. The first case his-
tory provides such an example.

Case History: Marge and Henry Dunbar

Marge and Henry are a Caucasian couple; he is 43 and she is 35. They
have been married for 10 years and have two children. This is the first mar-
riage for both. They began counseling to work on "communication prob-
lems." Henry had recently experienced the death of both his parents and
he had been very unwilling to discuss this with her, although it was clear
he was struggling with many issues of loss. Henry's reluctance to talk
about his loss was also part of a larger pattern of unwillingness to talk
about emotional issues, leaving Marge feeling isolated and alone. In re-
sponse, Marge turned to someone else. Her "emotional affair" drove them
to counseling. They completed the NEO PI-Rs and the CCICL 8 months
into counseling, after they had made significant progress. At this time,
however, Marge indicated a moderate amount of dissatisfaction in the re-
lationship (marking a 4 on a 1 [fIno dissatisfaction"] to 7 ["extreme dissat-
isfaction"] Likert scale). Given that Marge seems to be experiencing more
RATER VERSION APPLICATIONS 187

difficulty with the relationship than Henry, the results of her COA are pre-
sented in Figure 5-2.
This figure presents Henry's self-report along with Marge's rating of
him. As can be seen, there are several areas of disagreement. The overall
profile agreement coefficient for these two sets of ratings is .34, indicating
a low amount of convergence. Marge rates Henry significantly lower on
Neuroticism than he rates himself, suggesting that she sees him as being
very emotionally stable which is consistent with her belief that he exhibits
too much emotional control. The NEO PI-R computer report indicates sig-
nificant differences on the Neuroticism facets of Nl (Anxiety) and N6 (Vul-
nerability). Again, Marge believes that Henry's ability to deal with distress
in his life is functioning too well; it prevents him from experiencing feel-
ings that he should be sharing with her. This interpretation is consistent
with her ratings on the CCICL's Emotional section: She indicates that
Henry is emotionally bland, too calm, and exhibits too much emotional control.
On Extraversion, Marge sees Henry as being significantly more Gregar-
ious than he does himself. The low Extraversion self-rating is consistent with
someone who is not emotionally demonstrative and may appear to experi-
ence little joy. The El rating reveals what may be fueling the conflict; Marge's

FIGURE 5-2. NEO PI-R self-rating and observer rating for Henry Dunbar.
188 CHAPTERS

belief that Henry spends time with friends he is close to and may be sharing
his feelings with them rather than her; this belief may fuel the conflict. One
may wish to explore whether Marge harbors some jealousy toward Henry's
friends, who are few in number but may be very close. Although there is no
significant difference between these two ratings, the Et (Warmth) difference
is interesting in this context. Henry sees himself as a mostly formal and staid
individual, although Marge believes him to be quite approachable. Given this
perception of him as warm and inviting to others, she may be puzzled as to
why she is being excluded from his inner emotional world.
Marge's feelings of emotional abandonment are also supported by the
divergence on the Openness to Feelings facet (03; his self-rated T-score is
56 while her rated T-score is 43). A low score may indicate Marge's belief
that Henry has a limited affective range as well as a belief that he does not
consider emotions important. Again, this corresponds with her CCICL
Flexibility rating of Henry as lacking emotional depth. There is also a signif-
icant difference on the 02 facet (Aesthetics). This divergence, which may
at first appear odd, makes good sense here. The 02 scale evaluates the de-
gree to which someone is not only interested in art and beauty but is emo-
tionally moved by it as well. Henry's perceived artistic insensitivity may
reflect another aspect of his personality that Marge sees as being unre-
sponsive to external events.
The differences on Agreeableness are noteworthy, because Marge sees
Henry as being much more compassionate, caring, and considerate than
he sees himself. Significant differences emerge on the At (Trust), A2
(Straightforwardness), and A6 (Tender-Mindedness) scales. These ratings
may be double-edged in reference to Marge's issues. On the one hand, the
higher ratings may suggest why Marge wanted them to go to counseling.
After all, she sees him as a very decent person and, perhaps, in many ways
a very good spouse. Thus, she is intereste~ in maintaining the relationship
and giving it a chance to succeed despite the problems. As we noted in
Table 5-3, distortions in this direction are associated with higher levels of
marital satisfaction for the rater. These may be the qualities of Henry that
contribute to Marge's experienced happiness with the relationship. How-
ever, these ratings may be another part of Marge's frustrations as well. She
may question why a person who is trusting, straightforward, and caring
would refuse to discuss his feelings with her, especially when he knows
how important such revelations are to her. Why, then, does he refuse her?
These kinds of perceived inconsistencies may be underlying Marge's con-
flicts with Henry and motivating her to seek therapy: to find the answers
that seem to be eluding her.
Finally, on Conscientiousness, Marge rates Henry as being lower
overall on this domain than he himself does, indicating that she may per-
 RATER VERSION APPLICATIONS 189

ceive Henry as being self-centered and selfish, unresponsive to the needs

of others, perhaps using them for his own ends. Yet the one significant dif-
ference on this domain is found for C5 (Self-Discipline), where Marge rates
him higher on the scale than he rates himself. Consistent with the preced-
ing, this discrepancy may reflect Marge's perceptions of Henry as being
very stoic and in control of his feelings and behaviors. He may have a
sense of duty that he is trying to fulfill that keeps him from being suffi-
ciently emotionally vulnerable to Marge. Her rating on the CCICL in the
Personal Reliability section, lets work interfere with family time, suggests that
she may think that Henry tries to escape from his feelings by immersing
himself in his work.
Overall what emerges in this profile is Marge's belief that Henry is, es-
sentially, emotionally unavailable to her. She believes that he has many im-
portant feelings hidden inside him that he seems unwilling to discuss. She
may also believe that Henry has been "emotionally unfaithful" to her. She
sees him as very gregarious and perhaps sharing his feelings with others
rather than her. This may have provoked Marge's jealousy, leading her to,
in tum, seek out others with whom to confer and find succorance.
But what appears from Henry's self-rating is an individual who is not
a very emotional person. His low Extraversion score indicates that he does
not experience many positive emotions and his average score on Neuroti-
cism suggests no experiences of strong negative feelings (although any
negative affect centers on Anxiety and Depression). As a result, Henry may
appear quite unemotional. Feelings are something that he does not expe-
rience much, and his corresponding low scores on Openness (especially
03) indicate that feelings are also not important to him. Thus, he may ap-
pear very stoic and staid. It would certainly be difficult to get great emo-
tional reactions from him (Marge's "emotional affair" may have been such
an attempt). When he is overwhelmed with the stressors of life (high N6-
Vulnerability), he may not resort to emotional outbursts to help relieve
stress; rather, he is more likely to use the defense mechanisms of reaction
formation and rationalization. He may even underplay his physical and
emotional discomforts.
Marge certainly overemphasized the lack of emotional diversity that
Henry experiences. However, his self-rated scores on Nt and N3 indicate
that Henry experiences some negative affect, centering on Anxiety and De-
pression. It is interesting that Marge does not seem to pick this up, and this
may underscore Henry's success at hiding his feelings. Perhaps there is a
need for some exploration as to how Henry expresses his feelings of anxi-
ety and depression, or for some examination as to why he may not be
forthcoming in relating these feelings to his wife. This is perhaps the most
germane question to our analysis. Why is Henry reluctant to share his
190 CHAPTERS

feelings with his wife? To answer these questions, we need to tum to an ex-
amination of Marge's COA profile, which is presented in Figure 5-3.
Figure 5-3 presents Marge's self-report along with Henry's rating of
her on the NEO PI-R. These two profiles converge quite well, having a
profile agreement coefficient of .75, indicating a high level of agreement.
This is also consistent with Henry's marital dissatisfaction rating of "2" on
the CCICL. Both raters acknowledge that Marge experiences a high level
of negative affect, although she thinks she copes with these feelings better
than Henry thinks she does. Both agree on her overall Openness and the
pattern of the facets is similar, although Marge sees herself as being more
dogmatic and concrete.
Despite this overall level of agreement, there are several areas of dif-
ference that need to be considered. Overall, Henry sees Marge as much
more extraverted than she sees herself, especially when it comes to her
own sense of joyousness (E6). Significant differences are also noted on E2
(Gregariousness) and E5 (Excitement-Seeking). Henry does not see Marge
as interpersonally isolated as she sees herself to be. This may explain
Marge's need for more emotional intimacy from Henry; he does not really
appreciate how few social contacts she has. He may be her sole source of
emotional contact, although he may see more activity than there is, or
Marge may not realize the extent of her own social world.
Marge
Self Report
Henry
ranking Marge ----------------

FIGURE 5-3. NEO PI-R self-rating and observer rating for Marge Dunbar.
 RATER VERSION APPLICATIONS 191

Other differences appear on the Conscientiousness domain, on C3

(Dutifulness) and C6 (Deliberation), where Henry rates Marge lower.
Henry's high ratings on N5 (Impulsiveness) and E5 (Excitement-Seeking),
as well as the low rating on C6, may suggest a perception of Marge as very
impulsive and perhaps reckless.
But what is important in this profile is perhaps the reason for Henry's
emotional distance from Marge: Marge's high level of emotional dyspho-
ria. Both parties agree that Marge experiences much anxiety, frustration,
depression, and self-consciousness. There may be some question in
Henry's mind about her ability to manage all of this dysphoria. He may be
unwilling to share his strong feelings of loss and grief with his wife be-
cause he does not wish to burden her with additional, painful feelings. De-
spite Marge's own emphasis on emotions and feelings (see her very high
OI-Fantasy, and 03-Feelings scores), he may look elsewhere for emo-
tional expression.
This attempt at "protecting" Marge from his feelings will certainly be
experienced as a serious rebuff. Marge's high score on N4 (Self-Con-
sciousness) indicates a sensitivity to rejection. She may look to Henry for
affirmation and validation, especially as a wife. The degree to which he is
able to show his feelings to her may be one of her criteria for determining
her success and adequacy as a partner. His turning away from her may
challenge these kinds of feelings. This may lead her to be quite jealous of
his other social contacts; they are perceived as taking over her own role as
emotional helpmate. She may even feel threatened by his few associates.
As you may recall from Figure 5-2, Marge rated Henry as being very gre-
garious, while she rates herself as being very low in this quality. She may
see him as being quite skilled interpersonally, an area where she may feel
inadequate. Marge may feel that these perceived social skills also give
Henry opportunities to find better emotional contacts, opportunities that
she may be unable to pursue, or create, for herself.
Thus, we can begin to see how and why Henry and Marge's issues
have arisen. In therapy, one may wish to explore Henry's emotional world
and help him identify and discuss various emotions more effectively. On
the other hand, the therapist may wish to explore Marge's ability to cope
with negative emotions. Marge's efforts to manage her own affective dys-
phoria may leave her little room to work with others' problems. She may
have too high expectations for what she can handle herself. Thus, helping
her to cope better with her own emotions may make her more available to
provide assistance to Henry. One must also explore whether she really is
interested in Henry's talking to her about his feelings, or whether she re-
ally wants an opportunity to talk more about her own feelings.
Given that Henry is not a very emotional man himself, it is possible
that Marge overwhelms him with her own strong emotions. He may need
192 CHAPTERS

to distance himself in order to maintain his own sense of balance and per-
spective. Although Henry is clearly aware of the great amount of ongoing
emotional distress Marge experiences, some exploration of his responses
to these feelings would be welcome. Perhaps helping Henry affirm and
support Marge's personhood more clearly may assuage and comfort her
feelings of inadequacy.
Another area of exploration is the value systems of these two individ-
uals. Both present low self-report scores on 06 (Values) and acknowledge
such in each other. Low values on this facet indicate that the individual has
a very strong network of values, a clear sense of right and wrong. This
value system is also not open to negotiation or modification; it is a rigid set
of beliefs. To what degree do these two sets of values correspond with one
another? What are the expectations that such values create? Is Marge try-
ing to live up to an unrealistic image of a good wife? Has she set standards
too high for their own relationship? In completing the CCICL, Henry
checked only two items, can't say "no" to others and sets unrealistically high
standards for self. These two items may reflect Marge's strong need to be ac-
cepted by others; she looks for validation through serving others and liv-
ing up to high standards of success. Then again, what values does Henry
hold about marriage? Does he believe that a man should not appear emo-
tionally vulnerable? Does he consider discussions about his feelings inap-
propriate to share with his wife? These issues would need to be explored
with this couple.
Overall, this case history provides us with some important concepts.
First, dissatisfaction in a relationship can corne from many sources. In eval-
uating the COA profiles for Marge, we learn that there are three areas po-
tentially contributing to her dissatisfaction. First, her own high levels of
Neuroticism speak to a general, distressed lifestyle. Insecure and anxious
on the inside, she may find all of her relationships tenuous and difficult to
manage. As the data presented earlier showed, individuals high on Neu-
roticism experience less satisfaction with life in general and their marriages
in particular. Second, her perceptions of Henry are indeed skewed in cer-
tain areas, leading to interpretations of his behavior that reinforce her feel-
ings of isolation and emotional abandonment. Finally, Henry's self-report
indicates the presence of various characteristics <e.g., his own lack of emo-
tionality, his interpersonal aloofness) that also contribute to Marge's lack of
satisfaction. Obviously, all three of these areas need clinical attention.
A second insight from this case is that marital dissatisfaction does not
have to be a mutual experience. Clearly Henry does not have any major is-
sues with Marge, and finds the relationship to be meeting his needs. Given
Henry's profile and his tendency to dampen emotional issues, one must
raise the question as to whether Henry is merely denying some important
 RATER VERSION APPLICATIONS 193

issues in the relationship. It may also speak to his own need to avoid con-
flict or there may be some degree of secondary gain for him by keeping
Marge feeling insecure in the relationship. These issues need to be ex-
plored as well. One benefit of Henry's high level of marital satisfaction
may be his willingness to accept changes in the relationship in order to im-
prove its quality for Marge.
Even though only one person in this relationship was experiencing
marital distress, an exploration of both COA profiles can provide useful in-
formation about the relationship. I hope this case study begins to illustrate
the usefulness of COA analyses for understanding the dynamics of any
given relationship. This approach should be useful for working with pre-
marital couples as well; they can explore the motivational expectations of
their partners and begin to see the life directions each of their personalities
would wish to move toward. The COA process can help support the rela-
tionship by fostering greater self-insight, greater emotional insight into the
partner, and enhanced communication skills in dealing with intimate topics.
The following cases are presented to help extend your ability to eval-
uate COA profiles and infer useful clinical insights.

Case History: Sarah and Bob H.

This Caucasian couple has been married for 12 years, having no chil-
dren. Bob is a chief counsel (lawyer) who has been unemployed for the
past 21/2 years. Sarah is a human resources manager. Originally, Bob carne
to therapy to deal with his ongoing frustrations about being unemployed.
Eventually, though, he realized that he was experiencing a great deal of
friction in his relationship with his wife. He decided to bring Sarah into
treatment to help manage his problems associated with being unem-
ployed. The couple had been in treatment for 5 months when the COA
profiles were obtained. Because Bob is the identified patient, his COA pro-
files are presented first in Figure 5-4.
On his CCICL, Bob rated a moderate amount of dissatisfaction and
conflict in his relationship. Figure 5-4 presents Sarah's NEO PI-R self-report
and Bob's observer rating. The profile congruence index is only .48, sug-
gesting a low level of agreement. An evaluation of the two sets of ratings
indicates differences on 12 scales. In terms of the domains, Bob rates Sarah
higher on Neuroticism and Conscientiousness. As was pointed out in
Chapter 4, this type of profile indicates that Bob sees Sarah as a very com-
petitive person, one who uses her ambitiousness to compensate for internal
feelings of inadequacy. Meeting her high standards of success provides
some brief solace to her intrinsic feelings of unwantedness and inadequacy,
while failure provides a painful deluge of feelings that confirm her worst
194 CHAPTERS

Sarah H. with spouse's rating

Sarah
Spouse -----------.---.

FIGURE 5-4. NEO PI-R self-rating and observer rating for Sarah H.

fears: that she is useless and of no value. Thus, there is a strong emotional
edge to Sarah's ambitions. Sarah is seen as competitively climbing the pro-
fessional and social ladder. in order to find self-acceptance.
This interpretation is further confirmed when the facet scales are exam-
ined. Bob rated Sarah significantly higher on Depression, Self-Conscious-
ness, and Achievement. These ratings suggest that Bob sees Sarah as a very
focused and intense person who may have a very low tolerance for failure.
Another area of disagreement is on E5 (Excitement-Seeking). Sarah is
seen as a risk taker and thrill seeker. Coupled with the preceding percep-
tions, this rating may indicate that Sarah enjoys being "where the action
is." Her drive to succeed is attracted by opportunities for high profile in-
volvement in challenging tasks. Again, this perception contrasts starkly
with Bob's professionally lackluster situation.
A final difference of note is Bob's rating of Sarah on Al (Trust). Such a
score indicates a perception of Sarah as being cynical and skeptical, always
assuming that others are dishonest or dangerous. This perceived inability
to recognize the good in others may underscore Bob's belief that Sarah has
no belief in him or his ability to find gainful employment.
Overall, Bob perceives his wife as a very ambitious woman who
values success in herself and others. Distrustful of others, self-assertive,
 RATER VERSION APPLICATIONS 195

energetic, and hardheaded (low A6), Sarah is not seen as taking Bob's un-
employment well. Such "failure" may be completely unacceptable to
Sarah, who may see this situation as reflecting negatively on her as well.
Given Sarah's perceived need to present a highly successful face to the
world, Bob's unemployment may be an embarrassment to her that she is
trying to distance herself from. This picture is complemented by Bob's
CCICL ratings of Sarah as feeling inadequate, can't receive criticism, hostile at-
titude toward others, overcontrolling, always wants to be boss, cynical, always
does things her way, and, too neat.
However, Sarah's self-report does not reflect many of these percep-
tions. She does not indicate the levels of Self-Consciousness and Achieve-
ment Striving seen in the rating. She does not see herself as "driven" to
succeed in order to compensate for feelings of inadequacy. Rather, if she
experiences any negative affect, it is anger and frustration. Although she
feels herself quite competent (CI) and self-disciplined (C5), she does not
see herself as having a very high aspirationallevel (C4). Although Sarah
sees herself as being low on Agreeableness, her level of Trust (AI) is not as
low as Bob indicates; she may be skeptical of people's motives at times but
maintains a general positive attitude toward others.
These varying perceptions of Sarah are significant given Bob's current
employment status. He may be very sensitive to his own failures and in-
ability to find employment and may be projecting his own ambitiousness
and self-accusatory ideations onto Sarah. She may be quite frustrated at
the current situation, especially with the financial hardships Bob's unem-
ployment has caused. As a result, she may respond with anger and hostil-
ity, but this negative affect may not stem from her own sense of personal
inadequacy. Bob may interpret this anger not as frustration, but as a puni-
tive attack on him personally. He may even believe that she is gloating
over his loss of employment.
In order to better understand the areas of conflict in this relationship
we need to consider the other half of this profile: Bob's self-report com-
pared with Sarah's rating, which is presented in Figure 5-5. On the CCICL
Sarah also rated a moderate amount of conflict and dissatisfaction in the
relationship. The profile agreement coefficient for these two profiles is -.52,
indicating a very low level of agreement. In fact, this value shows that the
rating is quite the opposite of the self-report. Given the level of conflict in
this relationship, this amount of discrepancy is not surprising.
Bob saw Sarah as being very ambitious and not tolerating his appar-
ent failures very well. Looking at Bob's self-report we can see that he
scores very high on Self-Consciousness (N4) and Achievement Striving
(C4). Therefore, it is he who is ambitious and tries to succeed in order to
stave off his own personal feelings of inadequacy. Failure represents an
outcome that awakens very painful feelings that he may find difficult to
 196 CHAPTERS

Bob H. with spouse's rating

Bob
Spouse ----------------

FIGURE 5-5, NED PI-R self-report and observer rating for Bob H.

manage or even accept. Thus, our earlier hypothesis that Bob is projecting
his own competitiveness and punitive feelings onto Sarah seems sup-
ported here. His extemalization of his pain is placing a high degree of
stress on the relationship. His repeated failures to find employment may
create a number of frustrations (both interpersonal and financial) on Sarah,
who may respond with anger which is interpreted by Bob as an attack on
his adequacy as a person.
In Sarah's perceptions of Bob, there are a number of inconsistencies.
Sarah sees Bob significantly lower on overall Neuroticism, indicating a be-
lief that in general Bob is a rather emotionally stable individual who may
not show his emotions very clearly. This is consistent with her CCICL rat-
ings of unconcerned about impressions made on others and too much emotional
control. On the facets, her rating of him on N4 is significantly lower than
the self-report, suggesting that she may not see how inadequate he feels
about himself. She may not recognize the effect on his self-esteem his un-
employment has had. More than a professional setback, this situation af-
fects his sense of personal adequacy.
One potential area of conflict for Sarah may center on how well Bob is
able to manage the many negative feelings this situation is creating. Bob
 RATER VERSION APPLICATIONS 197

feels that he is coping well (N6) while Sarah certainly does not agree. There
may be some issues surrounding the best way for the couple to manage
Bob's unemployment. Looking to Openness, Sarah believes Bob to be
much more closed and rigid than he sees himself, especially on facets 02
(Aesthetics), 03 (Feelings), and 06 (Values). Bob has a set way of wanting
to do things (low 04) and enjoys structure and direction. However, there
may be a strong reluctance on Bob's part to try things (e.g., employment
opportunities) that may not conform to his expectations or that may not
provide as well-defined a job description as he would like.
This may lead Bob to forgo certain opportunities that present them-
selves. Given the length of his unemployment, Sarah may be urging him to
try new types of jobs that may be available, but Bob resists. He may be
holding out for a certain type of position that Sarah may see as unrealistic
or not possible at this time. As such, Sarah may be finding Bob's intransi-
gence and unwillingness to explore new types of positions frustrating.
Given his low Agreeableness scores (which she agrees with), he can be
quite bullheaded and stubborn. This is consistent with her CCICL ratings
of sees the world in black or white terms, too set in his/her ways, unimaginative,
and stubborn.
Another interesting area of divergence concerns Conscientiousness.
Overall, Sarah rates Bob significantly lower on this domain as well as
lower on the facets of C3 (Dutifulness), C5 (Self-Discipline), and C6 (De-
liberation). Bob sees himself as being organized, focused, and motivated.
Sarah believes just the opposite: He is unorganized and unambitious.
CCICL ratings on the Personal Responsibility domain were unorganized
and sloppy/messy. For Sarah, Bob's behavior is more self-centered and
passive. He does not seem to be energetically pursuing new career oppor-
tunities. Sarah's combined ratings of low Neuroticism and low Conscien-
tiousness suggest that she sees Bob as being informal, noncompetitive, and
easygoing. He may not share the same notions of success and material ad-
vancement that she has.
Sarah's image of Bob is one of a man who is not very ambitious or suc-
cess oriented. Rather, he has some very explicit ideas about how he needs
to pursue his goals which he is very committed to, but which may not be
the most efficient or productive from Sarah's perspective. She does not
think that his way of doing things is working well at all, yet he remains
doggedly attached to that process and will not consider any alternatives.
Rather than focus on the task at hand, Bob is perceived by Sarah as lashing
out at her and perhaps making her responsible for his inability to find
work. Although Bob sees himself as being assertive and proactive in this
situation, Sarah instead sees him as becoming involved with his own feel-
ings and circumstances and not really dealing with the issues at hand.
 198 CHAPTERS

There is indeed much turmoil in this relationship for both individuals,

and the sources of distress are different. Bob seems to be projecting his
own feelings of inadequacy onto Sarah, who is perceived as being castrat-
ing and demeaning. Sarah, on the other hand, is quite frustrated at Bob's
inability to organize himself to pursue employment. He seems to have
locked himself into a very small subset of opportunities that may not be
realistic for him at this time. Rather than acknowledging the faults with his
approach, he instead attacks her emotionally.
An interesting dynamic to consider with this couple is their individual
low scores on Agreeableness. Both of them are low on Compliance (A4);
they both like to have things their own way. They are both low on Straight-
forwardness (A2), indicating a desire to manipulate people to get what they
want from them. In some sense, these two people represent an irresistible
force meeting an immovable object. They are both out to have their own
way and will use whatever strategies it takes to accomplish their goals.
Both individuals want Bob to find a job that he is happy with, but they
probably have very different ideas about how this should happen. Also,
both individuals see themselves (and each other) as being quite competent,
so naturally each thinks that he or she is capable of finding the solution.
Having two antagonistic individuals in a relationship will certainly
lead to much conflict. After all, they are naturally confrontative individu-
als. They will invariably look for each other's weak spot. In this conflict,
Bob seems to focus on Sarah as being inadequate and perhaps overemo-
tional. Sarah responds by underestimating Bob's ability to really work
through his problem and succeed. Bob's perceptions of affective dysphoria
challenge Sarah's sense of Self-Discipline and Competence while her per-
ceptions of low Conscientiousness may challenge Bob's sense of personal
adequacy. Thus, the therapist may want to examine how these two indi-
viduals may be trying to hurt each other in an effort to manipulate the cir-
cumstances.

Case History: Alan and Karen G.

Karen G. is a 32-year-old Caucasian female who presented herself to
treatment because of conflict in her marriage. She believes that her hus-
band blames her for everything that seems to go wrong in their lives. She
believes that he expects her to "bow down to his needs" in all ways. Three
months into therapy her husband, Alan, joined in. He is a 33-year-old
African American who complains that he does not feel he gets respect from
his wife. Both parties feel that there is very little communication currently
going on in the relationship and that there is a great deal of interpersonal
conflict. They have been married for 11 years. They had been in treatment
RATER VERSION APPLICATIONS 199

for 51~ months at the time they completed the COA profiles. Because Karen
was the presenting client, her COA profile of Alan is presented first in Fig-
ure 5-6.
On the CCICL Karen indicates a moderate amount of dissatisfaction
and conflict in the relationship. The overall profile agreement between
Alan's self-report and Karen's ratings is .86, indicating a high level of
agreement. Nonetheless, there were 12 scales on which there is significant
difference.
On Neuroticism, Karen and Alan agree on Alan's overall low level.
On the CCICL Emotional domain Karen indicated issues about Alan hav-
ing too much emotional control, never shows concern for the future, and seems to
want to control my emotions/desires. On the CCICL she notes that Alan
"doesn't plan for much-will call at 4:00 P.M. to ask me to get a sitter for 7
P.M. so we can do whatever just came to him. But he won't PLAN a date."
Thus, Karen believes Alan has a wide range of interests that he wishes to
be involved in, and he wants to do it now. Notice Karen's high rating of
Alan on N5 (Impulsiveness) along with the low ratings on C5 (Self-Disci-
pline) and C6 (Deliberation), which are significantly lower than the self-re-
port scores. This pattern suggests a very impulsive approach to life, one

AlanG.
Self Report
Karen
ranking Alan c
N E 0 A 1 2 3 4
:
5
.
6

FIGURE 5-6. NEO PI-R self-rating and observer rating for Alan G.
200 CHAPTERS

characterized by a need for immediate gratification. He may wish to act on

whatever catches his eye for the moment, making it difficult for Karen to
anticipate his desires and to implement the logistical control of the house-
hold necessary to accommodate Alan's wishes.
In contrast to these very open aspects of Alan, Karen's very low rating
on 03 (Feelings), shows that there are also areas of rigidity that are causing
conflict. Alan does not seem to be very emotional nor to care for the feel-
ings of others. On the CCICL Flexibility domain Karen indicates that Alan
is insensitive to the feelings ofothers. The anger and frustration generated by
Karen is in response to Alan's not taking her feelings seriously. Because
emotions are not seen as valuable by individuals low on 03, Alan may be
ignoring not only Karen's affect but its underlying message.
On Agreeableness, Karen sees Alan as being higher on Al (Trust) and
A4 (Compliance). This combination suggests a portrait of Alan as naive,
gullible, and easily led. Rather than the wary, assertive individual he pre-
sents himself as, Karen sees Alan as being overly influenced by others. On
the CCICL she notes Alan as being gullible and overly trusting. Karen may
see Alan as giving too much consideration to the desires and wishes of his
friends and ignoring her own situation and needs. Rather than respond to
her needs, Alan instead expects Karen to make whatever arrangements are
necessary to allow him to do what he wants.
What emerges from this profile is a picture of a-man who sees himself
as an assertive, dominant individual who is certainly in control of his en-
vironment. He has a very clear set of values that he relies on to provide
him with a considered, well-intentioned view of the world. Although he is
warm and friendly toward others (high E1), his lower Agreeableness
scores indicate that he p.refers to have others accommodate his needs, and
he may perceive himself as the real "mover" behind any group. His low
06 (Values) score suggests a strongly held belief system, and the therapist
may wish to consider what types of expectations these values may gener-
ate concerning marriage and the roles of husband and wife.
This is in contrast to what Karen sees. She does not see this masterful,
ascendant individual who is in control of his environment. She perceives
Alan as a passive, follower type, tending to go along with the majority,
rather than as a leader. This passivity, though, does not seem to extend to
his marital relationship, where he does not seem to really care about his
wife's desires. Despite repeated attempts at making her wishes known,
Karen feels that Alan just does not listen. This is, perhaps, what is most
frustrating to Karen. Alan seems influenced by a wide range of people, ex-
cluding her. These contradictions in Alan also extend to his own personal
motivations. On the CCICL Karen notes "lazy about 'housework' and nor-
mal maintenance and a workaholic about hobbies." Alan seems to have
 RATER VERSION APPLICATIONS 201

the time to devote to his awn passions but does not seem to be interested
in investing in activities that do not center around his own needs. As Karen
also commented on the CCICL "doesn't seem to desire intimacy with chil-
dren-or me, unless he's in need sexually."
Thus Karen's issues may center around her feeling relegated to sec-
ond or third place in Alan's life. He seems to have emotionally disengaged
from their relationship, except for when his needs have to be met. But
Karen may not recognize the value system that Alan is working with. His
very low 06 score indicates someone very much committed to a particular
tradition and he may see marriage in a very different way from Karen.
These value differences would certainly need to be explored. One particu-
lar area of concern would be Alan's low Dutifulness (C3) score, indicating
someone who may be unreliable and undependable. Certainly this lack of
follow-through on Alan's part contributes to a perception in Karen that he
just does not care about, nor give much importance to, those things having
to do with their home life.
In contrast, the COA profile for Karen is presented in Figure 5-7. On
the CCICLAlan indicates a very high level of dissatisfaction and conflict in
his marriage (a rating of 6 out of a possible 7). The profile agreement coef-
ficient comparing the self-report and observer ratings was only .12, indi-
cating a very low degree of agreement. As readily can be seen in Figure 5-7,
Alan perceives his wife as being significantly higher on all aspects of Neu-
roticism. Alan's perception of Karen is as a very emotionally distressed
person who is unable to manage her emotional world. Although Karen be-
lieves herself to be stable and capable of coping with a wide range of stres-
sors, Alan believes her to be overly sensitive to life's issues. As Karen's
NEO PI-R interpretive report, based on these ratings, describes her, "In
coping with the stresses of everyday life, this individual is described as
being likely to react with ineffective responses, such as hostile reactions to-
wards others ... her general defensive style can be characterized as mal-
adaptive and self-defeating."
Much more convergence is seen on Extraversion, although there is a
significant difference on E2 (Gregariousness). Karen is perceived as being
much less sociable than she sees herself. On the CCICL Alan noted that
Karen has poor social skills. These scores portray a woman who is relatively
reserved and quiet, who takes a very passive role in groups. Not particu-
larly cheerful or optimistic, Karen prefers to pursue a more mundane, low-
key lifestyle. It may be that Alan's preferences for spontaneous adventures
prove to be overstimulating for Karen or, given her very low scores on 04
(Actions), may threaten to disrupt her own schedule and routine. Her low
scores on 05 (Ideas) and 06 (Values) suggest an individual who also has a
very clear set of values that guide her perceptions of the world. She is,
 202 CHAPTERS

Karen G.
Self Report
Alan
ranking his wife N E 0 A C

N E 0 A

1f
-<::I:
." ~

:I:
.c . .r
:;:'" ,,-
:~
~.

"e'" ~:
~ .
:" .
~
iil
'""
~
.~

1f
-<
~

FIGURE S-7. NEO PI-R self-report and observer rating for Karen G.

herself, quite dogmatic and has established for herself a very clear and
well-delineated world. Alan's behaviors may too frequently disrupt these
cherished patterns.
The Agreeableness domain portrays Karen as being quite considerate
and caring. Of particular interest is her very high score on A4 (Compli-
ance). In conjunction with the low Assertiveness (E4) rating, this indicates
someone who is very passive and accommodating, who wishes to follow
along with the group and do what is expected. This is curious given
Karen's complaints that Alan always wants to do things his own way with-
out much concern for her feelings. Why would such an accommodating,
passive person have so much difficulty with following along with some-
one's requests? The answer to this question may be found in Karen's very
low 04 (Actions) and 06 (Values) scores. She has certain expectations
about how her clearly structured world should operate. Being asked to do
things that fall outside these boundaries may be perceived as threatening
or inappropriate. In response Karen may refuse to acquiesce.
Finally, on Conscientiousness, Karen is perceived as being quite low
on all facets, suggesting a perception of Karen as being self-centered, care-
less, and disorganized. She is not seen as being very ambitious and does
 RATER VERSION APPLICATIONS 203

not always think through her situation before acting. She does not feel
very competent, nor does she feel that she is in control of events in her life.
She is not seen by Alan as someone who can be counted on to follow
through on her commitments. Alan notes on his CCICL that Karen is un-
organized.
Overall, Alan perceives his wife as someone who "bitches, moans,
and complains about everything." She is a mostly passive individual with
little drive or gumption, yet does not seem to be very receptive to his sug-
gestions for activity in the relationship. Alan's overtures are perceived by
Karen as being whimsical and unrealistic because he does not appreciate
all the work that needs to be done to do what he wants, especially when it
is demanded on the spur of the moment.
These COA profiles outline some important, self-maintaining areas of
conflict. Alan sees himself as an active, assertive person who likes to try
new and different things. Routines bore and frustrate him, so he likes to
give variety a try. His requests to have Karen join him in these new under-
takings may initiate a negative sequence of events. First, Karen is not a
well-organized woman and is not very task oriented. Thus when Alan sug-
gests that they go someplace different for dinner, he does not recognize
that Karen does not like to have to "pick up the details" to make the sug-
gestion work. She may doubt her ability to do it correctly or may fear mak-
ing a mistake (e.g., forgetting to make the reservations at the restaurant,
not calling the babysitter). Karen appears very sensitive to task require-
ments and feels that she may not have the energy or desire to make the
necessary efforts.
Asecond issue is Karen's desire for structure and routine. If they are to
go out to a restaurant, Karen will want to patronize her "tried and true" lo-
cations. Alan, on the other hand, will want to explore new cuisines and
restaurants. This will be an ongoing source of frustration for both. Once
Karen gets into her routine, she will not like to have to shift things around;
this is uncomfortable for her. Karen's resistance is perceived by Alan as
being an overreaction to his requests. Given Alan's very low score on 03
(Feelings), he is not very sensitive to feelings and emotions; they are mostly
irrelevant to him. Thus, Karen's affective responses may overwhelm him
and he may "tune out" Karen's complaints, demeaning them as just so
much nonsensical ranting. Thus he rates Karen high on Neuroticism.
Not wanting to be locked into a particular pattern, Alan may retreat
into his relationships with others or his personal hobbies to find the stim-
ulation he needs. Although Karen believes that she is able to cope reason-
ably well with Alan's detachment, her normally compliant, externally
oriented nature feels the lack of a directing force. Her attempts to obtain
Alan's attention may only exacerbate the situation.
 204 CHAPTERS

A third issue to consider is the value systems each of these individu-

als hold. Both have very low self-report scores on 06, indicating very rigid
belief systems. Some of the conflict experienced here may be due to the
varying expectations these people have of their partners. Alan expects to
receive unquestioning loyalty from his wife, who should comply with all
of his requests. Although Karen's low E3 and high A4 scores may not tem-
peramentally oppose her to this expectation, demands to change the ways
things have been done may create more emotional resistance.
Karen's expectations may focus on more down-to-earth tasks. Given
her generally low Conscientiousness scores, she feels a need for more sup-
port from Alan around the house and in getting daily chores completed.
Alan also needs to show increasing sensitivity to Karen's more gentle na-
ture and the ways she may make requests of him; he may tend to easily
dismiss her more reserved and submissive style. He needs to learn how to
balance his more bottom-line-oriented information style with her more in-
tuitive approach. Given Alan's dominant, surgent style, his lack of empa-
thy (low 03) may lead him to unintentionally bruise Karen's feelings as
well as making him insensitive to any emotional appeals she makes to
him. On the other hand, Karen's more agreeable, passive nature makes her
unable to break through into Alan's world. She may need to find more
forceful ways of confronting Alan and communicating her needs to him.

CLINICAL YIELD FROM CROSS-OBSERVER AGREEMENT ANALYSIS

Cross-Observer Agreement analysis provides a tremendous amount

of information about a couple. The contrast between the image an indi-
vidual presents and the impression generated in the spouse opens a win-
dow into the personal characteristics of both individuals that may be
underlying their dissatisfaction and conflict. There are three benefits of
using this approach.
First, the assessment process itself provides an opportunity for indi-
viduals to clarify the images they hold of their spouses and themselves.
During times of acute marital distress many dynamics may be operating
and many emotions may be in play, all of which create a sometimes chaotic
presenting picture. Conducting the COA analysis enables clients to sift
more carefully through their feelings and then to present them in a
medium very amenable to discussion.
Second, COAanalysis allows a therapist to contrast various aspects of
each person to their partner (e.g., self-report versus self-report; observer
rating versus self-report). Each of the contrasts provides new insights into
the dynamic nature of the relationship. Presenting these profiles to couples
can help them engage in a systematic, focused discussion of any areas of
 RATER VERSION APPLICATIONS 205

discrepancy. Such a dialogue can help each member better understand the
issues his or her partner experiences in the relationship. The five-factor
model provides a clear and common language to use in this discussion.
Each person should come away with a new understanding of their partner
and themselves.
Finally, COAanalysis provides the therapist with important informa-
tion about where and how to intervene with the couple. As we saw in the
previous case histories, the discrepancies between self- and observer rat-
ings indicate specific, salient issues emerging in the relationship. These are
the points at which intervention is necessary. With Alan G., the "where" of
an intervention is with his low Conscientiousness ratings. A therapist may
want to work toward making him appear less self-centered and fickle, and
more sensitive to Karen's emotional needs and the various strategies she
uses in expressing them. The "how" to intervene would want to consider
this couple's low Openness scores. Such individuals may appreciate a be-
havioral approach rather than a cognitive or dynamic one.
COA analysis is a relatively new area. There are few data available
that contrast various personality ratings. More work needs to be done to
establish a better empirical foundation for using test scores in this manner.
The CCICL was useful in this process: It highlighted the specific issues
these couples were experiencing. Such an instrument can pinpoint areas of
conflict. Although such an instrument is useful, it is by no means required
for doing a COA analysis. The data presented here are encouraging and
certainly showcase the heuristic value of this approach.

OBSERVER RATINGS AS PREDICTORS OF OUTCOME

As noted in the beginning of this chapter, most assessment protocols

rely on self-report inventories. When we want to know something about a
person or to make predictions about his or her life, we tum, naturally, to
the individual for information. Other sources of information, such as ob-
server ratings, are seen as adjuncts to assessment rather than as sources of
information useful in their own right. In the first part of this chapter, ob-
server ratings were used to show the validity of a self-report that may have
been unnecessarily rejected. Then, with couples, ratings were shown to
serve as a counterpoint to the self-report, outlining areas of perceived dis-
tortion between two people. In this section I take a different tack and look
at an observer rating as a stand-alone source of information that can be
useful in making predictions about a person.
Hogan (1991) outlined the fundamental differences between a self-
report and observer rating of personality. For an observer, personality
206 CHAPTERS

represents the target's public self or social reputation. For a self-rating, per-
sonality represents the structures, dynamics, and processes inside the per-
son that explain why he or she acts in a certain way. Thus, information
obtained from these two sources convey, to some degree, very different as-
pects of the individual. As Hogan noted, in some circumstances where the
assessment goal is prediction of behavior (e.g., personnel selection), repu-
tation may be a most important quality to measure. After all, the best pre-
dictor of future behavior is past behavior (Wernimont & Campbell, 1968);
one's reputation, which emerges over one's history of behavior with oth-
ers, should be a singularly useful predictive index.
A number of studies have evaluated the predictive utility of observer
ratings. Digman (1972) reported correlations in the .50s between elemen-
tary school teachers' ratings of Conscientiousness and high school grade
point average. Scheier, Buss, and Buss (1976) showed that observer ratings
of personality had greater predictive validity than did self-ratings con-
cerning the level of awareness of one's own aggressive behavior and re-
lated affective reactions. John and Robbins (1993) suggested that under
highly evaluative circumstances an individual may bias his or her self-
report and thus attenuate the validity of the measure. Yet an observer rat-
ing is not susceptible to the ego-involving motivations of the target, and
therefore would not suffer a decline in accuracy. These authors showed
that self-peer congruence declined when the rating was on an evaluative
dimension while peer-peer congruence remained at comparable levels of
convergence regardless of the evaluative nature of the term being rated.
Mount, Barrick, and Strauss (1994) provided the most empirically
compelling support for observer ratings. They gathered a sample of 105
sales representatives who completed a self-rating on the Big Five person-
ality domains. These individuals also had their supervisor and up to five
coworkers and five customers rate them on the same personality domains
as well as on several work performance indices. The results indicated two
important findings. First, the observer ratings of personality were signifi-
cant predictors of performance outcome (only rated Neuroticism did not
correlate with any of the performance criteria). Interestingly, the only self-
report score that correlated with performance was Conscientiousness.
Thus, ratings may provide more criterion-relevant information in this con-
text than self-reports.
The next finding concerned the incremental validity of the observer
ratings. Incremental validity (discussed in more detail in the next chapter)
determines whether a score provides any additional information over and
above that already obtained by existing measures. Hierarchical multiple
regression analyses were conducted in which the performance ratings
were the dependent variable. On the first step of the analysis, self-reported
 RATER VERSION APPLICATIONS 207

personality scores were entered. On the second step the observer ratings
were entered. A partial F-test determined whether the variables entered on
Step 2 added any explanatory variance over that already accounted for on
Step 1. The results showed that for the four personality domains of Extra-
version, Openness, Agreeableness, and Conscientiousness, the observer
ratings provided additional, explained variance in the performance crite-
ria over and above whatever contribution was made by the self-reports.
These findings highlight the value of observer ratings as predictors in
their own right. They provide useful information about an individual that
is not entirely redundant with a corresponding self-report. This is because
observers have a different perspective on the target than the target has of
him- or herself, and these insights have their own predictive validity. Fur-
ther, an observer rating is not influenced or distorted by the same kinds of
motivations as a self-report. Thus, under certain circumstances one may
prefer to use a rating in place of the self-report.
Concerning validity research, these data emphasize the need to dis-
tinguish between the validity of a construct and the manner in which the
construct is measured. As Mount and colleagues (1994) showed, at times
an observer rating was a stronger predictor of performance than a self-
report. Therefore, determining the actual validity of a particular construct
in predicting some outcome on the basis of only one information source
(e.g., self-report) may underestimate the true strength of the relationship.
As Mount and colleagues showed, the relationship between personality
and performance was significantly stronger when both the self-report and
observer ratings were included in the prediction equation.
Although these data were obtained in a work setting, the lesson
should not be lost for the clinical environment. The best evaluations are
those that rely on multiple sources of information and look for points of
convergence. Obtaining similar diagnostic indicators from diverse mea-
sures and informants increases confidence in the accuracy of the inferences
drawn. Observer ratings are a useful piece of any assessment process and
complement information obtained from self-reports. Under circumstances
where the validity of self-reported data may be questioned, obtaining in-
formation from knowledgeable informants provides a useful substitute.
Certainly more research is needed for examining the utility of ob-
server ratings. Although ratings do not suffer from the same sources of
bias and distortion as self-reports, they do have their limitations and some
evaluation of their sources of bias would be helpful. Observer ratings are
an assessment technology that has been underutilized. The availability of
an established, normed, rating instrument in Form R of the NEO PI-R
should help make this type of data more accessible, and useful, to re-
searchers and practitioners alike.
 CHAPTER 6

RESEARCH APPLICATIONS
WITH THE NEO PI-R

Clinical work is a very time-consuming endeavor filled with many pres-

sures and concerns. With all the time constraints it seems reasonable to ask
whether the introduction of a new assessment instrument is worth the
time it consumes. Although I have already outlined how the NEO PI-R can
be useful in providing clinically relevant information, there are three other
benefits that also need to be considered. First, clinicians must and do have
an investment in assessing the quality of their own work. How effective
are one's interventions? In what ways does a therapist impact their clients?
Answers to these questions help therapists fine-tune their therapeutic
skills, pointing out particular strengths and weaknesses. Second, in work-
ing with clients on growth issues, it becomes important to have measures
that will provide information on the adaptive capacities of the individual.
Unlike many clinical instruments, the NEO PI-R can speak meaningfully
about coping strategies and interpersonal styles, among other qualities.
These insights focus on who a person is and the directions they are likely
to move toward rather than emphasizing deficiencies and weaknesses. Fi-
nally, there are the pressures of managed care, which continually call for
greater documented efficiency in treatment.
Although a great deal of energy is devoted to this last area, and more
is said about managed care later, the first issues are quite important. Many
times clinicians rely on clinical interviews to gather client information, but
much of this dialogue usually surrounds finding out the problems or is-
sues that have led the person to treatment. It is not until after therapy is
well under way that attention focuses on more latent personal dimensions
and their role in causing and maintaining the presenting problem. It is here

209
210 CHAPTER 6

that the NED PI-R can be quite useful. It can generate tremendous
amounts of information about the client's characteristic ways of living that
can help both inform and direct clinical querying and intervening. Thus,
the energy expended in obtaining personality assessment data is rewarded
when treatment moves into its middle and end phases. This chapter pro-
vides some strategies and materials to use in meeting these challenges.
This chapter also outlines some of the ways that the NEO PI-R can be
used as a research tool in areas such as cross-cultural psychology, scale val-
idation, and program evaluation. Given the strong psychometric founda-
tions of the instrument, it has a significant contribution to make in
expanding our knowledge base in any area that concerns personality
structure and development. Perhaps the two greatest potential contribu-
tions of this relatively comprehensive taxonomy is its ability to clarify the
personological content of existing constructs and to point out ways of
identifying new domains of individual-difference variables. This chapter
outlines some of the techniques and strategies for using the NEO PI-R to
realize these goals.

THE NEED AND ROLE OF RESEARCH

OVERVIEW

Today clinicians are facing great pressures from third-party payers to

provide more effective services, in less time, and with lower cost. Many
practitioners greet this new mental health-care landscape with frustration
and nervousness. The seemingly rigid treatment standards call for inter-
ventions that do not always seem appropriate to the needs of the client.
These pressures are augmented by both the growing consumer movement
interested in having practitioners demonstrate some value for their ser-
vices (Lyons, Howard, O'Mahoney, & Lish, 1997), and the maturing ethical
dimensions of practice that call for therapists to explore and document the
effectiveness of their treatments (Ogles, Lambert, & Masters, 1996).
But behind this seemingly black cloud lies an important silver lining:
the need for empirical documentation of clinical outcomES. Underlying all
these pressures on practice is the need for clinicians to provide greater ac-
countability for the effectiveness of their interventions, both in terms of
client satisfaction and more objective indices of relief from symptomatol-
ogy. Responding to this need requires clinicians to collect data to support
the value of their services. Although this may appear to be cumbersome
and inconvenient in a milieu that is already overburdened with paper-
work and time commitments, the fact is this need for accountability will
 RESEARCH APPLICATIONS 211

not just go away. It provides a real impetus for the social sciences to begin
to put in place a paradigm for the collection of information that can speak
meaningfully to the quality of our interventions and the conditions under
which they can be expected to work. All of this interest in creating greater
therapist accountability has a singular focus on obtaining outcome assess-
mentdata.
Outcome assessment data speak to the very heart of practice, in that
they provide information about how effective our treatments are, and with
what type of clients. Collecting relevant clinical outcome data will (a) help
improve the quality and efficacy of treatment, (b) enhance the empirical
foundation of the clinical sciences, and (c) provide greater accountability
(Barlow, Hayes, & Nelson, 1984). With so many types of professionals
vying for a slice of the mental health pie the emphasis on empirical data
will only grow, enabling those with the data to claim the largest share.
Thus the pressure will increase for clinicians to create a clinical envi-
ronment that includes a significant assessment component. There are
numerous works currently available that outline how relatively easy out-
come assessments can be performed (e.g., Lyons et al., 1997; Ogles et al.,
1996; Stout, 1997). All of these works stress the need for more information
on treatment efficacy. Research has to be seen as our friend. It alone will be
able to stave off the onslaught of managed care and its ubiquitous call for
more efficiency. Only data will be able to speak persuasively about how
much treatment various "problems" will need to receive, and how much
improvement can be expected with different kinds of clients.
This need for more outcome data must be seen as an irresistible force
in our field. This information must be collected and someone will have to
do it; either managed care will collect its own data or we will need to do it
ourselves. If it is us, social scientists, both clinicians and researchers, who
undertake this research challenge, then we will receive an added bonus for
our efforts: We will have the ultimate control of how we define ourselves
professionally. It is the selection of variables and outcomes that will have
the largest influence on marking out the territory we as social scientists call
our own, the constructs we find of interest, and the goals we pursue with
our interventions. It would be to our advantage if we make these selec-
tions rather than having some outside agency impose them on us.
The basic paradigm for conducting outcome research is really quite
straightforward. It may consume some additional time of the client and
therapist, but its payoff is certainly worth the investment. The NEO PI-R
can playa pivotal role in this process. The dimensions assessed speak
meaningfully about the motivations and capacities of individuals. As was
highlighted in Chapter 2, the NEO PI-R facets and domains are very
closely linked to a wide array of clinical dimensions and outcomes. Scores
212 CHAPTER 6

on the NEO PI-R can be used to match treatments to clients, to make dif-
ferential diagnoses, and to formulate expectations of treatment outcome,
to name a few of its potential benefits. But this is only the beginning.

How CLINICIANS CAN PROVIDE IMPORTANT INSIGHTS

Clinicians are in an ideal position to collect important clinical informa-
tion. Dealing directly with the consumers of their services, clinicians can
collect assessments at ideal and timely moments. Although the choice of in-
struments to use is quite large, many of the basic tools for assessing out-
come are largely context-specific questionnaires that are tailored to reflect
the particular services that were rendered. The last section in this chapter
provides some examples of forms that I have used in my own outcome
evaluation studies. They are included to provide a template for those inter-
ested in performing their own work. Readers should feel free to modify and
adapt these questionnaires to meet the needs of their own situation.
Several developments have made the current field ready for these
types of investigations: first, the presence of a wide range of psychometri-
cally useful clinical instruments that are relatively simple to administer and
score; second, the availability of high-speed computers that can easily ana-
lyze data sets and support scoring programs to interpret clinical data; and,
finally, the increasing sophistication of the social sciences themselves. We
are building better theoretical models and employing more mature empir-
ical designs to test them. As a consequence, the field is moving toward a
new level of development, one that is far more data driven than at any time
in its past. Thus the question is no longer whether we need to obtain clinical
outcome data, but rather when and how this information will be obtained.
In order to fully enjoy the fruits that outcome research can provide for
us, we need to become familiar with the many issues that emerge in doing
research with the NEO PI-R. The sections that follow outline several of
these, including identifying the number of factors in a data set, evaluating
the generalizability of the five-factor structure in new samples, and estab-
lishing the validity of new and existing constructs. In writing this chapter
I tried to balance the need to provide more technical information with the
reality that many readers may be relatively unfamiliar with the quantita-
tive bases of these techniques. The information presented here provides a
blueprint to follow in using the NEO PI-R in its most empirically advan-
tageous way.
Ultimately, though, clinicians and researchers will need to forge a new
alliance that will allow for the collection of data that will help (re)establish
the value of the social sciences and to document empirically the efficacy of
our interventions. We need to do this in a way that maintains the integrity
 RESEARCH APPLICATIONS 213

of our discipline and brings forth our own sense of identity, rather than
having it imposed on us from outside sources. Finally, our ability to influ-
ence managed care and its reimbursement strategies will be entirely de-
pendent on the type and quality of empirically based information that we
can bring to bear on the issues of treatment and outcome. Because we are
at the applied frontier of our field, we need to take a more active and in-
volved role in this process.

ISSUES RELATING TO THE FACTOR ANALYSIS

OF NEO PI-R INFORMATION

The factor structure of the NEO PI-R is well established empirically in

the research literature, both for the English version (Costa & McCrae,
1992c; Costa, McCrae, & Dye, 1991; McCrae & Costa, 1987; Piedmont,
1994) and for foreign language versions (Capara, Barbaranelli, & Comrey,
1995; Katigbak, Church, & Akamine, 1996; Piedmont & Chae, 1997; Silva et
al., 1994; Vassend & Skrondal, 1995). Nonetheless, there remain some tech-
nical debates reverberating in the literature relating to factor analytically
derived information from the NEO PI-R. This section provides an
overview of the technical issues underlying these controversies and the re-
sponses that have been made to address them. The ultimate aim of this sec-
tion, however, is to provide the reader with guidelines for both selecting
and using factor analytic techniques with NEO PI-R data. It should be re-
membered that the best analytic endeavors are those that have a clear the-
oretical framework for guiding the investigations. This section addresses
two of the more basic questions surrounding the NEO PI-R: how many
factors really exist in this instrument and what is their nature?

FACTOR ORTHOGONALITY

Controversy continues to center on whether or not the five factors are

independent of one another as claimed. On the one hand, orthogonality is
a preferred quality of the data because it assures very little redundancy in
content. Information obtained from one factor has nothing in common
with the remaining dimensions. This assures efficacy in assessment and
enables one to build useful prediction models. In response to this, some
argue that such technical neatness is rarely observed in nature (e.g., J.
Block, 1995). As Loevinger (1994) has stated, "There is no reason to believe
that the bedrock of personality is a set of orthogonal (independent) factors,
unless you think that nature is constrained to present us a world in rows
and columns. That would be convenient for many purposes, particularly
214 CHAPTER 6

given the statistical programs already installed on our computers. But is it

realistic?" (p. 6).
Spirrison (1994) responded that the question of orthogonality is irrel-
evant. The five-factor model represents an abstraction of what we refer to
as personality. Whether this abstraction provides a faithful reproduction of
a real entity is another question. What is important is whether this ab-
straction has any empirical utility, "The question is whether the rows and
columns of orthogonal solutions provide an effective framework for orga-
nizing our perceptions of personality" (p. 582). From this perspective,
given that personality is in and of itself an abstraction developed by psy-
chologists to understand human behavior rather than a reflection of some
real, underlying psychic reality, then the "neatness" of the model should
not be grounds for dismissing it. Rather, its value should be determined by
its heuristic and predictive utility.
However, given the empirical basis of both the FFM and the NEO
PI-R, it seems necessary that any approach to determining factor orthog-
onality take an empirical route. Fortunately, whether the factors are corre-
lated can be empirically determined. If the five factors are really correlated,
then creating scales that contain that overlap should be better predictors of
external criteria than scales constrained to be independent. This is because
the independent factors are losing information and their construct validity
will suffer. McCrae and Costa (1989b) demonstrated that when the five fac-
tors are treated as independent entities, the factors' convergent correla-
tions with external criteria were higher than when the factors were
allowed to correlate. Further, patterns of discriminant validity were also
lower for the orthogonal factors.
But the persistent question is "Why do we continue to find significant
correlations between the various five factors?" If the underlying structure
is in fact orthogonal, why is that not witnessed in real data? There are a
number of studies, usually confirmatory in nature, that show substantial
overlap among factors (e.g., Mooradian & Nezlek, 1996; Vassend & Skron-
dal, 1995). I believe one possible reason for this observed overlap is re-
searchers' reliance on only a single source of information for obtaining the
personality ratings: self-reports. Consequently, the observed interfactor
correlations may be the result of correlated method error. If this potential
source of error can be controlled for, then one would expect to find that the
correlations among the five domains would become zero.
This hypothesis can be tested by obtaining personality ratings from
multiple information sources. I evaluated this hypothesis using a sample
of 178 undergraduate and graduate students who completed the Multi-
dimensional Personality Questionnaire (MPQ; Tellegen, 1982) and a NEO
PI-R for themselves as well as having two people rate them on the ob-
 RESEARCH APPLICATIONS 215

server form of the NEO PI-R (see Piedmont, 1994, for a description of the
sample and methods). The purpose of these analyses was to evaluate the
overlap between the Neuroticism and Extraversion domains. As was out-
lined earlier, these two factors are strongly related to experienced Well-
Being and there continues to be some controversy over the relationship
between these two domains. Some argue that Positive and Negative Affect
(Extraversion and Neuroticism, respectively) are correlated (e.g., Chen,
Dai, Spector, & Jex, 1997; Green, Goldman, & Salovey, 1993), while others
believe the two to be truly orthogonal (e.g., Goldstein & Strube, 1994).
The MPQ contains two global domains of interest here, Positive Affect
and Negative Affect. I tested two models using LISREL 8 (Joreskog & Sor-
bom, 1993). The first model is presented in Figure 6-1, and outlines two
latent dimensions, Negative Affect (NEG-AFF) and Positive Affect (POS-
AFF). The former was defined by self-rated measures of Neuroticism
(SELFN) and Negative Affect (NEGAFF) as well as by observer-rated Neu-
roticism (RATEN). The latter latent dimension was defined by self-rated Ex-
traversion (SELFE) and Positive Affect (POSAFF) and observer-rated
Extraversion (RATEE). As can be seen, the estimated correlation between
the two latent dimensions was -.31, suggesting that Positive and Negative
Affect may reflect a singular bipolar dimension rather than two distinct do-
mains. The overall model chi square with 11 degrees of freedom was 25.47
(p = .0078), suggesting a relatively poor model fit, although the Normed Fit
Index (NFI), Goodness of Fit Index (GFI) and Adjusted Goodness of Fit
Index (AGFI) were all over .90, indicating that the model does match the
data well. This apparent inconsistency is a regular problem noted in confir-
matory analyses using personality-based data and is considered later.
A second model was also tested and it is presented in Figure 6-2. Here
two additional latent dimensions were included, one that represented self-
report variance and the other representing observer-rating variance. These
dimensions were added to represent the impact of information source on
scores. The overall model chi square with 4 degrees of freedom was 2.27
(p = .69) indicating a very good fit of the model to the data. Secondary fit
indices were all uniformly close to unity (NFl = .99; GFI = 1.0; AGFI = .98)
This model also fits the data significantly better than the previous one
(Ll = X2 = 23.2, P < .01).
Two conclusions can be drawn from these data. First, by including the
dimensions representing the information source for the ratings (e.g., self
versus observer), we obtain a model that fits the data much better than
when these dimensions are not included (model chi squares of 2.27 versus
25.47, respectively). Second, the estimated correlation between the latent
personality domains drops to a nonsignificant -.18, supporting the con-
tention that these domains are independent. Thus, the observed covari-
216 CHAPTER 6

71.176. .
--1 ~.n SELFN

m.-1 NEGAFF rU

l
~71

n., r'·
RATEE

SELFE
~M
J
u

m.u4 POSAFF ~
FIGURE 6-1. Estimated Confirmatory Factor Analysis Parameters for the Personality Only
Model.

ance among the personality scales noted in these data can be attributed to
correlated method error arising from a single information source (note that
the correlation between rating sources is r =.75 which attests to the cross-
observer validity of these personality dimensions). Thus, two personality
scales will correlate both because they share substantive content and be-
cause it is the same person generating the information. Any distortions in-
troduced in one scale by a person will also be included in all other
measures he or she completes. Hence, all these measures will correlate to
some degree.
These data support my contention that observed correlations among
the five personality domains noted in several studies is the result of corre-
lated error introduced by relying on a single information source. By using
multiple raters and including latent dimensions that represent these
 RESEARCH APPLICATIONS 217

n.~ RA~N k~

4.71 SELFN

11.
... ~ HECIAFF K1

n.~,-

...
FIGURE 6-2. Estimated confirmatory factor analysis parameters for the Personality and In-
formation Source Model.

sources of method variance in CFAmodels, it may be possible to remove

this overlap from the estimated correlations among the substantive per-
sonality dimensions.
Thus, it appears that the five factors do represent independent aspects
of personality. In order to capitalize on this reality, Costa and McCrae
(1992c) provide weights for creating factor scores for each of the five fac-
tors. These factor scores are based on varimax rotations of NED PI-R data
from the standardization sample and result in factors that are indeed more
orthogonal than the domain raw scores. The computer scoring program
provides these values in lieu of the raw domain scores. These factors have
also been shown to be better predictors of external criteria because they
rely on information from all 30 facet scales. When conducting research
with the NED PI-R, it is preferable that these values be calculated and
218 CHAPTER 6

included rather than simply relying on values obtained by summing the

facets for each domain.

NUMBER OF FACTORS

The issue of how many factors are needed to explain the full spectrum
of personality continues to be debatable in the field. As noted in Chapter 1,
there are numerous personality models advocating any number of per-
sonality domains. Discussions of the merits of these models is beyond the
scope of this book, but the issue of how many factors there are does raise
two concerns for us. The first issue relates to evaluating various factor
structures and the second centers on one's ability to obtain the appropriate
factor structure from NEO PI-R data.

Evaluating Factor Structures

In prefacing this section, a few comments on factor analysis are nec-
essary. Factor analysis is a data reduction process. Its aim is to reduce a
large set of items into homogeneous clusters of items known as factors. As
is the case with all statistical analyses, such procedures are tools designed
to help investigators obtain insights from a data set. As a tool, factor analy-
sis can never provide "the answer" to a question one may hold of a data
set. Rather, it can only suggest possible interpretations. A statistical test or
analysis should never be seen as providing the final word; it can never do
that. Even with an instrument as empirically sound as the NEO PI-R, fac-
tor analysis can never provide complete certainty for conclusions drawn
from a set of data (see Vassend & Skrondal, 1995).
Also, as a tool, statistical results are only as good as the data they an-
alyze. The term GIGO is appropriate here. It stands for "Garbage In,
Garbage Out." Poor data, no matter how well analyzed or evaluated will
always lead to poor statistical conclusions. Statistical analyses do not
change the fundamental value of data. Instead, they merely try to make as
much sense out of data as possible. In the social sciences, much of the data
collected is weak. Samples are rarely randomly selected, the subjects to
variables ratios are frequently low, distributions are not truly normal, and
the ordinal nature of the data analyzed (to name a few problems) will al-
ways play havoc with the data analyses to an unknown degree. In terms of
factor analysis, these problems manifest themselves through the non-
replicability of factor structure across different samples, the emergence of
method or error factors, and the encountering of anomalous statistical dif-
ficulties (e.g., negative variances, singular matrices). Therefore, great care
needs to be taken in the interpretation of the results from factor analyses.
Perhaps the most important criteria for evaluating the worth of any factor
 RESEARCH APPLICATIONS 219

analysis is the degree to which the obtained structure can be replicated

across different samples. Other than that, caveat emptor.
With that said, it should be realized that there are researchers who be-
lieve that the five-factor model specifies either too many factors or too few.
For example, Eysenck believed that there are only three factors: Neuroti-
cism, Extraversion, and Psychoticism (Eysenck, 1991; Eysenck & Eysenck,
1985). Neuroticism and Extraversion correspond well to their five-factor
model equivalents; however, Psychoticism is seen as a "superfactor"; it
combines Agreeableness and Conscientiousness into one domain. Open-
ness to Experience is not represented in this model. Still others believe that
there should be more than five factors. For example, Jackson, Ashton, and
Tomes (1996) have suggested that there are six factors. They found evi-
dence for N, E, a, and A. Conscientiousness, however, was found to be
broken into two specialized dimensions, Industriousness and Methodical-
ness. Tellegen and Waller (in press) believe that there are seven dimen-
sions: the already acknowledged five plus Positive and Negative Valence
(see also Almagor, Tellegen, & Waller, 1995; Benet & Waller, 1995). These
additional two factors represent more of the evaluative nature of person-
ality assessment. Unlike the adjectives that were used to generate the five-
factor model, the terms subsumed by these new factors represent qualities
that were originally considered unsuitable for analysis (e.g., excellent, im-
pressive, wicked, cruel).
All of these different personality models claim support for them-
selves, and do so in relation to the five-factor model. Frequently support-
ing data is provided that includes either the NEO PI-R or markers for the
five factors. How can such models exist given the very strong empirical
basis of the five-factor model? Is the five-factor model incorrect in its spec-
ification of personality dimensions? These are important questions that
need to be answered. But it is reassuring that these questions are empiri-
cal questions. In other words, data can be used to settle the issues that
these studies raise. The way to analyze data to answer these kinds of ques-
tions is at the heart of this section.

When Five Factors Are Not Enough

In evaluating whether there are more than five factors, the first test of
the larger factor set's viability should be replication: Can the factor struc-
ture be obtained in multiple samples? If the kinds of statistical artifacts
mentioned earlier (e.g., low subjects to variables ratio, non-normal distrib-
utions) are operating in a data set to create additional factors, these forces
will be much less influential or entirely absent in another data set contain-
ing the same variables. Thus, obtaining a similar factor structure in a dif-
ferent sample will be very difficult.
220 CHAPTER 6

Factor comparability is usually obtained through the use of a congru-

ence coefficient (see Gorsuch, 1983, for how to calculate this value and for
other comparability indices). Acongruence coefficient is a value between 0
and 1 that indexes the degree to which the pattern and magnitude of the
factor loadings from two analyses that include the same variables mea-
sured over different samples are equivalent. Usually a value of .90 or
greater is considered high enough to warrant the conclusion that two so-
lutions are identical (i.e., the obtained factors represent equivalent enti-
ties). However, there are other methods for determining how high a
congruence coefficient needs to be (see McCrae, Zonderman, et al., 1996).
If a replicated factor structure can be obtained, such an event, in and
of itself, does not call into question the comprehensiveness of the five-fac-
tor model. McCrae and Costa (1995) outlined three criteria that need to be
met before these additional factors can be accepted as new dimensions.
First, the new factor or factors must be shown to be independent of the
original five. Second, the new factor or factors should be of the same gen-
eral breadth as the other five; the new factor should encompass several
other, more specific qualities in its range of convenience in a way similar to
the way each domain of the NEO PI-R contains six facet scales. Third, the
new factor should be recoverable from multiple sources of information.
The qualities of personality represented in this new factor should also be
recoverable from other established personality instruments. Further, these
new qualities should be able to be consensually validated, being found in
both self-reports and observer ratings. When these criteria are applied,
many studies claiming more than five personality domains fail (see Mc-
Crae & Costa, 1995; Piedmont, McCrae, & Costa, 1991).
These criteria distinguish between a true, broad-based dimension of
personality and more specific, unique personality descriptions. As noted
in Chapter I, there are qualities of personality that are reliably obtained,
but reflect highly circumscribed aspects of personality (e.g., Goldberg,
1990). One needs to be careful not to overvalue additional, circumscribed
factors that may be encountered in a data set.

When Five Factors Are Too Many

The problem of fewer than five factors provides another set of issues
to be considered when working factor analytically with the NEO PI-R. As
noted earlier, Eysenck has argued that there are only three factors that de-
scribe personality; Neuroticism (N), Extraversion (E), and Psychoticism
(P). Clearly, in a system such as this the smaller model is either leaving out
dimensions of personality or is combining some of the NEO PI-R domains
into a single, more global dimension. If the smaller model is leaving out in-
 RESEARCH APPLICATIONS 221

formation that is contained in the NEO PI-R, then usually one can be con-
fident that the smaller model is overlooking important aspects of person-
ality. If it is combining domains, then some empirical considerations need
to be evaluated to determine whether the smaller model is to be preferred
due to its greater parsimony or whether the smaller model is inappropri-
ately combining independent dimensions.
For the three-factor model advocated by Eysenck, both issues emerge.
First, Eysenck's model leaves out the dimension of Openness to Experience.
However, Eysenck (1991) has suggested that Openness is really an aspect of
cognitive ability rather than personality. Although there is data showing
that Openness is not an aspect of cognitive functioning (McCrae, 1993-1994,
1994b), let us finesse this point for now by granting the issue to Eysenck
(see Costa & McCrae, 1995b, for a response). This still leaves us with the
need to evaluate the worth of a system with fewer than five factors.
Several studies have shown that Eysenck's dimension of Psychoticism
is inversely related to the domains of Agreeableness and Conscientious-
ness (e.g., Costa et a/., 1991; Goldberg & Rosolack, 1994) and Eysenck
(1992) has interpreted this to mean that these domains are merely facets of
the larger Psychoticism dimension. However, Costa and McCrae (1992b,
1995b) have argued that Psychoticism represents "a relatively arbitrary
conflation of two independent dimensions" (1995b, p. 310). As noted ear-
lier in this chapter, the question as to whether the five factors should be
considered orthogonal or correlated has important conceptual implica-
tions. If the five factors are correlated, then this leaves open the possibility
of obtaining "high-order factors," super dimensions that are comprised of
two or more of the five. This is what Eysenck claimed. However, if the five
domains are orthogonal, then no such super factors can exist (an identity
matrix cannot be factored) and Costa and McCrae would be correct. Al-
though the data suggest that orthogonal factors have better predictive va-
lidities than oblique (correlated) factors, there are other empirical criteria
that can be applied for evaluating the internal quality of a factor structure.
One important way of clarifying this issue is to evaluate the pattern of
external correlations these elements have. One of the important aspects of
factor analysis is that it identifies homogeneous groupings of items. As in
the NEO PI-R, the six facet scales are more specific articulations of the
larger domain to which they belong. Thus, these scales all correlate in a sim-
ilar way with relevant criteria. For example, all six Neuroticism facets are
related to low psychological well-being (e.g., Costa & McCrae, 1984). There-
fore, aggregating across the facets produces domain scores that are more re-
liable and stable than their constituent elements. If Agreeableness and
Conscientiousness are merely subfacets to Psychoticism, then they should
reveal a similar pattern of correlates to a given criterion and summing these
222 CHAPTER 6

two dimensions together should produce an index that is more strongly as-
sociated with this criterion than either of the domains separately.
Costa and McCrae (1995b) used this strategy in evaluating the relations
between the NEO PI-R and the Eysenck Personality Profiler (EPP). Corre-
lations between Agreeableness and Conscientious and the EPP scales
showed very different patterns of correlates for the two NEO PI-R domains.
Further, when Agreeableness and Conscientiousness were aggregated and
then correlated to the EPP scales, the resulting correlations were lower than
their individual associations. Thus, the Psychoticism dimension appears to
inappropriately combine independent personality qualities. This forced ag-
gregation ultimately impairs external validity, which should be regarded as
an important criterion for evaluating the utility of any personality model.
What should be taken from this discussion is that empirical methods
can and do generate a wide variety of statistical models for describing per-
sonality systems. Any review of the current literature will reveal numerous
taxonomic models ranging from 3 to as many as 16 factors. Often, the
quantitative basis of these models is quite compelling. This can leave one
quite confused about the state of affairs in personality measurement. How
many factors are there? But with a little care and patience, the discerning
reader can begin to ask the appropriate, sophisticated questions that can
provide an answer to that question. Given that the five-factor model has
been shown to be a reliable description of personality variables, it can and
should be the reference point for any critical evaluation of other models.
In the case of a model advocating more than five personality dimen-
sions, the questions that need to be asked include whether the factor struc-
ture can be replicated in a new sample, the independence of the new
factors from the established five, and the personological breadth of the
new domains. When a model advocates fewer than five factors, it needs to
be determined whether the model omits something already contained in
the five-factor model. Such an omission would be sufficient evidence to re-
ject the new model. However, if the model seems to combine two or more
of the Big Five's domains, then some evaluation of the predictive utility of
this global factor needs to be conducted. Does the "super factor" actually
evidence higher external validity than its components? If not, then the new
model does not provide an optimal specification of personality structure.

RECOVERING THE NEO PI-R STRUCTURE IN NEW SAMPLES

The NEO PI-R is extensively used in a wide array of applications, in-
cluding industrial and organizational, psychotherapeutic, medical, and
cross-cultural contexts. In each of these environments issues emerge con-
cerning the psychometric properties of the instrument with these new
 RESEARCH APPLICATIONS 223

populations. This is especially true when the scale is used internationally.

An important question that emerges concerns the cross-cultural general-
izability of the five-factor model: Is it as useful a description of personal-
ity when employed with individuals who do not have the same lexical and
historical heritage as the West? Do the basic concepts of the five-factor
model remain relevant in new cultures? Answers to these questions have
important implications for those who wish to construct a unified theory of
personality structure and development.
Empirically, answers to these questions involve a confirmatory data
analytic approach. In these studies, data are collected on the NEO PI-R in
a new sample and the questions are whether the five-factor structure can
be recovered and if the obtained structure is similar to normative values.
Thus, a set of observed values are compared to an a priori defined numer-
ical model. The closer the obtained values come to the ideal values, the
more likely that the data will be regarded as "fitting" the model. Programs
such as LISREL (Joreskog & Sorbom, 1993) are designed to mathematically
evaluate this kind of model fit. The value of this approach is that one must
specify some theoretical model and the relations among the respective el-
ements prior to conducting the research. This is a much more sophisticated
and powerful approach to data analysis than the usual post hoc, ex-
ploratory methods commonly employed. Exploratory methods capitalize
heavily on chance and sample specific factors that result in poor replica-
bility of findings.
Confirmatory Factor Analysis (CFA) has been used quite frequently in
NEO PI-R research, and has generated much controversy. Of particular in-
terest has been the apparent failure to find the anticipated five-factor struc-
ture. Borkenau and Ostendorf (1990) conducted the first CFAof NEO PI-R
data and found that the observed data did not fit the expected model very
well. Parker, Bagby, and Surnmerfeldt (1993) performed a CFAof the original
NEO PI-R normative data. They tested several models, including complex
ones that specified secondary loadings among the facets, and found that
none of their models provided satisfactory fit. In a later study by Panter,
Tanaka, and Hoyle (1994) it was found that the five factors were by no means
mutually independent. That these findings are at variance with other empir-
ical evidence that shows a clear, replicable factor structure raises two possi-
bilities. The first is that the five-factor model does not have a robust structure.
Studies finding this structure may have capitalized on sample specific vari-
ance, or the five-factor model may be a limited phenomenon, emerging only
in certain segments of our society (e.g., college students). The second possi-
bility is that CFA procedures may not be appropriate for use with personal-
ity data. Specifically, CFAmay require a model to explain more variance in
the data than is possible given the reliabilities of the measures used.
224 CHAPTER 6

In order to evaluate these two possibilities, Burke and his colleagues

(Church & Burke, 1994; Katigbak et al., 1996) systematically evaluated both
the five-factor structure in a cross-cultural sample (Filipinos) and CFA pro-
cedures relative to exploratory analyses. The first result of note in these
studies was that a five-factor model did emerge from NEO PI-R data when
exploratory factor analysis (EFA) was applied. This model was also found
to cross-validate strongly in another sample. Thus the data clearly showed
not only that the five factors are present but that their relevance emerged
in a sample very different from the original derivation sample. Once this
replicable structure was obtained, the data (and model) were subjected to
CFA to determine fit. Given the robustness of the model in these different
samples, it would not have been surprising to find convergence with the
CFAresults. However, the CFAresults were repeatedly disappOinting. As
Katigbak and colleagues (1996) noted,
In theory, confirmatory factor analysis would seem to be the ideal method
for testing the cross-cultural generalizability of a personality model. How-
ever, an increasing number of studies ... indicate that both its significance
tests and conventional goodness-of-fit criteria may be too stringent. ... In-
deed, we found here that exploratory factor solutions that are highly con-
gruent across samples ... were judged to have at best only a fair fit when
applied in these different (cross-validation) samples. (p. 111)

McCrae, Zonderman, et al. (1996) believed that the assumptions underly-

ing CFAmay not be appropriate to real personality-type data.
At the heart of CFA analyses is the chi square test for model fit. This
is a major advantage to the approach because it allows one to evaluate how
well observed data fit some a priori model. However, problems with the chi
square test have been known for years (e.g., Bentler & Bonnett, 1980;
Marsh, Balla, & McDonald, 1988). The most problematic issue concerns the
sensitivity of the statistic to sample size. Given a large enough sample, any
model will be rejected. As a result numerous other fit indexes have been
developed to address this issue, but their interpretation and statistical
properties are not fully understood. Consequently they do not have ac-
companying significance tests and this leaves their practical value in ques-
tion. As noted earlier, a common finding in CFAanalysis using personality
data is that an overall chi-square fit statistic is significant (indicating that
the model does not fit the data) yet secondary fit indexes show that there
is a high level of congruence. This type of outcome is awkward to manage
both statistically and interpretively.
Although CFA may not be the most appropriate method for evaluat-
ing how well a factor structure resembles the underlying five-factor model,
users of the NEO PI-R still need to have some confirmatory-based ap-
 RESEARCH APPLICATIONS 225

proach for evaluating structure in new samples. EFA methods spring to

mind, but they have their own limitations. First, they provide solutions
that are sample specific. Given the reality that research samples rarely are
representative of a population, it is possible that limited, nomeplicable fac-
tors could emerge in a specific data set. Therefore, without a clear statisti-
cal model driving the analysis itself, it becomes difficult for researchers to
discern the merits among a variety of potentially acceptable solutions.
One analytic strategy, using EFA techniques, could be to extract five
factors from a NEO PI-R data set in a new sample and to compare them
statistically with normative values presented in the manual. Such a com-
parison would use congruence coefficients (Gorsuch, 1983) to assess the
actual degree of comparability between the two sets of structures. Al-
though Katigbak and colleagues (1996) saw this approach as the preferred
strategy for those using personality data, they point out a disadvantage:
Sample specific variation in the covariance among NEO PI-R facets may
overstate differences between the observed sample and normative values.
This is because the NEO PI-R does not evidence true simple structure.
Many of the facet scales have significant secondary loadings on other fac-
tors. These cross-factor loadings make sense theoretically and have been
replicated numerous times (e.g., Church & Burke, 1994). For example,
Angry Hostility, a facet of Neuroticism, reflects one's tendency to experi-
ence anger. However, individuals low on Agreeableness also tend to expe-
rience and express anger directly; thus this facet loads negatively on
Agreeableness as well. Perhaps the most problematic area concerns the
facets for Extraversion and Agreeableness. These two domains comprise
the interpersonal circumplex, so the facets scales form a more complex, cir-
cular pattern around these two domains. Small, sample specific changes in
these scales' patterns of covariation could lead to very different factor
structures for these two domains.
In order to address this problem, McCrae, Zonderman, et al. (1996)
proposed a methodology for conducting confirmatory analyses using
NEO PI-R data that allows one to determine whether an obtained data set
matches a normative structure. The value of this approach is threefold.
First, it relies on EFA techniques which have shown themselves to be most
straightforward and useful for analyzing personality-type data. Second, it
provides an a priori empirical framework for factor analyzing the data ma-
trix. Finally, it provides useful measures of fit. The proposed methodology
uses regular EFAmethods for extraction of five factors, and then orthogo-
nal Procrustes rotations to obtain final structural fit of the data.
Orthogonal Procrustean rotation is a type of targeted rotation
(Schonemann, 1966). The goal is to rotate an observed set of data into a for-
mat that is determined a priori. The predetermined structure serves as the
226 CHAPTER 6

"target matrix" that the rotation attempts to approximate. The eigenvec-

tors (factor loadings) of the rotated solution are then compared to the nor-
mative (targeted) values through use of congruence coefficients. These
values determine how well the real data represents the model. Usually,
congruence coefficients greater than .90 are considered high enough to
convey adequate "fit." McCrae, Zonderman, et al. (1996) have developed
actual significance values for these statistics.
In order to evaluate the degree of fit, congruence coefficients are cal-
culated for not only the overall five factors, but also for the pattern of load-
ings for each facet scale across the five factors. It is possible for an overall
factor (which is based on the loadings for 30 facet scales) to have accept-
able fit, but individual facets may not show correspondence with their re-
spective normative values. Thus, these congruence coefficients can be used
as "modification" indexes for evaluating where and to what degree a par-
ticular element does not conform to its expected values. Table 6-1 presents
an SPSS Matrix program for performing a Procrustean rotation of NEO
PI-Rdata.
As can be seen, there are two principal data matrices. The norms ma-
trix is already provided. These values are the factor loadings for the NED
PI-R in the normative sample. This matrix will serve as the target matrix
for rotating obtained data. The other matrix, loadings, is for the actual data.
It should be entered as a 30 (rows) by 5 (columns) matrix. The first column
should be the factor loadings for each of the facet scales on Neuroticism,
the next column is for Extraversion, then Openness, Agreeableness, and
Conscientiousness, in that order. The entered loadings can be either ro-
tated or unrotated loadings from the original EFA. The program will auto-
matically print out a table of its results. It will calculate congruence
coefficients for each factor and facet scale, and an overall fit congruence co-
efficient. McCrae, Zonderman, et al. (1996) provided .05 and .01 critical val-
ues for these congruence coefficients for each of these statistics. The 95th
and 99th percentile cutoff values for the domain factor coefficients are .55
and .65, respectively. Similar cutoff values for the facet congruence coeffi-
cients are .86 and .94, respectively. Finally, the critical values for the over-
all congruence coefficient are .42 and .46, respectively.
Orthogonal Procrustes Rotations serve as a useful compromise strat-
egy for doing confirmatory analyses using NED PI-R data. It capitalizes
on the straightforward, familiar techniques of EFA and avoids the many
restrictive assumptions of CFA. Most notably, it provides a method of eval-
uating structural fit in a way that is not overly dependent on sample spe-
cific variance. In other words, whatever distortions may be in the data at
hand, the influence on the final factor structure is minimized because the
data are being rotated to a form determined a priori.
 RESEARCH APPLICATIONS 227

TABLE 6-1. SPSS Program for Conducting an Orthogonal Procrustean Rotation

of NED PI-R Factor Data
matrix.
compute loadings=!
Insert your own principal components derived, orthogonally rotated five factors
Must be a 30 (rows) by 5 (columns) matrix. Follow format of next section.
}.
compute norms=!
.81, .02, -.01, -.01, -.10 ;
.63, -.03, .01, -.48, -.08 ;
.80, -.10, .02, -.03, -.26 ;
.73, -.18, -.09, .04, -.16 ;
.49, .35, .02, -.21, -.32 ;
.70, -.15, -.09, .04, -.38 ;
-.12, .66, .18, .38, .13 ;
-.18, .66, .04, .07, -.03 ;
-.32, ,44, .23, -.32, .32 ;
.04, .54, .16, -.27, ,42 ;
.00, .58, .11, -.38, -.06 ;
-.04, .74, .19, .10, .10 ;
.18, .18, .58, -.14, -.31 ;
.14, .04, .73, .17, .14 ;
.37, ,41, .50, -.01, .12 ;
-.19, .22, .57, .04, -.04 ;
-.15, -.01, .75, -.09, .16 ;
-.13, .08, ,49, -.07, -.15 ;
-.35, .22, .15, .56, .03 ;
-.03, -.15, -.11, .68, .24 ;
-.06, .52, -.05, .55, .27 ;
-.16, -.08, -.00, .77, .01 ;
.19, -.12, -.18, .59, -.08 ;
.04, .27, .13, .62, .00 ;
-.41, .17, .13, .03, .64 ;
-.04, .06, -.19, .01, .70 ;
-.20, -.04, .01, .29, .68 ;
-.09, .23, .15, -.13, .74 ;
-.33, .17, -.08, .06, .75 ;
-.23, -.28, -.04, .22, .57} .
compute s=t(loadings)·norms.
compute wl=s·t(s).
compute vl=t(s)·s.
call eigen(wl,w,evalwl).
call eigen(vl,v,evalvl).
compute o=t(w)·s·v.
compute ql=o &/abs(o).
compute kl=diag(ql).
compute k=mdiag(kl).
(continued)
228 CHAPTER 6

TABLE 6-1. (Continued)

compute ww=w*k.
compute U=ww*t(v).
compute procrust=loadings*t1.
compute cmlm2=t(procrust)*norms.
compute ca=diag(cmlm2).
compute csum2ml=cssq(procrust).
compute csum2m2=cssq(norms).
compute csqrtl1=sqrt(csum2ml).
compute csqrt12=sqrt(csum2m2).
compute cb=t(csqrtll)"csqrt12.
compute cc=diag(cb).
compute cd=ca&/cc.
compute faccongc=t(cd).
compute rmlm2=procrust"'t(norms).
compute ra=diag(rmlm2).
compute rsum2m 1=rssq(procrust).
compute rsum2m2=rssq(norms).
compute rsqrtll =sqrt(rsum2m1).
compute rsqrtI2=sqrt(rsum2m2).
compute rb=rsqrtll"t(rsqrt12).
compute rc=diag(rb).
compute faccongr=ra&/rc.
compute top={ca;ra}.
compute bot={cc;rc}.
compute ctop=csum(top).
compute cbot=csum(bot}.
compute total=ctop I cbot.
compute . procrust={procrust,faccongr;faccongc,total}.
print procrust ltitle = "FACTOR CONGRUENCE COEFFICIENTS"
/format fS.2
Iclabels= "NEURO" "EXTRA" "OPEN" "AGREE" "CONSC" "ITEMCONG"
/RLABELS= "Nl" "N2" "N3" "N4" "NS" "N6" "El" "E2" "E3" "E4"
liES" "E6" 1101" "02" "03" "04" "05" "06" "Al" IIA2"
II A3" "A4" "AS" II A6" "e1" "C2" IIC3" IIC4" lies" "C6"
"FACTCONG"
/ space=newpage.
END MATRIX.

An example of this process may help highlight the methodology. The

following data are taken from Piedmont and Chae (1997), who were inter-
ested in developing a Korean version of the NEO PI-R. The emphasis was
on determining whether the measure in its new form could be considered
structurally equivalent to its English relative. A translated version was given
to 653 native Koreans and the resulting information was factor analyzed
 RESEARCH APPLICATIONS 229

using EFA techniques. Five factors were extracted and rotated and the re-
sults of this analysis are presented in Table 6-2. Congruence coefficients were
calculated between this solution and factor loadings presented in the NEO
PI-R manual (Costa & McCrae, 1992c). Several findings of interest emerge.
First, the overall congruence coefficient of .82 is significant, suggest-
ing that the general structure is comparable with normative values. In fact,
an examination of each of the five factor congruence coefficients (located at
the bottom of the columns) suggests that all five factors are recovered in a
form comparable to the normative structure. The domains of Neuroticism,
Openness, and Conscientiousness more strongly resemble their English
counterparts than do Extraversion and Agreeableness. However, an ex-
amination of the facet congruence coefficients indicates that none of the
facet scales for the latter two domains is well replicated. The insignificant
coefficients tell us that the pattern of loadings for these scales across the
five domains is different from that found in the normative structure.
Clearly, in using EFA there is a failure to replicate two of the five domains.
But there are several reasons for this apparent failure. One reason is
that the five factors do not operate in a way that was observed in America.
Given the cross-cultural nature of this sample, this different factor structure
may have something important to say about Korean culture and how per-
sonality is expressed through it. A second explanation deals with the fact
that the NEO PI-R does not have perfect simple structure and that subtle
shifts in the facet scales' interrelatedness could impact overall structure.
Given, as noted earlier, that the Extraversion and Agreeableness domains
constitute the interpersonal circumplex, a complex ordering of traits that is
designed to combine various quantities of these two dimensions, this sec-
ond explanation may be likely. The only way to determine which of these
two hypotheses is correct is to conduct an Orthogonal Procrustes rotation
of the results. If such an analysis still could not recover the five domains in
a way consistent with normative data, then this would be strong evidence
that personality may indeed be organized differently in Korean culture.
However, if the normative structure could be better approximated, then the
observed distortions most likely are due to sample specific factors.
These data were then analyzed using the statistical program pre-
sented in Table 6-1. The result of this targeted rotation procedure was the
factor structure presented in Table 6-3. As can be seen, the obtained factor
structure more closely approximates normative values. The overall con-
gruence coefficient is much higher than previously found (.96 versus .82).
The congruence coefficients for the five factors are all quite high, including
the Extraversion and Agreeableness domains, whose values have risen
to .95. Finally, the facet congruence coefficients also show significant
 230 CHAPTER 6

TABLE 6-2 .. Principal Components Analysis Using a Varimax Rotation of the

Korean NEO PI-R Facet Scales.
Varimax factor loadings
Korean Factor Factor Factor Factor Factor Facet
NEO PI-R Facets 1 2 3 4 5 congruence
Nl: Anxiety .81 .02 .02 -.02· -.14 .99"
N2: Angry Hostility .65 .35 .05 -.25 -.16 .84
N3: Depression .82 -.07 .09 -.11 -.19 .99"
N4: Self-Consciousness .81 -.06 -.03 -.10 -.06 .96"
N5: Impulsiveness .52 .36 .15 -.05 -.49 .95"
N6: Vulnerability .71 -.11 -.12 -.02 -.42 .99"

El: Warmth -.05 .20 .19 .79 -.08 .68

E2: Gregariousness -.15 .40 -.14 .63 -.12 .62
E3: Assertiveness -.31 .63 .08 .09 .22 .79
E4: Activity .13 .63 .08 .20 .23 .74
E5: Excitement Seeking .09 .61 .18 .19 -.18 .64
E6: Positive Emotions -.26 .35 .41 .44 -.08 .66

01: Fantasy .21 .10 .53 .01 -.32 .97'

02: Aesthetics .09 .00 .72 .19 .01 .98"
03: Feelings .17 .24 .70 .13 .11 .89"
04: Actions -.27 .31 .37 .08 -.24 .88a
as: Ideas -.10 .21 .72 -.05 .14 .96"
06: Values -.27 -.18 .51 .10 -.09 .83

AI: Trust -.32 -.21 .09 .60 .24 .78

A2: Straightforwardness -.16 -.52 .07 .19 .33 .57
A3: Altruism -.08 .34 .14 .68 .22 .39
A4: Compliance -.09 -.71 -.03 .21 -.04 .39
AS: Modesty .23 -.46 -.25 .15 -.01 .60
A6: Tendermindedness .11 -.29 .36 .48 .17 .56

Cl: Competence -.48 .18 .13 .07 .54 .99"

C2: Order -.09 .06 -.13 -.02 .70 .99"
C3: Dutifulness -.04 -.12 -.03 .23 .80 .96b
C4: Achievement -.07 .42 .07 -.02 .67 .95b
C5: Self-Discipline -.43 .04 .01 .04 .71 .97"
C6: Deliberation -.25 -.23 .05 -.04 .64 .92a

Factor Congruence .97' .60" .93b .61 a .95' .82'

N = 653. Loadings above I .40 I are in bold.
'Congruence higher than that of 95% of rotations from random data.
'Congruence higher than that of 99% of rotations from random data McCrae, Zonderman, el al.
Nole. From "Cross-Cultural Generalizability of the Five-Factor Model of Personality: Development and
Validation of the NEO PI-R for Koreans," by R. L. Piedmont and J. H. Chae, 1997, Journal ofCross-Cul!ural
Psychology, 28, p. 141. Copyright 1997 by Western Washington University. Adapted with permission.
 RESEARCH APPLICATIONS 231

TABLE 6.3. Principal Components Analysis Using an Orthogonal Procrustes

Rotation of the Korean NEO PI-R Facet Scales
Procrustean factor loadings
Korean Factor Factor Factor Factor Factor Facet
NEO PI-R Facets 1 2 3 4 5 congruence
Nl: Anxiety .81 .01 -.06 -.03 -.15 .99&
N2: Angry Hostility .66 .03 .03 -.43 -.10 .99&
N3: Depression .82 -.11 .02 -.02 -.21 .99&
N4: Self-Consciousness .80 -.11 -.11 -.01 -.09 .99&
N5: Impulsiveness .54 .23 .15 -.36 -.42 .95&
N6: Vulnerability .69 -.05 -.17 .01 -.45 .98"

El: Warmth -.04 .76 .14 .33 -.04 .96&

E2: Gregariousness -.17 .75 -.15 .03 -.06 .96"
E3: Assertiveness -.28 .43 .12 -.40 .34 .98&
E4: Activity .15 .51 .06 -.33 .34 .97&
ES: Excitement Seeking .12 .54 .19 -.37 -.06 .98"
E6: Positive Emotions -.21 .58 .41 .02 .01 .90"

01: Fantasy .26 .12 .53 -.06 -.27 .98&

02: Aesthetics .16 .18 .68 .19 .05 .97"
03: Feelings .25 .26 .66 -.02 .19 .94"
04: Actions -.23 .29 .42 -.19 -.15 .88'
05: Ideas -.02 .10 .73 -.10 .22 .97"
06: Values -.22 .00 .52 .24 -.08 .84

AI: Trust -.32 .33 .04 .56 .21 .95'

A2: Straightforwardness -.16 -.19 .02 .57 .24 .96"
A3: Altruism -.08 .32 .05 .72 .16 .93'
A4: Compliance -.11 -.25 -.07 .67 -.17 .94"
AS: Modesty .19 -.16 -.31 .43 -.11 .95"
A6: Tendermindedness .14 .20 .27 .58 .13 .94"

Cl: Competence -.46 .12 .14 -.02 .58 .99"

C2: Order -.10 -.06 -.16 .02 .69 .98"
C3: Dutifulness -.04 .03 -.11 .34 .76 .96"
C4: Achievement -.05 .17 .05 -.24 .74 .98"
C5: Self-Discipline -.42 -.01 .00 .09 .71 .96&
C6: Deliberation -.24 -.23 .02 .25 .60 .99"

Factor Congruence .98& .95" .94& .95" .97" .96"

N = 653. Loadings above 1.40 I are in bold.
"Congruence higher than that of 95% of rotations from random data.
'Congruence higher than that of 99% of rotations from random data
Note. From "Cross-Cultural Generalizability of the Five-Factor Model of Personality: Development and·
Validation of the NEO PI-R for Koreans," by R. L. Piedmont and J. H. Chae, 1997, Journal of Cross-Cultural
Psychology, 28, p. 142. Copyright 1997 by Western Washington University. Adapted with permission.
232 CHAPTER 6

improvement: All of the 12 facets for Extraversion and Agreeableness now

present themselves in a way that is identical to normative values. Thus, we
can be confident that the NEO PI-R is quite similar structurally in both Ko-
rean and American samples. The Procrustean rotation helped disentangle
sample specific error from more substantive relationships among these
variables.
Those interested in evaluating the generalizability of the NED PI-R to
new samples should use this approach for several reasons. First, CFA tech-
niques appear too stringent for use with personality data. All too often ad-
equate fit is not found for proposed models. The resulting "modifications"
that are introduced into the analyses in order to find a better solution are
frequently post hoc in nature and, at times, theoretically uninterpretable.
However, EFA techniques are not entirely appropriate either. EFAdoes not
capitalize on the fact that an a priori model exists and this information does
not fully inform the analytic process. Finally, this approach is the one rec-
ommended by the authors of the NED PI-R as being the most useful (Mc-
Crae, Zonderman, et al., 1996). Thus, any discussion of the factorial
stability of the NEO PI-R will need to be done within this medium if it is
to be seriously considered.

APPLICATION OF NEO PI-R DATA TO

VALIDITY RESEARCH

Because the five-factor model represents a comprehensive framework

of personality qualities, it is a useful standard for evaluating the persono-
logical content of scales. Correlating an instrument with markers of the
five factors provides a personological "fingerprint," enabling users to bet-
ter understand those aspects of the individual represented in the scale and
the kinds of outcomes that can be anticipated. As Ozer and Riese (1994)
have noted, mapping scales onto the five personality domains helps to
clarify a scale's relations to other psychological constructs. As was dis-
cussed in Chapter 2, evaluating personological content solely on the basis
of a scale's name can be quite misleading. The fact that two scales have the
same name does not mean that they assess similar constructs. Similarly,
different scale names does not mean that the scales measure different qual-
ities. Rather, the similarity or dissimilarity between two scales can be in-
ferred from the pattern of correlates each has to the five-factor model.
The value of the NED PI-R in this process is that it provides measures
of not only the five major personality dimensions, but also 30 facet scales
that can more specifically highlight the motivational basis of scales. In this
 RESEARCH APPLICATIONS 233

section, I expand on the value of using the NEO PI-R in validity research,
especially as it pertains to the development of a scale's construct and in-
cremental validity.

CONSTRUCT VALIDITY

The world of personality assessment is indeed very rich, if one were

to use the number of existing scales as a measure of affluence. As I re-
viewed in Chapters 1 and 2, there are literally thousands of measures de-
signed to capture a multitude of personality characteristics. There can be
no doubt that among those instruments there is a tremendous amount of
redundancy in content. Therefore, it becomes necessary for test developers
to show the uniqueness of the constructs they develop. Although the NEO
PI-R was not developed to replace all other assessment instruments, be-
cause it provides a relatively comprehensive sampling of personality traits
it is in a unique position to enable test developers to think with more so-
phistication about the theoretical basis of their constructs and about the
range of phenomena that the scale will predict.
Of course, forcing a greater degree of precision and accuracy into the
development of scales increases the difficulty of making new instruments,
but the payoff is certainly greater. It would be much to the benefit of the
field to have fewer scales, with clearly understood, nonredundant person-
ological content, than a haphazard amalgam of poorly defined measures
(see Meehl, 1978, p. 823, for an excellent discussion of this and related is-
sues). Bringing such empirical and conceptual clarity is where the NEO
PI-R comes in. Its value is particularly relevant in areas where researchers
are only beginning to delineate individual-difference constructs that de-
fine their field of interest. One such area is the field of religious research,
which is used here to exemplify how the NEO PI-R can be fruitfully ap-
plied to the construct validation process.
Religious researchers are interested in determining how religion and
spirituality influence human development and behavior. Recently this
area of endeavor has literally exploded with interest, as attested by the de-
velopment of literally dozens of new scales. This interest in scale develop-
ment reflects a real need for researchers in this area to get a handle on the
spiritual dimensions of the person. But this wholesale development of
scales has proceeded without much interest in integrating all these mea-
sures into some cohesive model. Thus, researchers in this field are left with
a rather chaotic mishmash of measures. The situation is so bad that Gor-
such (I988) has argued that no new religious measures be developed until
the construct validity of the existing instruments can be better ascertained.
234 CHAPTER 6

The real question in this area is whether these putatively spiritual con-
structs represent qualities of people that are distinct from established per-
sonality measures or are rather some new combination of them. Also,
without any larger framework for classifying spiritual constructs, the pos-
sibility of redundancy in content is very high. Although the five-factor
model is not religiously based, it can provide a useful starting point for
evaluating the personological content of religious constructs. As I have
noted elsewhere (Piedmont, 1996b), if religious variables have any overlap
with personality, then correlating such scales with the five factors can help
illuminate the underlying personological motivations of the scale as well
as highlight the types of behavioral outcomes to be anticipated. Further,
correlations with the five-factor model can also help to nomologically link
religious constructs with one another.
In order to address this question, I gave the NEO PI-R along with sev-
eral popularly used religious scales to a mixed-sex undergraduate sample
of 493 people. The first scale was the Hood Mysticism Scale (Hood, 1975),
a measure of reported mystical experiences. It contains two factors, a gen-
eral mystical experience dimension and the joyful or noetic qualities asso-
ciated with religious experiences. Also given was the Faith Maturity Scale
(Benson, Donahue, & Erickson, 1993). It reflects "the degree to which a per-
son embodies the priorities, commitments, and perspectives characteristic
of vibrant and life-transforming faith" (p. 3). There are two subscales, the
Vertical, which evaluates the degree to which one has a close relationship
with God, and the Horizontal, which looks at the degree to which one's
faith influences relationships with others. The third measure was the In-
trinsic/Extrinsic Religiosity Scale (Gorsuch & Venable, 1983). Intrinsic Reli-
giosity reflects a religious orientation that is internalized and done for its
own sake. Extrinsic Religiosity reflects an attitude toward religious in-
volvement for the sake of achieving some other goal. The final measure was
the Spiritual Well-Being scale (Ellison, 1983). This scale attempts to measure
spirituality as it speaks to overall well-being. There are two subscales, Reli-
gious Well-Being, which examines the level of satisfaction one experiences
in his or her relationship with God, and Existential Well-Being, which eval-
uates the level of personal meaning one has derived for his or her life. The
correlation of these scales to the NEO PI-R is given in Table 6-4.
Two conclusions can be drawn from these data. First, the number of
significant correlations attests to the relevance of the five-factor model for
understanding religious-oriented constructs. For example, the Intrinsic Re-
ligiosity scale correlated positively with Agreeableness and· Conscien-
tiousness. People who have an internalized view toward religion may be
perceived by others as being moral, reverent, considerate, responsible, and
dependable (Hofstee, de Raad, & Goldberg, 1992). The negative correlation
 RESEARCH APPLICATIONS 235

TABLE 6-4. Correlations between Select Religious Constructs and the NEO PI-R
Domains
NEO PI-R Domain
Religious scale N E 0 A C R2 a a-R2 I-a ~R2

Hood Mysticism I" .02 .08 .23' -.04 .02 .06' .90 .84 .10 .001
Hood Mysticism 2h -.04 .13d .20' .01 .09'" .06' .82 .76 .18 .002
Intrinsic Religiosity .00 -.04 -.09' .09'" .11' .03' .71 .68 .29 .007
Extrinsic Religiosity .11' -.04 -.07 -.07 -.09' .02 .69 .67 .31 .000
Existential Well-Being -.51' .34' -.05 .20' .39' .35' .87 .52 .13 .000
Religious Well-Being -.04 .05 -.11' .11' .13d .05' .93 .88 .07 .001
FMS - Horizontal -.06 .08 .17" .25' .16' .09' .74 .65 .26 .12'
FMS - Vertical -.09'" .11' .04 .13d .17" .05' .81 .76 .19 .02d
'General Mysticism Factor
'Joyful Religious Expression Factor
'lp<.01
< .05

"p < .001 (all correlations are two-tailed).

Note. "Strategies for Using the Five-Factor Model in Religious Research," by R. L. Piedmont, 1996. In R.
L. Piedmont (Chair), The Five-Factor Model and Its Valuefor Religious Research (1996, p. 14), Toronto, Ontario,
Canada, Symposium presented at the annual convention of the American Psychological Association.
Copyright 1996 by Ralph L. Piedmont. Reprinted with permission.

with Openness suggest a strong commitment to the person's values. This

pattern is very consistent with Gorsuch's (1994) definition of Intrinsic Re-
ligiosity, "the motivation for experiencing and living one's religious faith
for the sake of the faith itself. The person's religion is an end unto itself, a
goal pursued in the absence of external reinforcement" (p. 317). An Ex-
trinsic Religious orientation reflects mostly an anxious and insecure tem-
perament. Such individuals may engage in religious practices in an
attempt to minimize or reduce ongoing feelings of guilt and inadequacy.
Correlating these measures with the NEO PI-R provides insights into
their construct validity. It also enables one to look across scales to evaluate
redundancy. For example, the Existential Well-Being scale clearly captures
aspects of well-being and life satisfaction as evidenced by its very strong
correlations with (low) Neuroticism and (high) Extraversion. However, the
Religious Well-Being scale appears not to measure a similar construct. It
reflects not so much well-being as a sense of commitment to core, inter-
nalized values. Thus, the scale may be misleading in its "well-being" label
because it does not seem to capture those qualities that are motivationally
fundamental to the experience of well-being. In fact, the Religious Well-
Being scale appears personologically quite similar to the Intrinsic Reli-
giosity scale; both measure have correlates with the same NEO PI-R
domains of similar magnitude.
236 CHAPTER 6

A similar pattern of correlates raises the real possibility of redundancy

between the Religious Well-Being and Intrinsic Religiosity scales. These
scales appear to reflect similar motivational qualities of the individual. If
this is true, then one needs to question whether both instruments are really
necessary. However, similar domain correlates may mask still more subtle
differences. Any full analysis of redundancy would need to rest on a facet
analysis, correlating each scale with all 30 NEO PI-R facet scales to garner
a more nuanced interpretation of the measures. Such an analysis indicates
that in terms of Openness, both scales have a single significant correlation
with 06 (Openness to Values). On Agreeableness, both scales have a sig-
nificant correlation with A2 (Straightforwardness), although the Intrinsic
scale also correlates with A6 (Tender-Mindedness) while the Religious
Well-Being scale has additional correlates with Al (Trust) and A3 (Altru-
ism). Finally, on Conscientiousness both scales correlate significantly with
C4 (Achievement Striving) and C6 (Deliberation). Religious Well-Being
also correlates with CI (Competence).
Thus, even on the facet level these two instrument seem to reflect the
same personological content. Future research will need to focus on the dis-
criminant validity of these two scales. It may be that each predicts differ-
ent types of religious-spiritual outcomes. Without any compelling
evidence of this nature, one would do well to avoid using both scales in the
same study. Here is another example of how instruments with very differ-
ent names and intended uses capture redundant content. In fact, one
would make a serious interpretive error if he or she were to conclude that
a positive correlation between these two instruments (which should occur
given their overlap with the five-factor model) reflected the greater spiri-
tual satisfaction derived from those with an Intrinsic Religiosity. The cor-
relation would not reflect the association between two separate constructs,
but rather the redundancy in psychological content.
A second conclusion that can be drawn from the data in Table 6-4 is
that although the religious constructs share something in common with
the five-factor model, the multiple R2'S shown in the sixth column indicate
that these religious constructs are not completely redundant with the
model. These values were obtained by using each of the religious scales as
the dependent variable in a regression equation. The five NEO PI-R do-
main scores were entered as a single block. The R2 values represent the
amount of variance each religious scale has in common with the NEO PI-R
domain scores. Although there is a significant amount of overlap between
the two sets of constructs, it is clear that each religious variable contains a
significant amount of unique, reliable variance. This indicates that scores
on the religious scales contain information about people that is not shared
with the five-factor model. Thus, these scales have the potential to offer
new insights into psychological phenomena not already assessed by tradi-
 RESEARCH APPLICATIONS 237

tiona I personality scales. However, documenting this capability requires a

more rigorous empirical paradigm, one that includes the systematic eval-
uation of the relatedness of these two sets of variables to various criteria.
This involves an incremental validity paradigm.

THE INCREMENTAL VALIDITY PARADIGM

With so many scales available in the marketplace today and the very
real question of redundancy in content, it is critical that one evaluate the
unique value of a scale; its "bang for the buck," so to speak. What does a
scale tell us about some outcome that cannot be obtained from already es-
tablished measures? This question is particularly salient when one is in-
terested in developing new personality dimensions. The incremental
validity paradigm is useful in determining the common and unique pre-
dictiveness of various classes of variables. Its ultimate value is that it pro-
vides a direct index of what a scale adds to our predictive ability.
For religious research, incremental validity paradigms would be help-
ful in outlining those individual-difference qualities distinctive to religious
constructs that are predictive of important outcomes (e.g., prosocial be-
havior, altruism, racism, sexual behavior) over and above the more tradi-
tional personality variables as represented by the five-factor model and the
NEO PI-R. In this manner, researchers can begin to articulate more clearly
those aspects of human functioning that are singularly captured by reli-
gious constructs.
At the heart of this approach is stepwise regression analysis. Once a
suitable criterion is selected, measures of the five-factor model (i.e., the do-
main scores from the NEO PI-R) can serve as markers of traditional per-
sonality qualities. They would be entered on step 1 of the analysis. Then,
the religious construct or constructs of interest can be entered on step 2.
This variable would then identify that variance in the criterion not associ-
ated with the personality traits. A partial F-test (J. Cohen & Cohen, 1983)
would then determine whether the variance accounted for by the religious
variable (or the construct entered on step 2) added significantly to that al-
ready accounted for by the personality dimensions entered on the first step.
The data in Table 6-4 provide a systematic evaluation of the unique
variance of each of the religious constructs. The Cronbach alpha (a) for
each scale is given for this sample. This value outlines the amount of reli-
able variance in the observed scores. The column labeled "a - R2" indicates
the amount of unique, reliable variance in the observed scores. This value
subtracts the amount of shared variance between the religious and per-
sonality constructs from the total amount of reliable variance. The larger
this value, the more scores on this scale capture qualities of the individual
not shared by the five-factor model. A. S. Kaufman (1975) and Silverstein
238 CHAPTER 6

(1982) suggest that to conclude a scale has enough reliable variance to war-
rant a separate interpretation, its unique variance component should be
larger than its error (1 - (1) and at least 25% of the total variance. All of
these scales have a pattern consistent with these guidelines. Thus, they
have something to say independent of the five personality dimensions.
However, the most important question is whether this unique, reliable
variance has something meaningful to contribute to the prediction of im-
portant outcomes. This is where the incremental validity paradigm comes
into play. For the data presented in Table 6-4, those subjects also completed
a behavior-based index of prosodal behavior. This variable served as the
outcome for the stepwise regression analysis. The NEO PI-R domain scores
were entered on the first step, and then each of the religious constructs was
systematically entered on step 2. The last column of Table 6-4 (~R2) indi-
cates the increase in explained variance the religious constructs added to
predicting prosodal behavior over personality's contribution. As can be
seen, only the FMS scales added significantly to the explained variance. The
other measures, despite large amounts of reliable, unique variance, had
nothing to contribute uniquely to this outcome. Thus, any association be-
tween these religious constructs and prosocial behavior would be due en-
tirely to variance that overlaps with the more traditional personality scales.
These findings are surprising given the importance of the religious con-
structs. They are measurement staples of religious research, yet any predic-
tive value they may have in regard to nonreligious outcomes may be due to
their overlap with standard personality variables. It should also be men-
tioned that other studies have evaluated the predictiveness of these con-
structs against a wide range of outcomes (e.g., burnout among clergy, racism,
sexual attitudes, quality of life) and the results have been equally lackluster
(Csarny, 1997; Rodgerson, 1994). The one, and to date in my research, only ex-
ception has been the Faith Maturity Scale, which has consistently shown it-
self to possess significant incremental validity (e.g., Chen, 1996).
The incremental validity paradigm provides an empirically sophisti-
cated methodology for evaluating the utility of individual-difference con-
structs in providing additional, nonredundant information about people.
It very directly assesses what a new scale tells us about others over and
above what can be determined with existing scales. The NEO PI-R serves
as the anchoring point in these ventures. Although in this example I used
only the domain scores as the reference point, it is equally appropriate to
enter the facets instead. This will generate a larger portion of variance at-
tributable to personality. Scales that fail to improve predictiveness need to
be reconsidered. As seen in the example data here, many of the religious
constructs failed the incremental validity test. Thus, researchers in the area
of spirituality need to demonstrate the value of many of their constructs;
 RESEARCH APPLICATIONS 239

their relatedness to important psychological outcomes seems to be based

solely on their overlap with personality constructs. Merely repackaging es-
tablished constructs under new labels does not, ultimately, serve the cause
of science, the purpose of which is to expand our understanding of human
nature not merely rehash it.
In all fairness, though, the Faith Maturity Scale offers promise and
hope to religious researchers who are interested in identifying constructs
unique to their area of inquiry. This measure, although having something
in common with the NEO PI-R, does have sufficient unique variance to
provide additional explanatory power for a number of psychologically rel-
evant outcomes. Greater attention should be given to examining exactly
what this uniqueness is and to plumb its personological depths empiri-
cally. It may, in time, provide a new source of individual-difference con-
structs with the potential to expand both our current models of personality
and our ability to predict important psychological outcomes.

OUTCOME RESEARCH

An important contribution of psychological tests is their ability to

help us anticipate important life outcomes. Ultimately test data need to tell
us about the direction a person's life will move in and the kinds of events
that he or she is likely to experience. Ideally, it would be useful if assessors
could evaluate individuals on dimensions that would indicate not only the
client's current strengths and weaknesses, but also the risk potential to ex-
perience certain negative life outcomes, such as health problems, poor life
satisfaction, interpersonal difficulties, and so on. From my introductory re-
marks in this chapter, social scientists would also want to know how per-
sonality may dispose someone to experience a mental disorder, or what
their suitability for therapy may be, or their potential to benefit from an in-
tervention. These are all important questions, and the NED PI-R has some-
thing to contribute to all of these outcomes. For the purposes of this
section, I focus on using the NEO PI-R in an applied clinical context for as-
sessing change and predicting outcome.
Conducting outcome studies is a rather straightforward process.
What is needed is the ability to collect data at two points in time: at the be-
ginning of treatment and then again at the end. Certainly, the NEO PI-R
should be given at both assessment intervals. However, the NED should
not be considered all that is necessary. One would want to complement the
NEO with more specific clinical measures, perhaps the Brief Symptom In-
ventory (BSI; Derogatis, 1993) to assess symptomalogical distress, and a
measure of psychosocial issues. At the end of treatment one would repeat
240 CHAPTER 6

these measures as well as additional ratings that focus on the important

therapeutic elements of the treatment. This rating form should be com-
pleted by the therapist. The form would include subjective ratings of how
well the client has progressed along a number of relevant dimensions. An
example of such a form I have used is presented in Figure 6-3.
This form was used in a study evaluating the efficacy of an outpatient
substance abuse program that was spiritually based and vocationally fo-
cused. It dealt with chronic substance abusers who were primarily from a
transient population. The program ran for 6 hours a day, 5 days a week for
6 weeks. The goal at the end of the program was to have participants se-
cure some type of full-time employment. The program itself involved nu-
merous types of interventions, including individual and group counseling,
spiritual discussions, cultural trips, and vocational training. The form pre-
sented in Figure 6-3 was specifically designed, with consultation with the
staff, to assess issues that were relevant to their own goals.
In developing any type of outcome study, one should consider the
specific goals pursued, the types of clients treated, and the interventions
used. This information must be included in the outcome rating survey if it
is to have any relevance. Thus, the rating form in Figure 6-3 may not gen-
eralize to other clinical contexts, but readers should feel free to modify this
form to address the unique features of their contexts. Nonetheless, there
are some general features in the form that should be considered.
First, there are a number of simple Likert rating scales evaluating the
effectiveness of various aspects of the program for the client. These include
the perceived attitude of the clients toward the treatment and their success
in reaching various spiritual criteria. Each of these items is valuable in its
own right, because the items address very specific aspects of the interven-
tion program. Such single-item scales carry a respectable level of reliabil-
ity and validity (see Wanous, Reichers, & Hudy, 1997). However, one can
build in more reliability to these assessments by aggregating multiple
items to form scales and obtaining multiple ratings for each client. In col-
lecting multiple ratings, three raters are usually optimal, but two will work
very well.
A second value of this form is the presentation of a check list of vari-
ous therapeutic techniques. Therapists indicate those that seemed to work
well with the client by putting a 1/+" before it and indicate those interven-
tions that did not seem to be as effective by placing a "-" in front of it.
These types of ratings can be very useful for identifying which type of
client appears to respond to which type of treatment. This person-by-treat-
ment assessment is at the very heart of current clinical research. It repre-
sents the next step that the field will be taking as it hones its techniques
and defines their range of application.
RESEARCH APPLICATIONS 241

COUNSELOR'S FINAL EVALUATION

NAME: __________________________________________ DATE: _ __

I. Client's attitude towards program: From "I' Very little

Acceptance to "7" Very much Accepting . . . . . . . . . . . . . 2 3 4 5 6 7
2. Client's efforts in treatment: "I" very little effort to "7"
very diligent effort ........................... 2 3 4 5 6 7

3. Treatments which seemed most helpful (Place a plus sign (" + ") before all treatments that were used
SUCCESSFULLY; a minus sign (".") before any treatments that were used UNSUCCESSFULLY.
OGestalt o Vocational OSpiritual Direction OCultural Trips
OCfient Centered OGroup SeSsions o Probl.m Solving OJournaling
OCognitive Treatment OR.laxation Sellions DArt Thar.py
Olnsight DSystematic O.sans. DAA/NA Program

4. TlIERAPEUnC SUBPROCESSES USED (Place a plus sign (" +") before all subprocesses that
were used SUCCESSFULLY; a minu sign (" 0") before any subprocesses that were used
UNSUCCESSFULLY.
DSupport and r....ur.nce Dlnt.rpr.tation: Current life oCognitive restructuring
DAdvic., problem solving DProbl.m SOlving .dvlce OF ree association
OExperiencing techniques OStr••s manag.m.nt oPrayer/meditation
oInterpretation: early life techniques oSpiritual "adings
oOther:
---------------------------------------------------------
S. Length of program for client (in weeks):
6. Reason for termination (check only one):
Orelapse a other priorities a violation of progrem rules
Oloss of interest a legal difficulti.s a complet.d program
o Other:
----------------------------
7. How successfully were treatment Ilolls reached? "I" not
successful "4" partially successful "7" fully successful .... 2 3 4 5 6 7
8. What degree of recovery was noted? "I" likely to
relapse "4" no change "7" likely to stay clean . . . . . . . . . . 2 3 4 5 6 7
9. Spiritual development: "I" rigid (willfulness), "4' no
chanlle "7" flexihility (willingness) ................ 2 3 4 5 6 7
10. What was the client's attitude towards vocational
training? "I" resistive "4· complacent "7" proactive ..... 2 3 4 5 6 7
II. How well did the client learn the job searching/
vocational techniques provided? "I" not well, "7" very
well .................................... 2 3 4 5 6 7
12. Overall, how much personal growth (i.e., emotional,
vocational. spiritual) did this person experience? "I" not
much growth "7" a lot of growth . . . . . . . . . . . . . . . . . . 2 3 4 5 6 7

(Continued)
242 CHAPTER 6

13. Please rate the c1ient's motivation to want to make a

change in their lifestyle. "I" very little motivation to
change "7" very hillh motivation to change . . . . . . . . . . . 1 234 5 6 7
14. Please rate the degree to which you believe the client
CAN make a change in their lifestyle. "I" unable to make a
personal change "7" very capable of making a personal
change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . " 1 234 567
IS. Please rate the degree to which you believe the client
can maintain any change in their lifestyle. "I" unable to
mairnain any changes, will revert to earlier lifestyle "7" no
difficulty in maintaining any changes . . . . . . . . . . . . . . . 1 2 3 4 5 6 7

FIGURE 6-3. An example of a therapist's rating scale for treatment efficacy.

The size of this type of list is limited only by the kinds of interventions
that are being used. This type of approach appreciates the reality that most
clinicians are eclectic in their dealings with clients. Thus, rather than trying
to classify therapeutic interventions by the theoretical orientation of the
therapist, this type of list can provide a more detailed insight into exactly
what interventions were applied.
Another useful question asked is why clients dropped out of the pro-
gram. Oftentimes in clinical research one goal is to identify those who
complete from those who fail to complete the course of treatment. This
type of question seeks to identify the reason or reasons for early termina-
tion. Are certain types of individuals more prone to "quit" than others?
But, as this item illustrates, there are numerous possible reasons for some-
one's dropping out. In this example, one reason could be a relapse to using
drugs. Identifying those who are at risk for repeat drug use and those more
likely to stick with the program is of fundamental interest to substance
abuse counselors. However, people left this program for other reasons, in-
cluding a loss of interest, legal problems (such as being arrested for an ear-
lier crime) or life circumstances that took them out of the area. These types
of cases should not be included in any analyses of recidivism.
A similar kind of form can also be given to clients to complete on their
own. Such a questionnaire can access their perceptions of the therapeutic
program and how well each element was received. Clients can also rate how
successful they found the treatment to be. Other questions can be included
about the client's view of a number of useful dimensions, all of which speak
to his or her satisfaction with the services that were rendered. Figure 6-4 pro-
vides an example of such a questionnaire that I have used, although there
are a number of standardized client satisfaction questionnaires already
available (e.g., Larsen, Attkisson, Hargreaves, & Nguyen, 1979).
As can be seen in Figure 6-4, clients are directly asked to indicate how
successful they found various aspects of the treatment and how much im-
 RESEARCH APPLICATIONS 243

PARTICIPANT'S EVALUAnON
NAME: _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ DATE: _ _ _ _ __
As pan %ur evaillOtion o/the tjftctilltrltss o/the CHRP program, we would Ilkt your view o/the
counseling you Milt rtcttvtd. PlellSe answer thefollowlng questions.
I. How helpful did you find your participation in the CHRP program in dealill8 with the problem that
brought you in for treatment?
OVery helpful Osomewhat helpful [Jnot very helpful o not helpful at all
2. In general, how do you feel now compared to how you felt when you began the program?
Omuch better abetter Othe same o worse
3. Do you have more self-esteem and feel better about yourself?
Omuch better Obetter Othe same o worse
4. Do you think you are able to get along better with other people?
Qmuch better abetter Othe same o worse
S. Which aspects of the program seemed the most helpful to you? Please check one response for each
aspect.
very somewhat not
helpful helpful helpful
or don't
know
Gettill8 advice about what to do ................... . Q o Q
Feelill8 accepted by the group . . . . . . . . . . . . . . . . . . . . . . Q a o
Learning to understand yourself in relation to the God of
your understandill8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . o a o
Getting reassurance and support ................... . Q a o
Learnill8 about other people . . . . . . . . . . . . . . . . . . . . . . . Q a o
Expressing your feelings .......•................. o o a
Taking responsibility for your life .................. . o o a
The mornill8 dialogue .......................... . a a a
The employment readiness training ................. . Q a a
The spiritual element to the program ................ . o a a
Learning about" Action, Attitude, Awareness" .......... . o a a
Learning about "Serendipity" ..................... . Q a a
If you knew someone with similar problems, would you recommend the program to
them? .................................................... aVe. ONo

FIGURE 6-4. Example of a client satisfaction questionnaire given at the end of treatment.

provement they feel they have experienced. From a consumer point of

view, these types of questions provide prima facie evidence of treatment ef-
ficacy. If clients are not happy with what they experienced, no matter what
the therapist may claim, then the value of treatment must be questioned.
244 CHAPTER 6

This type of questionnaire can also provide very useful feedback informa-
tion for therapists. A therapist may believe that a certain intervention is
important or even critical to the process of therapy, but clients may indi-
cate that it is the least useful. Finding out why such discrepancies exist can
be helpful in tailoring a treatment program to better meet the needs of the
clients. It may also provide insights as to why resistance may be develop-
ing in some or all clients at certain points in the program.
The utility of both these questionnaires is that they are easy to admin-
ister and interpret. They do not take much time to complete and can pro-
vide a wide range of valuable information. Given the high face validity of
the items, these measures can be readily interpreted by simply reading
over the responses. Further, these responses can be easily entered into data
programs for later statistical analyses. These scores become very interest-
ing when aggregated over many clients and then correlated to the psycho-
metric information that is collected in tandem with the forms. The next
sections outline some of the kinds of data I have obtained from using
scales like the ones just described.

UNDERSTANDING CHANGE OVER TIME

As we noted earlier in this book, the dimensions of the NEO PI-R re-
main very stable in adulthood; as Costa and McCrae (1994) noted, 25-year
test-retest values hover around .80, and it is estimated that over a 50-year
adult life span 60% of the variance remains stable. These stability coeffi-
cients are complemented by analyses of mean level change which also
show no significant shifts over time. Based on this data Costa and McCrae
noted that personality is pretty much "set like plaster" after age 30.
Such evidence for stability raises a number of theoretical, philosophi-
cal, and ideological questions. How, then, do we define the concepts of
psychological growth and freedom? To what degree do individuals have
the capacity to make choices? Is personality destiny? From an applied per-
spective, what purpose do psychological interventions serve? Should clin-
icians attempt to affect clients by helping them make shifts in their
fundamental psychological structures or through improving their psy-
chosocial instrumentality (e.g., coping ability)? Answers to these questions
certainly address our field at its most fundamental level and force us to
think most seriously about concepts of change and development (these is-
sues were touched on in Chapter 1 in the discussion on genotypic versus
phenotypic change).
But even if adulthood is not characterized by any natural psychologi-
cal metamorphoses, are such transformations possible given the interdic-
tion of certain events? There are three types of events that could arguably
RESEARCH APPLICATIONS 245

provide transmuting opportunities: traumatic personal experiences (e.g.,

war), religious conversions, and therapeutic interventions. These rather
radical experiences may provide a portal into the psyche that capitalizes
on our innate psychic plasticity.
In order to address the ability of individuals to experience genotypic
change as a function of psychotherapy, I evaluated clients participating in
a 6-week outpatient drug rehabilitation program (see Piedmont, 1996a;
Piedmont & Ciarrocchi, in press). In this study I was able to measure par-
ticipants on the NEO PI-R both before and after their involvement in the
program. A total of 99 individuals (58 men and 41 women) successfully
completed the 6-week program. All subjects had been drug-free for at least
30 days prior to entering this program. Most were alcohol and cocaine
abusers with an average of 14 years of substance involvement. The treat-
ment they received included individual sessions, CDAgroups, discussions
of spirituality, and vocational counseling. The program was conducted 5
days a week for 6 hours a day. The results are presented in Table 6-5.
As can be seen, over the 6-week treatment interval, significant shifts
occurred in all the major domains of personality. The largest gains (about
one-half standard deviation) were noticed for Neuroticism and Conscien-
tiousness; individuals decreased their overall levels of emotional dyspho-
ria and increased feelings of competence and self-discipline. Also, the
changes in Extraversion and Agreeableness indicate a shift toward an in-
terpersonal style characterized by greater affability and trust. The increase
noted on Openness reflects greater awareness of potential opportunities
and a broader awareness of values.
Overall these results are extremely encouraging. Although "normal"
adulthood may not be characterized by any quantitative shifts in personal-

TABLE 6-5. t-Test Evaluating Mean Level Change in NEO

PI-R Domain Scales over a 6-Week Interval.
NEO PI-R domains TImet TIme 2 t Difference
Neuroticism 61.34 54.22 -6.98"
Extraversion 50.36 52.75 3.68"
Openness to Experience 54.00 56.25 3.70"
Agreeableness 41.69 44.44 3.77"
Conscientiousness 41.27 46.72 5.59"
N=99
'p < .001, two-tailed.
NEO PI-R scores presented as T-scores with a mean of 50 and standard deviation
of 10, based on normative data from unpublished manuscript.
Note. From Psychometric Utility of the NEO PI-R in an Outpatient, Drug Rehabilita-
tion Context (p. 32), by R. L. Piedmont and J. W. Ciarrochi (in press). Adapted
with permission.
246 CHAPTER 6

ity, those seeking changes may be able to find them. These data indicate
that psychological interventions may be able to induce positive changes in
one's underlying dispositions. The magnitude of these shifts is indeed quite
high (see Trull, Useda, Costa, & McCrae, 1995; Bagby, Joffe, Kalemba, &
Harkness, 1995 for comparisons) and suggests that the more time invested
in treatment, the larger the personological gains experienced by the client.
Of course, further follow-up data is necessary to determine whether
these treatment changes last. Are the noted changes temporary, with the
individuals returning to premorbid levels in time? Also, we need to deter-
mine whether the NEO PI-R scores obtained at posttreatment continue to
be as predictive of life outcomes as found with individuals who do not
show any change. More longitudinal research with clinical samples needs
to be undertaken to answer these questions. However, for our purposes,
two conclusions can be gleaned from these data. First, it appears that the
plaster cast can be broken. Those seeking serious changes in their lives
may be able to find it through therapy. The magnitude of effect for some of
the NEO PI-R domains was indeed quite large, as compared with other
outcome studies using this instrument. But the intervention was quite in-
tense; these individuals experienced 30 hours of therapy a week for 6
weeks. Individual treatment usually is not this focused and does not last
so long. The 180 hours of treatment correspond to about 3 years of tradi-
tional one-on-one therapy. Thus, the more put into therapy, the more ben-
efit that may be received.
The second value of these data is that they suggest that the NEO PI-R
may be a useful measure of psychotherapy outcome. Its ability to change
in response to therapy indicates that it is sensitive to treatment interven-
tions and can be a very useful index for gauging the extent of experienced
change. The fact that clients changed on Extraversion and Agreeableness
makes a powerful statement about the kinds of shifts in interpersonal
style. Not only do these data validate the treatment that was received, but
they also establish very useful expectations for future behavior on the part
of clients. With a more approachable and engaging style that will lend it-
self to developing more mutually satisfying emotional relationships with
others, these people can anticipate a deeper and more fulfilling quality to
their lives. The increased scores on Conscientiousness lead one to expect a
greater capacity on the part of clients to obtain and maintain vocational
opportunities.
By simply collecting NEO PI-R data pre- and posttreatment, one can
obtain a useful description of where clients are at these points. Shifts on the
domains and facet scales can indicate how and to what degree therapy af-
fected clients. As I pointed out in the beginning of this chapter, this type of
information can be very useful in documenting efficacy for third-party
 RESEARCH APPLICATIONS 247

payers. It is also useful for showing the magnitude of intervention neces-

sary to bring about useful, sustainable change in clients. The next section
looks at how outcome data can be used to select treatments for clients.

SELECTING TREATMENTS FOR PERSONS

Perhaps the frontier for clinical research is found in research efforts to

match treatments to persons. It is here that the greatest efficiency in pro-
viding clinical services can be found. Individuals no doubt bring their own
issues to therapy, but they also bring their own characteristic ways of man-
aging events in their environment. These styles of adaptation determine
the what and how of therapy. Personality dispositions no doubt influence
how clients present themselves in therapy and determine their responses
to interventions. Personality styles also bring with them certain therapeu-
tic opportunities and pitfalls. Therefore, understanding clients along the
dimensions of the five-factor model can help orient treatment in ways that
capitalize on clients' strengths.
Perhaps the most thorough treatment of personality by treatment in-
teractions using the NEO PI-R is presented by Miller (1991). Drawing on in-
formation from 101 of his own clients, Miller provided useful clinical
insights into how individuals high and low on each of the five domains
would present themselves to therapy. He also outlines some of the key
problems these clients are likely to experience along with potential treat-
ment opportunities and pitfalls. For example, individuals high on Neuroti-
cism present themselves with a variety of negative affects. Their presenting
problems span the full spectrum of neurotic pains. Although such individ-
uals may always experience personal pain regardless of how much therapy
they receive, such emotional distress can certainly motivate patient com-
pliance with treatment. On the other hand, those low on Neuroticism pre-
sent themselves as emotionally bland. Most of their issues may arise from
situational problems; these individuals can and do benefit from advice and
values clarification. Therapists should be cautious not to interpret the emo-
tional blandness as defensiveness. Table 6-6 provides the strengths and
weaknesses Miller (1991) associated with each of the five factors.
Of particular interest in the Miller paper was his finding that (low)
Neuroticism and (high) Conscientiousness were significantly related to
positive ratings of treatment outcome. Certainly high Conscientious indi-
viduals will make efforts to improve and to comply with treatment proto-
cols. Such efforts may not be undermined by higher levels of negative
affect, which compromises self-confidence and one's ability to tolerate
the personal discomfort associated with any change. The lower levels of
Neuroticism may reflect clients who come to therapy to address situation-
 248 CHAPTER 6

TABLE 6-6. Clinical Issues Surrounding Each of the Five Factors.

NEOFactor Potential Strength Potential Weakness
Neuroticism
High: Psychological Pain motivates Existence likely to remain
compliance uncomfortable; high N
cannot be interpreted away
Low: Wants and can benefit from Emotional Blandness may
advice and values be misunderstood as
clarification defensiveness

Extraversion
High: Comfortable with less Talkativeness can blunt
structured approaches; treatment focus
optimistic and energetic
Low: Comfortable with structured Lacks enthusiasm for
approaches interaction with therapist

Openness
High: Prefers imaginative Excessive curiosity can
approaches scatter resources
Low: Responds well to practical Rigidity and lack of
approaches; Education, curiosity can be mis-
support, behavior therapy understood as resistance

Agreeableness
High: Treatment alliance easily Accepts interpretations
formed uncritically. Need to
please therapist inter-
feres with disclosure of
transference
Low: Assertiveness and clear Hostility and skepticism
thinking about self-interest toward therapist; diffi-
facilitate problem solving cult to form treatment
alliance

Conscientiousness
High: Works hard to benefit from (possibly none)
treatment. Willing to
tolerate discomfort and
fustration
Low: (possibly none) Unlikely to do homework;
likely to reject interven-
tions that require hard
work or toleration of
discomfort
From Miller. T. (1991). The psychotherapeutic utility of the five-factor model of personality: A clinician's
experience. Journal ofPersonality Assessment, 57, 418-419. Copyright 1991 by Lawrence Erlbaum Associates,
Inc., Publishers. Adapted with permiSSion.
 RESEARCH APPLICATIONS 249

ally induced problems. The task with these people is to find ways of mar-
shaling already existing personal resources to manage the new stressor,
which may in time pass away. No doubt, the more personal resources a
client brings to therapy, the more likely they are to experience a positive
outcome. The real goal is to identify ways to engage clients with fewer psy-
chological resources and still bring about a better clinical result.
Using the evaluation form presented in Figure 6-3, I conducted a pro-
gram evaluation of an outpatient substance abuse program (Piedmont &
Ciarrocchi, in press). Data from this project have been presented through-
out this book. One of the interesting parts of this project was to correlate
NEO PI-R scores at Time 1 (entrance into the program) with counselor rat-
ings obtained at Time 2 (at the end of treatment). Of particular interest
were the correlations with items in questions 3 and 4 in Figure 6-3. These
associations speak to those initial qualities of the clients that seemed to
benefit from specific treatment interventions.
Our results indicated that those high on Neuroticism benefited from
client-centered therapy and systematic desensitization, while problem-
solving advice was not seen as effective with these types of clients. Open-
ness correlated positively with vocational techniques. Those high on
Agreeableness were rated as responding well to the AAand NA programs,
while those low on Agreeableness responded well to relaxation sessions,
art therapy, and joumaling. Those high on Conscientiousness responded
well to Gestalt techniques. Openness did not correlate with any of the
treatment techniques.
These results should not be interpreted as generalizing to all thera-
peutic contexts; more research is needed before that conclusion can be
drawn. Rather, these findings represent patterns of relationships that hold
for a particular program, and this is the data's greatest value. In conducting
outcome evaluations, clinicians are interested in doc\imenting the efficacy
of their particular interventions; for determining which of their specific
clients will respond to the presented treatments. This is why outcome re-
search is so important for individual clinicians. It provides a framework for
better understanding the nuances of their clinical practice. There are so
many factors that influence how a person high on some personality di-
mension responds to a treatment, such as the demeanor of the therapist and
the way the treatment is presented. These unique features of a therapist's
treatment environment are captured in these types of ratings and provide
important feedback on how well various interventions are working.
These data also provide some support for the preference of therapists
for eclectic approaches to treatment. Certainly there are no generic treat-
ment models that work for all people equally well. Therapists need to re-
spond to a myriad of nonspecific factors that clients present in order to
250 CHAPTER 6

find an intervention that seems to be congruent with the client's needs. To

date, however, there has been no systematic way of linking client charac-
teristics to therapeutic techniques or to treatment outcomes. Using simple
evaluative rating scales like the one presented in Figure 6-3, clinicians can
obtain data that can be effective in identifying a wealth of curative factors
previously unnoticed.

UNDERSTANDING OUR CLIENTS

Another value of the NEO PI-R is that it provides an opportunity to

collect data not just on individuals, but on groups. Who, in general, are the
people presenting themselves for treatment? What kinds of clients do I
serve? By answering these questions, therapists can begin to understand
the kinds of needs people bring for treatment. It provides a useful psycho-
logical orientation to those we treat.
For example, Table 6-7 provides an overall description of the 132 in-
dividuals who began the outpatient substance abuse program outlined
earlier in this chapter. The average score for everyone on each of the 35
NEO PI-R scales was calculated and then entered into the scoring software
for evaluation, using the combined norms for reference. The value of this
type of analysis is that it gives an overview of the entire group's personal-
ity presentation.
The resulting profile indicates that these individuals are likely to pre-
sent a defensive facade of superiority and may use defense mechanisms
such as acting out and projection. These individuals may have a marginal
level of life satisfaction, being more sensitive to life's problems than its re-
wards. Overall, these individuals may prefer an interpersonal style that
can be described as cold, unfeeling, dominant, assured, and especially
arrogant and calculating. As a group, this personality profile quite closely
resembles that of the Borderline Personality Disorder. Thus, these individ-
uals may be skeptical and antagonistic in treatment and reluctant to estab-
lish a treatment alliance with the therapist.
Aggregating data at the group level has three important implications.
First, it allows one to appreciate who is presenting for treatment. As we see
in Table 6-7, those coming for substance abuse treatment present a specific
personality profile. As we saw in the previous section, these personality di-
mensions are differentially responsive to various treatment interventions.
Therefore, certain types of interventions will be more effective than others.
It was not surprising that therapists in this situation frequently used AA
and NA groups, and Gestalt and client-centered techniques: These tech-
niques seemed to have the best results with these types of clients. This
information is useful for anticipating issues that will emerge over the
 RESEARCH APPLICATIONS 251

TABLE 6-7. Personality Characteristics of Clients

Entering an Outpatient Substance Abuse Program
NEO PI-R domain and facet T-Score Range
Neuroticism 60 HIGH
Anxiety 55 AVERAGE
Hostility 58 HIGH
Depression 59 HIGH
Self-Consciousness 58 HIGH
Impulsiveness 55 AVERAGE
Vulnerability to Stress 60 HIGH
Extraversion 51 AVERAGE
Warmth 45 AVERAGE
Gregariousness 49 AVERAGE
Assertiveness 48 AVERAGE
Activity 51 AVERAGE
Excitement Seeking 57 HIGH
Positive Emotions 50 AVERAGE
Openness to Experience 53 AVERAGE
Fantasy 53 AVERAGE
Aesthetics 56 HIGH
Feelings 54 AVERAGE
Actions 52 AVERAGE
Ideas 52 AVERAGE
Values 51 AVERAGE
Agreeableness 44 LOW
Trust 37 LOW
Straightforwardness 40 LOW
Altruism 45 AVERAGE
Modesty 45 AVERAGE
Compliance 50 AVERAGE
Tendermindedness 54 AVERAGE
Conscientiousness 45 AVERAGE
Competence 41 LOW
Order 50 AVERAGE
Dutifulness 37 LOW
Achievement 46 AVERAGE
Self-Discipline 43 LOW
Deliberation 44 LOW

course of therapy. Given the interpersonal style of the group, this outpa-
tient program had to build in techniques that would address the blustery
yet insecure and untrusting interpersonal nature of the group.
A second value of these types of data is that they allow clinicians to cre-
ate local norms for their psychological instruments. Knowing the overall
population characteristics, it becomes useful to evaluate each new client in
252 CHAPTER 6

reference to these qualities. Yes, these clients are suspicious and non trust-
ing, but how does this new client compare on these dimensions relative to
others who have come through the program? Is he or she more detached or
less so? These specific questions provide more texture and nuance to psy-
chological evaluations. Developing local norms allows for a more fine-
grained analysis of personality as it applies in a particular treatment context.
The final value of aggregate data works out of the first two: It enables
therapists to conduct a needs assessment of their clients. Given the overall
portrait of the clients, what are the needs they bring to treatment? Certainly,
the overall profile presented here indicates individuals in need of several
things. Interpersonally, they need to learn how to better relate to others; to
be able to initiate and maintain emotionally satisfying relationships. People
cannot always be seen as threatening objects, but rather as possible sources
of support and encouragement. Intrapersonallr- these individuals need to
create better self-images and to develop more efficient coping skills. They
need to become less impulsive and more self-controlled. Treatment with
these individuals needs to proceed on multiple levels simultaneously.
Group data can provide a useful summary statement of the personal-
ity qualities in one's treatment population. Such information can certainly
inform treatment selection and identify potential areas of intervention. The
use of group data can be extended to reflect different subgroups as well.
For example, it may be possible to identify normatively those who re-
sponded well to treatment and those who did not. These aggregated pro-
files can be used in two ways. First, the personality differences between the
two groups can be plotted and interpreted for its clinical significance. Why
do clients high on dimension X and low on dimension Y do poorly, while
those with the opposite pattern do well? Is it something about our center,
or therapist, or treatment modality?
A second way that these group profiles can be used is in the selection
of clients. Some programs, like the substance abuse treatment program dis-
cussed earlier, have limited resources and must evaluate prospective
clients prior to accepting them for treatment. The effort is to identify those
who appear most likely to benefit from the intervention. In such a situa-
tion, the profiles of new clients can then be compared to these aggregated
norms of those who successfully completed the program. The degree of fit
can be used as part of the selection process.

CONCLUSIONS

The NEO PI-R is an empirically powerful instrument that has numer-

ous applications in both clinical and research contexts. The strength of
the instrument is its foundation in the five-factor model of personality, a
 RESEARCH APPLICATIONS 253

comprehensive taxonomy of personality traits. The NEO PI-R thus can be

useful in evaluating the validity of new constructs and for organizing and
integrating personality information in a given context. Clinically, the in-
strument provides a useful summary of personality dispositions that have
numerous treatment implications.
The information presented in this chapter should be used as a para-
digm for employing the NEO PI-R in various contexts. Strategies were
given that address important questions about the underlying structure of
the instrument as well as for using the measure in new areas, such as cross-
cultural research. From an applied perspective, the NEO PI-R has much to
contribute to the advancement of clinical practice. Clinicians should feel
free to use and adapt the forms presented for assessing client satisfaction
and measuring treatment outcome to suit their own particular needs.
The need for clinical data will not go away in this current managed
care environment. If we are ever going to be able to stem the effects of the
unmitigated pressures for cost cutting that the third-party payer system
presents, then we will need to collect informative, rigorous data. Doing
this requires the investment of only a minimal amount of time. The ease
and simplicity of current spreadsheet programs and statistical packages
makes the collection, storage, and analysis of data more user-friendly than
at any other time. The benefits far outweigh these limited costs.
My purpose in writing this book was to provide a basic introduction
to the NED PI-R and its interpretive and empirical applications. The in-
strument can add much depth and clarity to any endeavor. Yet, the current
research has only scratched the surface of what this instrument can do and
new applications continue to be found. Perhaps the greatest value of the
NEO PI-R is that it places the user in the very center of current personal-
ity research and theory. Although the NED PI-R is formulated on a trait
perspective, it provides a way of interfacing with any number of psycho-
logical theories or perspectives. Its robust empirical nature provides re-
searchers with a useful reference point for evaluating constructs and their
predictive utility.
Ultimately, though, as the five-factor model of personality continues to
become more widely used, the social sciences may approach the goal of cre-
ating a truly cumulative knowledge base. Finally, research studies can be
meaningfully related to one another by virtue of the personality domains
they assess. Areas of redundancy and overlap can be quickly observed,
while new dimensions of investigation can be highlighted, evaluated, and
reconciled empirically to established models. The social sciences will in-
deed take on a greater focus and harmony.
However, one should not interpret this to mean that the five-factor
model or the NEO PI-R is intended to replace all other assessment instru-
254 CHAPTER 6

ments. Rather, one should consider the five-factor model the first step of
scientific inquiry: the accurate description of personality. By providing a
common language for talking about personality-related phenomena, we
lay the foundation for the development of more precise theories and more
powerful empirical tests of the hypotheses these theories generate.
 REFERENCES

Aldenderfer, M. S., & Blashfield, R. K. (1984). Cluster analysis. Beverly Hills, CA:
Sage.
Allport, G. W. (1937). Personality: A psychological interpretation. New York: Holt.
Allport, G. W. (1961). Pattern and growth in personality. London: Holt, Rinehart &
Winston.
Allport, G. w., & Odbert, H. S. (1936). Trait names: A psycho-lexical study. Psycho-
logical Monographs, 47(1, Whole No. 211).
Almagor, M., Tellegen, A., & Waller, N. G. (1995). The big seven model: Across-cul-
tural replication and further exploration of the basic dimensions of natural lan-
guage trait descriptors. Journal of Personality and Social Psychology, 69, 300-307.
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental
disorders (4th ed.). Washington, DC: Author.
Andrews, F. M., & Withey, S. B. (1976). Social indicators ofwell-being: Americans' per-
ceptions oflife quality. New York: Plenum.
Arbisi, P. A, & Ben-Porath, Y. S. (1995). An MMPI-2 infrequent response scale for
use with a psychopathological populations: The infrequency-psychopathol-
ogy scale, F(p). Psychological Assessment, 7,424-431.
Avia, M. D., Sanz, }., Sanchez-Bemardos, M. L., Martinez-Arias, M. R., Silva, E, &
Grana, J. L. (1995). The five-factor model-II. Relations of the NEO-PI with
other personality variables. Personality and Individual Differences, 19, 81-97.
Bagby, R. M., Joffe, R. T., Parker, J. D. A, Kalemba, Y., & Harkness, V. (1995). Major
depression and the five-factor model of personality. Journal of Personality Dis-
orders, 9, 224-234.
Bailey, K. D. (1994). Typologies and taxonomies: An introduction to classification tech-
niques. Thousand Oaks, CA: Sage.
Bakan, D. (1966). The duality of human existence: Isolation and communion in Western
man. Boston: Beacon.
Bandura, A (1989). Human agency in social cognitive theory. American Psychologist,
44,1175-1184.

255
 256 REFERENCES

Barlow, D. H., Hayes, S. c., & Nelson, R O. (1984). The scientist-practitioner: Research
and accountability in clinical and educational settings. New York: Pergamon.
Barrick, M. R, & Mount, M. K. (1991). The Big Five personality dimensions and job
performance: A meta-analysis. Personnel Psychology, 44,1-26.
Barry, W. A. (1970). Marriage research and conflict: An integrative review. Psycho-
logical Bulletin, 73, 41-54.
Ben-Porath, Y. S., & Waller, N. G. (1992). Five big issues in clinical assessment: Are-
joinder to Costa and McCrae. Psychological Assessment, 4, 23-25.
Bendig, A. W. (1958). Comparative validity of objective and projective measures of
need achievement in predicting students' achievement in introductory psy-
chology. Journal of General Psychology, 59, 51-57.
Benet, v., & Waller, N. G. (1995). The Big Seven factor model of personality de-
scription: Evidence for its cross-cultural generality in a Spanish sample. Jour-
nal of Personality and Social Psychology, 69, 701-718.
Benson, P. L., Donahue, M. J., & Erickson, J. A. (1993). The faith maturity scale: Con-
ceptualization, measurement, and empirical validation. Research in the Social
Scientific Study of Religion, 5, 1-26.
Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness of fit in the
analysis of covariance structures. Psychological Bulletin, 88, 588-606.
Bergeman, C. S., Chipuer, H. M., Plomin, R, Pedersen, N. L., McClearn, G. E., Nes-
selroade, J. R, Costa, P. T., Jr., & McCrae, R R (1993). Genetic and environmen-
tal effects on Openness to Experience, Agreeableness, and Conscientiousness:
An adoption/twin study. Journal ofPersonality, 61, 159-179.
Berne, S. L. (1961). Transactional analysis in psychotherapy. New York: Grove.
Block, J. (1965). The challenge of response sets: Unconfounding meaning, acquiescence,
and social desirability in the MMPI. New York: Appleton-Century-Crofts.
Block, J. (1971). Lives through time. Berkeley, CA: Bancroft Books.
Block, J. (1995). A contrarian view of the five-factor approach to personality de-
scription. Psychological Bulletin, 117, 187-215.
Block, J. H., & Block, J. (1980). The role of ego-control and ego-resiliency in the or-
ganization of behavior. In W. A. Collins (Ed.), Minnesota symposia on child psy-
chology (pp. 39-101). Hillsdale, NJ: Erlbaum.
Bond, M. H. (1979). Dimensions used in perceiving peers: Cross-cultural compar-
isons of Hong Kong, Japanese, American, and Filipino university students. In-
ternational Journal of Psychology, 14,47-56.
Bond, M. H., Nakazato, H., & Shiraishi, D. (1975). Universality and distinctiveness
in dimensions of Japanese person perception. Journal of Cross-Cultural Psy-
chology, 6, 346-357.
Borkenau, P. (1992). Implicit personality theory and the five-factor model. Journal
of Personality, 60, 295-327.
Borkenau, P., & Liebler, A. (1992). Trait inferences: Sources of validity at zero ac-
quaintanceship. Journal of Personality and Social Psychology, 62, 645-657.
Borkenau, P., & Liebler, A. (1993). Convergence of stranger ratings of personality
and intelligence with self-ratings, partner ratings, and measured intelligence.
Journal of Personality and Social Psychology, 65,546-553.
 REFERENCES 257

Borkenau, P., & Ostendorf, F. (1990). Comparing exploratory and confirmatory fac-
tor analysis: A study on the 5-factor model of personality. Personality and Indi-
vidual Differences, 11, 515-524.
Botwin, M. (1995). Review of the NEO PI-R In J. Conoley & J. Impara (Eds.), Men-
tal measurement yearbook (12th ed., pp. 862-863). Lincoln: University of Ne-
braska Press.
Bouchard, T. J. (1968). Convergent and discriminant validity of the Adjective Check
List and the Edwards Personal Preference Schedule. Educational and Psycho-
logical Measurement, 28, 1165-1171.
Bouchard, T. J., Jr., & McGue, M. (1990). Genetic and rearing environmental influ-
ence on adult personality: An analysis of adopted twins reared apart. Journal
of Personality, 58, 263-292.
Boyle, G. J. (1989). Re-examination of the major personality-type factors in the Cat-
tell, Comrey, and Eysenck scales: Were the factor solutions by Noller et al. op-
timal? Personality and Individual Differences, 10, 1289-1299.
Brooner, R K, Herbst, J. H., Schmidt, C. W., Bigelow, G. E., & Costa, P. T., Jr. (1993).
Antisocial personality disorder among drug abusers: Relations to other per-
sonality diagnoses and the five-factor model of personality. Journal of Nervous
and Mental Disease, 181,313-319.
Buros,o. K (Ed.). (1978). Eighth mental measurements yearbook (Vol. 1). Highland
Park, NJ: Gryphon.
Buss, D. M. (1984). Marital assortment for personality dispOSitions: Assessment
with three different data sources. Behavior Genetics, 14, 111-123.
Buss, D. M. (1991a). Conflict in married couples: Personality predictors of anger
and upset. Journal of Personality, 59, 663-688.
Buss, D. M. (1991b). Evolutionary personality psychology. In M. Rosensweig & L.
Porter (Eds.), Annual review of psychology (pp. 459-451). Palo Alto, CA: Annual
Reviews.
Buss, D. M. (1992). Manipulation in close relationships: Five personality factors in
interactional context. Journal of Personality, 60, 477-499.
Buss, D. M., & Craik, K H. (1980). The frequency concept of disposition: Domi-
nance and prototypically dominant acts. Journal of Personality, 48, 379-392.
Buss, D. M., & Craik, K H. (1983). The act frequency approach to personality. Psy-
chological Review, 90, 105-126.
Butcher, J. N., & Rouse, S. V. (1996). Personality: Individual differences and clini-
cal assessment. Annual Review of Psychology, 47, 87-111.
Butcher, J. N., Dahlstrom, W. G., Graham, J. R, Tellegen, A., & Kaemmer, B. (1989).
MMPI 2: Manual for administration and scoring. Minneapolis: University of
Minnesota Press.
Cantor, N. (1990). From thought to behavior: "Having" and "doing" in the study of
personality and cognition. American Psychologist, 45, 735-750.
Cantor, N., Norem, J., Langston, c., Zirkel, S., Fleeson, W., & Cook-Flannagan, C.
(1991). Life tasks and daily life experience. Journal of Personality, 59, 425-451.
Cantril, H. (1965). The pattern ofhuman concerns. New Brunswick, NJ: Rutgers Uni-
versity Press.
258 REFERENCES

Capara, G. V., Barbaranelli, c., Borgogni, L., & Perugini, M. (1993). The "Big Five
Questionnaire": Anew questionnaire to assess the five-factor model. Person-
ality and Individual Differences, 15,281-288.
Capara, G. v., Barbaranelli, c., & Comrey, A. L. (1995). Factor analysis of the NEO-
PI inventory and the Comrey Personality Scales in an Italian sample. Person-
ality and Individual Differences, 18, 193-200.
Carmichael, C. M., & McGue, M. (1994). A longitudinal family study of personality
change and stability. Journal of Personality, 62, 1-20.
Caspi, A., & Bern, D. J. (1990). Personality continuity and change across the life
course. In L. A. Pervin (Ed.), Handbook of personality theory and research (pp.
549-575). New York: Guilford.
Cattell, R. B. (1933). Temperament tests: II. Test. British Journal of Psychology, 23,
308-329.
Cattell, R B. (1944). Interpretation of the twelve primary personality factors. Char-
acter and Personality, 13, 55-91.
Cattell, R B. (1947). Confirmation and clarification of the primary personality fac-
tors. Psychometrika, 12, 55-91.
Cattell, R B. (1948). The primary personality factors in women compared with
those in men. British Journal of Psychology, 38, 114-130.
Cattell, R B. (1994). Constancy of global, second-order personality factors over a
twenty-year-plus period. Psychological Reports, 75, 3-9.
Chae, J-H, Piedmont, R. L., Estadt, B. K., & Wicks, R. J. (1995). Personological eval-
uation of Clance's Impostor Phenomenon Scale in a Korean sample. Journal of
Personality Assessment, 65,468-485.
Chen, M. C. (1996). Psychosocial correlates ofprosocial behavior among college students
in Taiwan. Unpublished doctoral dissertation, Loyola College, Baltimore.
Chen, P. Y., Dai, T., Spector, P. E., & lex, s. M. (1997). Relation between negative af-
fectivity and positive affectivity: Effects of judged desirability of scale items
and respondents' social desirability. Journal of Personality Assessment, 69,
183-198.
Church, A. T., & Burke, P. J. (1994). Exploratory and confirmatory tests of the Big
Five and Tellegen's three- and four-dimensional models. Journal of Personality
and Social Psychology, 66, 93-114.
Clance, P. R (1985). The impostor phenomenon: Overcoming the fear that haunts your
success. Atlanta, GA: Peachtree.
Clance, P. R., & Imes, S. A. (1978). The impostor phenomenon in high achieving
women: Dynamics and therapeutic intervention. Psychotherapy Theory, Re-
search and Practice, 15,241-247.
Clark, L. A. (1990). Toward a consensual set of symptom clusters for assessment of
personality disorder. In J. Butcher & C. Spielberger (Eds.), Advances in person-
ality assessment (Vol. 8, pp. 243-266). Hillsdale, NJ: Erlbaum.
Cloninger, C. R. (1987). Asystematic method for clinical description and classifica-
tion of personality disorders. Archives of General Psychiatry, 44, 573-588.
Cloninger, C. R, Svrakic, D. M., & Przybeck, T. R. (1993). Apsychobiological model
of temperament and character. Archives of General Psychiatry, 50, 975-990.
 REFERENCES 259

Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the be-
havioral sciences. Hillsdale, NJ: Erlbaum.
Cohen, S., Doyle, W. J., Skoner, D. P., Fireman, P., Gwaltney, J. M., Jr., & Newsom,
J. T. (1995). State and trait negative affect as predictors of objective and sub-
jective symptoms of respiratory viral infections. Journal of Personality and So-
cial Psychology, 68, 159-169.
Colvin, C. R, & Funder, D. C. (1991). Predicting personality and behavior: A
boundary on the acquaintanceship effect. Journal of Personality and Social Psy-
chology, 60, 884-894.
Costa, P. T., Jr. (1996). Work and Personality: Use of the NEO PI-R in Industrial/Or-
ganizational psychology. Applied Psychology: An International Review, 45,
225-241.
Costa, P. T., Jr., & McCrae, R R (1980a). Influence of extraversion and neuroticism
on subjective well-being: Happy and unhappy people. Journal of Personality
and Social Psychology, 38, 668-678.
Costa, P. T., Jr., & McCrae, R R (1980b). Somatic complaints in males as a function
of age and neuroticism: A longitudinal analysis. Journal of Behavioral Medicine,
3,245-257.
Costa, P. T., Jr., & McCrae, R R (1984). Personality as a lifelong determinant of
well-being. In C. Z. Malatesta & c. E. Izard (Eds.), Emotion in adult development
(pp. 141-157). Beverly Hills, CA: Sage.
Costa, P. T., Jr., & McCrae, R R (1985). Concurrent validation after 20 years: The
implications of personality stability for assessment. In J. N. Butcher & c. D.
Spielberger (Eds.), Advances in personality assessment (pp. 31-54). Hillsdale, NJ:
Erlbaum.
Costa, P. T., Jr., & McCrae, R R (1986). Age, personality, and the Holtzman inkblot
technique. International Journal of Aging and Human Development, 23, 115-125.
Costa, P. T., Jr., & McCrae, R R (1987). Neuroticism, somatic complaints, and dis-
ease: Is the bark worse than the bite? Journal of Personality, 55, 299-316.
Costa, P. T., Jr., & McCrae, R R (1988a). From catalogue to classification: Murray's
needs and the five-factor model. Journal of Personality and Social Psychology, 55,
258-265.
Costa, P. T., Jr., & McCrae, R R (1988b). Personality in adulthood: A six-year lon-
gitudinal study of self-reports and spouse ratings on the NEO Personality In-
ventory. Journal of Personality and Social Psychology, 54, 853-863.
Costa, P. T., Jr., & McCrae, R. R (1989a). Personality continuity and the changes of
adult life. In M. Storandt & G. R VandenBos (Eds.), The adult years: Continuity
and change (pp. 45-77). Washington, DC: American Psychological Association.
Costa, P. T., Jr., & McCrae, R R (1989b). Personality, stress, and coping: Some
lessons from a decade of research. In K. S. Markides & c. L. Cooper (Eds.),
Aging, stress, and health (pp. 269-285). New York: Wiley.
Costa, P. T., Jr., & McCrae, R R (1990). Personality disorders and the five-factor
model. Journal of Personality Disorders, 4, 362-371.
Costa, P. T., Jr., & McCrae, R R (1992a). Normal personality assessment in clinical
practice: The NEO Personality Inventory. Psychological Assessment, 4, 5-13.
260 REFERENCES

Costa, P. T., Jr., & McCrae, R R (1992b). Reply to Eysenck. Personality and Individ-
ual Differences, 13,861-865.
Costa, P. T., Jr., & McCrae, R R (1992c). Revised NED Personality Inventory: Profes-
sional manual. Odessa, FL: Psychological Assessment Resources.
Costa, P. T., Jr., & McCrae, R. R (1992d). Trait psychology comes of age. In T. B. Son-
deregger (Ed.), Nebraska symposium on motivation: Psychology and aging (pp.
169-204). Lincoln: University of Nebraska Press.
Costa, P. T., Jr., & McCrae, R R (1994). "Set like plaster"? Evidence for the stability
of adult personality. In T. F. Heatherton & J. L. Weinberger (Eds.), Can person-
ality change? Washington, DC: American Psychological Association.
Costa, P. T., Jr., & McCrae, R R (1995a). Domains and facets: Hierarchical person-
ality assessment using the revised NEO Personality Inventory. Journal of Per-
sonality Assessment, 64, 21-50.
Costa, P. T., Jr., & McCrae, R. R. (1995b). Primary traits of Eysenck's P-E-N system:
Three- and five-factor solutions. Journal of Personality and Social Psychology, 69,
308-317.
Costa, P. T., Jr., & McCrae, R R (in press). The revised NEO Personality Inventory
(NEO-PI-R). In S. R. Briggs & J. Cheek (Eds.), Personality measures (Vol. 1).
Greenwich, CT: JAI.
Costa, P. T., Jr., & Widiger T. A. (1994). Personality disorders and thefive-factor model of
personality. Washington, DC: American Psychological Association.
Costa, P. T., Jr., McCrae, R. R, & Norris, A. H. (1981). Personal adjustment to aging:
Longitudinal prediction from neuroticism and extraversion. Journal of Geron-
tology, 36, 78-85.
Costa, P. T., Jr., McCrae, R R, & Holland, J. L. (1984). Personality and vocational in-
terests in an adult sample. Journal of Applied Psychology, 69, 390-400.
Costa, P. T., Jr., Zonderman, A. B., McCrae, R R, & Williams, R B., Jr. (1985). Con-
tent and comprehensiveness in the MMPI: An item factor analysis in a normal
adult sample. Journal of Personality and Social Psychology, 48, 925-933.
Costa, P. T., Jr., Busch, C. M., Zonderman, A. B., & McCrae, R R. (1986). Correla-
tions of MMPI factor scales with measures of the five-factor model of person-
ality. Journal of Personality Assessment, 50, 640-650.
Costa, P. T., Jr., McCrae, R. R, & Zonderman, A. B. (1987). Environmental and dis-
positional influences on well-being: Longitudinal follow-up of an American
national sample. British Journal of Psychology, 78, 299-306.
Costa, P. T., Jr., Zonderman, A. B., McCrae, R. R, Comoni-Huntley, J., Locke, B.
Z., & Barbano, H. E. (1987). Longitudinal analyses of psychological well-
being in a national sample: Stability of mean levels. Journal of Gerontology,
42,50-55.
Costa, P. T., Jr., McCrae, R R, & Dye, D. A. (1991). Facet scales for Agreeableness
and Conscientiousness: A revision of the NED Personality Inventory. Person-
ality and Individual Differences, 12, 887-898.
Costa, P. T., Jr., Fagan, P. J., Piedmont, R. 1., Ponticas, Y., & Wise, T. N. (1992). The
five-factor model of personality and sexual functioning in outpatient men and
women. Psychiatric Medicine, 10, 199-215.
Creamer, M., & Campbell, I. M. (1988). The role of interpersonal perception in
dyadic adjustment. Journal of Clinical Psychology, 44, 424-430.
 REFERENCES 261

Crowne, D. P., & Marlowe, D. (1960). Anew scale of social desirability independent
of psychopathology. Journal of Consulting Psychology, 24, 349-354.
Csarny, R. J. (1997). The incremental validity of religious constructs in predicting quality
of life, racism, and sexual attitudes. Unpublished doctoral dissertation, Loyola
College, Baltimore.
Deniston, W. M., & Ramanaiah, N. V. (1993). California Psychological Inventory
and the five-factor model. Psychological Reports, 73,491-496.
Derogatis, L. R. (1993). Brief Symptom Inventory, manual. Minneapolis, MN: National
Computer Systems.
Digman, J. M. (1972). The structure of child personality as seen in behavior ratings.
In R. Dreger (Ed.), Multivariate personality research (pp. 587-611). Baton Rouge,
LA: Claitor's.
Digman, J. M. (1989). Five robust trait dimensions: Development, stability, and util-
ity. Journal of Personality, 57, 195-214.
Digman, J. M. (1990). Personality structure: Emergence of the five-factor model.
Annual Review of Psychology, 41, 417-440.
Digman, J. M. (1996). The curious history of the five-factor model. In J. S. Wiggins
(Ed.). Thefive-factor model ofpersonality. New York: Guilford.
Digman, J. M., & Takemoto-Chock, N. K. (1981). Factors in the natural language of
personality: Re-analysis, comparison, and interpretation of six major studies.
Multivariate Behavioral Research, 16, 149-170.
Duijsens, 1. J., & Diekstra, R. F. W. (1996). DSM-III-R and ICD-lO personality disor-
ders and their relationship with the big five dimensions of personality. Per-
sonality and Individual Differences, 21, 119-133.
Edwards, A. L. (1957). The social desirability variable in personality assessment and re-
search. New York: Dryden.
Edwards, A. L. (1959). Edwards Personal Preference Schedule manual. New York: Psy-
chological Corp.
Edwards, A. L., & Heathers, L. B. (1962). The first factor of the MMPI: Social desir-
ability or ego strength. Journal of Consulting Psychology, 26, 99-100.
Edwards. J. E., & Waters, L. K. (1983). Predicting university attrition: A replication
and extension. Educational and Psychological Measurement, 43, 233-236.
Ellenberger, H. (1970). The discovery of the unconscious. New York: Basic Books.
Ellison, C. W. (1983). Spiritual well-being: Conceptualization and measurement.
Journal of Psychology and Theology, 11, 330-340.
Entwisle, D. R. (1972). To dispel fantasies about fantasy-based measures of achieve-
ment motivation. Psychological Bulletin, 77, 377-39l.
Eysenck, H. J. (1991). Dimensions of personality: 16,5, or 3?-Criteria for a taxo-
nomic paradigm. Personality and Individual Differences, 12, 773-790.
Eysenck, H. J. (1992). Four ways five factors are not basic. Personality and Individual
Differences, 13,667-673.
Eysenck, H. J., & Eysenck, M.W. (1985). Personality and individual differences. Lon-
don, Plenum.
Fagan, P. J., Wise, T. N., Schmidt, C. W., Ponticas, Y., Marshall, R. D., & Costa, P. T.,
Jr. (1991). A comparison of five-factor personality dimensions in males with
sexual dysfunction and males with paraphilia. Journal of Personality Assess-
ment, 57, 434-448.
262 REFERENCES

Fiske, D. W. (1949). Consistency of the factorial structure of personality ratings

from different sources. Journal of Abnormal and Social Psychology, 44, 329-344.
Frances, A J., First, M. B., Widiger, T. A, Miele, G. M., TIlly, S. M., Davis, W. W., &
Pincus, H. A (1991). An a to z guide to DSM-IV conundrums. Journal of Ab-
normal Psychology, 100, 407-412.
Friedman, M., & Rosenman, R. H. (1959). Association of a specific overt behavior
pattern with increases in blood cholesterol, blood clotting time, incidence of
arcus senilis and clinical coronary artery disease. Journal of the American Med-
ical Association, 2208, 828-836.
Funder, D. c., & Colvin, C. R. (1988). Friends and strangers: Acquaintanceship,
agreement, and the accuracy of personality judgment. Journal of Personality
and Social Psychology, 55,149-158.
Furnham, A (1986). Response bias, social desirability, and dissimulation. Personal-
ity and Individual Differences, 7,385-400.
Furnham, A (1994). The big five versus the big four: The relationship between the
Myers-Briggs Type Indicator (MBTI) and NED-PI five factor model of per-
sonality. Personality and Individual Differences, 21, 303-307.
Gerbing, D. W., & Tuley, M. R. (1991). The 16PF related to the five-factor model of
personality: Multiple-indicator measurement versus the a priori scales. Mul-
tivariate Behavioral Research, 26, 271-289.
Goldberg, L. R. (1981). Language and individual differences: The search for uni-
versals in personality lexicons. In L. Wheeler (Ed.), Review of personality and so-
cial psychology (pp. 141-165). Beverly Hills, CA: Sage.
Goldberg, L. R. (1982). From Ace to Zombie: Some explorations in the language of
personality. In C. D. Spielberger & J. N. Butcher (Eds.), Advances in personality
assessment (pp. 203-224). Hillsdale, NJ: Erlbaum.
Goldberg, L. R. (1990). An alternative "description of personality": The big-five
structure. Journal of Personality and Social Psychology, 59,1216-1229.
Goldberg, L. R. (1992). The development of markers for the big-five factor struc-
ture. Psychological Assessment, 4, 26-42.
Goldberg, L. R. (1993). The structure of phenotypic personality traits. American Psy-
chologist, 48, 26-34.
Goldberg, L. R., & Digman, J. M. (1994). Revealing structure in the data: Principles
of exploratory factor analysis. In S. Strack & M. Lorr (Eds.), Differentiating ab-
normal and normal personality (pp. 216-242). New York: Springer.
Goldberg, L. R., & Rosolack, T. K. (1994). The Big Five factor structure as an inte-
grative framework: An empirical comparison with Eysenck's P-E-N model. In
C. F. Halverson, Jr., G. A Kohnstamm, & R. P. Martin (Eds.), The developing
structure of temperament and personality from infancy to adulthood (pp. 7-35).
Hillsdale, NJ: Erlbaum.
Goldstein, M. D., & Strube, M. J. (1994). Independence revisited: The relation be-
tween positive and negative affect in a naturalistic setting. Personality and So-
cial Psychology Bulletin, 20, 57-64.
Gorsuch, R. L. (1983). Factor analysis (2nd ed.). Hillsdale, NJ: Erlbaum.
Gorsuch, R. L. (1988). Psychology of religion. Annual Review of Psychology, 39, 201-221.
 REFERENCES 263

Gorsuch, R L. (1994). Toward motivational theories of intrinsic religious commit-

ment.Journal for the Scientific Study of Religion, 33, 315-325.
Gorsuch, R L. (1997). Exploratory factor analysis: Its role in item analysis. Journal
of Personality Assessment, 68, 523-560.
Gorsuch, R L., & Venable, G. D. (1983). Development of an "age universal" I-E
scale. Journalfor the Scientific Study of Religion, 22, 181-187.
Gottman, J. M. (1979). Marital interaction: Experimental investigations. New York:
Academic Press.
Gough, H. B., & Hall, W. B. (1975). An attempt to predict graduation from medical
school. Journal of Medical Education, 30, 940-950.
Gough, H. B., & Heilbrun, A. B. (1980). The Adjective Check List manual. Palo Alto,
CA: Consulting Psychologists Press.
Gough, H. G. (1987). California Psychological Inventory administrator's guide. Palo
Alto, CA: Consulting Psychologists Press.
Graham, J. R (1990). MMPI 2: Assessing personality and psychopathology. New York:
Oxford University Press.
Green, D. P., Goldman, S. L., & Salovey, P. (1993). Measurement error masks bipo-
larity in affect ratings. Journal ofPersonality and Social Psychology, 64, 1029-1041.
Gruber-Baldini, A. L., Schaie, K. w., & Willis, S. L. (1995). Similarity in married cou-
ples: A longitudinal study of mental abilities and rigidity-flexibility. Journal of
Personality and Social Psychology, 69, 191-203.
Guilford, J. P., & Guilford, R B. (1936). Personality factors 5, E, and M, and their
measurement. Journal ofPersonality, 34, 21-36.
Haan, N., Millsap, R, & Hartka, E. (1986). As time goes by: Change and stability in
personality over fifty years. Psychology and Aging, 1, 220-232.
Hahn, R, & Comrey, A. L. (1994). Factor analysis of the NEO-PI and the Comrey
Personality Scales. Psychological Reports, 75, 355-365.
Harkness, A. R (1992). Fundamental topics in the personality disorders: Candidate
trait dimensions from lower regions of the hierarchy. Psychological Assessment,
4,251-259.
Harkness, A. R, & McNulty, J. L. (1994). The personality psychopathology five
(PSY-5): Issue from the pages of a diagnostic manual instead of a dictionary. In
S. Strack & M. Lorr (Eds.), Differentiating normal and abnormal personality (pp.
291-315). New York: Springer.
Hartmann, H. (1939/1958). Ego psychology and the problem ofadaptation. New York:
International Universities Press.
Heath, A. c., Neale, M. c., Kessler, R. c., Eaves, L. J., & Kendler, K. S. (1992). Evi-
dence for genetic influences on personality from self-reports and informant
ratings. Journal of Personality and Social Psychology, 63, 85-96.
Heaven, P. C. L., Connors, J., & Stones, C. R. (1994). Three or five personality di-
mensions? An analysis of natural language terms in two cultures. Personality
and Individual Differences, 17, 181-189.
Helmes, E., & Jackson, D. N. (1994). Evaluating normal and abnormal personality
using the same set of constructs. In S. Strack & M. Lorr (Eds.), Differentiating
normal and abnormal personality (pp. 341-360). New York: Springer.
 264 REFERENCES

Helson, R., & Moane, G. (1987). Personality change in women from college to mid-
life. Journal at Personality and Social Psychology, 53, 176-186.
Hershberger, S. L., Plomin, R., & Pedersen, N. L. (1995). Traits and metatraits: Their
reliability, stability, and shared genetic influence. Journal at Personality and So-
cial Psychology, 69, 673-685.
Hirschfeld, R. M. A., Klerman, G. L., Clayton, P., Keller, M. B., McDonald-Scott, P.,
& Larkin, B. (1983). Assessing personality: Effects of depressive state on trait
measurement. American Journal of Psychiatry, 140, 695-699.
Hofer, S. M., Hom, J. L., & Eber, H. W. (1997). A robust five-factor structure of the
16PF: Strong evidence from independent rotation and confirmatory factorial
invariance procedures. Personality and Individual Differences, 23, 247-269.
Hofstee, W. K. B., de Raad, D., & Goldberg, L. R. (1992). Integration of the Big Five
and circumplex approaches to trait structure. Journal at Personality and Social
Psychology, 63, 146-163.
Hogan, R. (1991). Personality and personality measurement. In M. D. Dunnette &
L. M. Hough (Eds.), Handbook at industrial and organizational psychology (2nd
ed., pp. 873-919). Palo Alto, CA: Consulting Psychologists Press.
Hogan, R., & Hogan, J. (1989). How to measure employee reliability. Journal at Ap-
plied Psychology, 74, 273-279.
Hogan, R., & Hogan, J. (1992). Hogan Personality Inventory, manual. Tulsa, OK:
Hogan Assessment Systems.
Holden, R. R. (1992). Associations between the Holden Psychological Screening In-
ventoryand the NEO Five-Factor Inventory in a nonclinical sample. Psycho-
logical Reports, 71, 1039-1042.
Hood, R. W. (1975). The construction and preliminary validation of a measure of
reported mystical experience. Journal for the Scientific Study at Religion, 14,
29-41.
Horner, M. S. (1968). Sex differences in achievement motivation and performance in com-
petitive and non-competitive situations. Unpublished doctoral dissertation, Uni-
versity of Michigan, Ann Arbor.
Homer, M. S. (1972). Toward an understanding of achievement-related conflicts in
women. Journal atSocial Issues, 28,157-175.
Horney, K. (1950). Neurosis and human growth. New York: Norton.
Horowitz, L. M. (1996). The study of interpersonal problems: A Leary legacy. Jour-
nal at Personality Assessment, 66, 283-300.
Hough, L. M. (1992). The "big five" personality variables-Construct confusion:
Description versus prediction. Human Performance,S, 139-155.
Hough, L. M., Eaton, N. K., Dunnette, M. D., Kamp, J. D., & McCloy, R. A. (1990).
Criterion-related validities of personality constructs and the effect of response
distortion on those validities. Journal at Applied Psychology, 75,581-595.
Hsu, L. M. (1986). Implications of differences in elevations of K-corrected and non-
K-corrected MMPI T scores. Journal at Consulting and Clinical Psychology, 54,
552-557.
Isaka, H. (1990). Factor analysis of trait terms in everyday Japanese language. Per-
sonality and Individual Differences, 11, 115-124.
 REFERENCES 265

Jackson, D. N. (1989). Basic Personality Inventory manual. Port Huron, MI: Sigma As-
sessment Systems.
Jackson, D. N., Aston, M. c., & Tomes, J. L. (1996). The six-factor model of person-
ality: Facets from the big five. Personality and Individual Differences, 21, 391-402.
Jang, K. L., Livesley, W. J., & Vernon, P. A. (1996a). The genetic basis of personality
at different ages: Across-sectional twin study. Personality and Individual Differ-
ences, 21, 299-30l.
Jang, K. L., Livesley, W. J., & Vernon, P. A. (1996b). Heritability of the Big Five per-
sonality dimensions and their facets: A twin study. Journal of Personality, 64,
577-591.
Jessor, R. (1983). The stability of change: Psychosocial development from adoles-
cence to young adulthood. In D. Magnusson & V. L. Allen (Eds.), Human de-
velopment: An interactional perspective (pp. 321-341). New York: Academic
Press.
John, O. P. (1990). The "Big Five" factor taxonomy: Dimensions of personality in
the natural language and in questionnaires. In L. Pervin (Ed.), Handbook ofper-
sonality theory and research (pp. 66-100). New York: Guilford.
John, O. P., & Robbins, R. W. (1993). Determinants of inter-judge agreement on per-
sonality traits: The Big Five domains, observability, evaluativeness, and the
unique perspective of the self. Journal of Personality, 61, 521-551.
John, O. P., Goldberg, L. R., & Angleitner, A. (1984). Better than the alphabet: Tax-
onomies of personality-descriptive terms in English, Dutch, and German. In
J. J. c. Bonarius, G. L. M. van Heck, & N. G. Smid (Eds.), Personality psychology
in Europe: Theoretical and empirical developments (pp. 83-100). Lisse, Switzer-
land: Swets & Zeitlinger.
John, O. P., Angleitner, A., & Ostendorf, F. (1988). The lexical approach to person-
ality: A historical review of the trait taxonomic research. European Journal of
Personality, 2, 171-203.
Jones, E. E., & Davis, K. E. (1965). From acts to dispositions: The attribution process
in person perception. In L. Berkowitz (Ed.), Advances in experimental social psy-
chology (Vol. 2, pp. 78-103). New York: Academic Press.
Jones, E. E., & Nisbett, R. E. (1972). The actor and the observer: Divergent percep-
tions of the causes of behavior. In E. E. Jones, D. E. Kanouse, H. H. Kelley, R. E.
Nisbett, S. Valins, and B. Weiner (Eds.), Attribution: Perceiving the causes ofbe-
havior (pp. 79-94). Morristown, NJ: General Learning Press.
Joreskog, K. G., & Sorbom, D. (1993). LISREL 8 user's reference guide. Chicago: Sci-
entific Software International.
Juni, S. (1995). Review of the NEO PI-R. In J. Conoley & J. Impara (Eds.), Mental
measurement yearbook (12th ed., pp. 863-868). Lincoln: University of Nebraska
Press.
Kaiser, R. T., & Ozer, D. J. (1997). Emotional stability and goal-related stress. Per-
sonality and Individual Differences, 22, 371-379.
Karney, B. R., & Bradbury, T. N. (1995). The longitudinal course of marital quality
and stability: A review of theory, method, and research. Psychological Bulletin,
118,3-34.
 266 REFERENCES

Karney, B. R., & Bradbury, T. N. (1997). Neuroticism, marital interaction, and the
trajectory of marital satisfaction. Journal of Personality and Social Psychology, 72,
1075-1092.
Katigbak, M. S., Church, A. T., & Akamine, T. X. (1996). Cross-cultural generaliz-
ability of personality dimensions: Relating indigenous and imported di-
mensions in two cultures. Journal of Personality and Social Psychology, 70,
99-114.
Kaufman, A S. (1975). Factor analysis of the WISC-R at 11 age levels between 6%
and 161h years. Journal of Consulting and Clinical Psychology, 43,135-147.
Kaufman, L., & Rousseeuw, P. J. (1990). Finding groups in data: An introduction to
cluster analysis. New York: Wiley.
Keller, M. c., Thiessen D., & Young, R. K. (1996). Mate assortment in dating and
married couples. Personality and Individual Differences, 21, 217-221.
Kosek, R. B. (1996a). Criss-cross ratings of the Big Five personality dimensions as an
index ofmarital satisfaction. Unpublished doctoral dissertation, Loyola College,
Baltimore.
Kosek, R. B. (1996b). The quest for a perfect spouse: Spousal ratings and marital
satisfaction. Psychological Reports, 79, 731-735.
Kozma, A, & Stones, M. J. (1987). Social desirability in measures of subjective well-
being: Asystematic evaluation. Journal of Gerontology, 42, 56-59.
Lanning, K. (1994). Dimensionality of observer ratings on the California adult Q-
Set. Journal of Personality and Social Psychology, 67, 151-160.
Larsen, D. L., Attkisson, C. c., Hargreaves, W. A, & Nguyen, T. D. (1979). Assess-
ment of client/patient satisfaction: Development of a general scale. Evaluation
and Program Planning, 2, 197-207.
Larson, R. (1978). Thirty years of research on the subjective well-being of older
Americans. Journal of Gerontology, 33,109-125.
Leary, T. (1957). Interpersonal diagnosis ofpersonality. New York: Ronald.
Lester, D., Haig, c., & Monello, R. (1989). Spouses' personality and marital satis-
faction. Personality and Individual Differences, 10, 253-254.
Levin, J., & Montag, I. (1991). Relationship between the Basic Personality Inventory
and the NEO-Personality Inventory in a nonpatient sample. Psychological Re-
ports, 69, 1176-1178.
Levin, J., & Montag, I. (1994). The five factor model and psychopathology in non-
clinical samples. Personality and Individual Differences, 17, 1-7.
Linehan, M. M., & Nielsen, S. L. (1983). Social desirability: Its relevance to the mea-
surement of hopelessness and suicidal behavior. Journal ofConsuiting and Clin-
ical Psychology, 51,141-143.
Livesley, W. J., West, M., & Tanney, A (1985). Historical comment on the DSM-III
schizoid and avoidant personality disorders. American Journal of Psychiatry,
142,1344-1347.
Livesley, W. J., Jackson, D. N., & Schroeder, M. L. (1992). Factorial structure of traits
delineating personality disorders in clinical and general population samples.
Journal of Abnormal Psychology, 101, 432-440.
Livesley, W. J., Schroeder, M. L., Jackson; D. N., & Jang, K. L. (1994). Categorical dis-
tinctions in the study of personality disorder: Implications for classification.
Journal of Abnormal Psychology, 103, 6-17.
 REFERENCES 267

Locke, H. J., & Wallace, K. M. (1959). Short marital-adjustment and prediction tests:
Their reliability and validity. Marriage and Family Living, 21, 251-255.
Loevinger, J. (1994). Has psychology lost its conscience? Journal of Personality As-
sessment, 62, 2-8.
Lorr, M. (1996). The Interpersonal Circle as a heuristic model for interpersonal re-
search. Journal of Personality Assessment, 66, 234-239.
Lorr, M., & Strack, S. (1993). Some NEO PI five-factor personality profiles. Journal
of Personality Assessment, 60,91-99.
Lorr, M., Youniss, R P., & Kluth, C. (1992). The Interpersonal Style Inventory and
the five-factor model. Journal of Clinical Psychology, 48, 202-206.
Lubin, B., Larsen, R. M., & Matarazzo, J. D. (1984). Patterns of psychological test
usage in the United States: 1935-1982. American Psychologist, 39, 451-454.
Lyons, J. 5., Howard, K. I., O'Mahoney, M. T., & Ush, J. D. (1997). The measurement
and management of clinical outcomes in mental health. New York: Wiley.
Marsh, H. w., Balla, J. R, & McDonald, R P. (1988). Goodness-of-fit indexes in con-
firmatory factor analysis: The effect of sample size. Psychological Bulletin, 103,
391-410.
Marshall, G. N., Wortman, C. 8., Vickers, R R, Jr., Kusulas, J. W., & Hervig, L. K.
(1994). The five-factor model of personality as a framework for personality-
health research. Journal of Personality and Social Psychology, 67, 278-286.
McAdams, D. P. (1992). The five-factor model in personality: A critical appraisal.
Journal of Personality, 60, 329-361.
McClelland, D. c., Atkinson, J. w., Clark, R w., & Lowell, E. L. (1953). The achieve-
ment motive. New York: Appleton-Century-Crofts.
McCrae, R R (1982). Consensual validation of personality traits: Evidence from
self-reports and ratings. Journal ofPersonality and Social Psychology, 43, 292-303.
McCrae, R R (1986). Well-being scales do not measure social desirability. Journal of
Gerontology, 41, 390-393.
McCrae, R R (1989). Why I advocate the five-factor model: Joint factor analyses of
the NEO-PI with other instruments. In D. Buss & N. Cantor (Eds.), Personal-
ity psychology: Recent trends and emerging directions (pp. 237-245). New York:
Springer-Verlag.
McCrae, R R (1990). Traits and trait names: How well is Openness represented in
natural languages? European Journal of Personality, 4, 119-129.
McCrae, R R (1993). Moderated analyses of longitudinal personality stability. Jour-
nal of Personality and Social Psychology, 65, 577-585.
McCrae, R R (1993-1994). Openness to experience as a basic dimension of per-
sonality. In K. S. Pope & J. L. Singer (Eds.), Imagination, cognition, and personal-
ity: Consciousness in theory-research-clinical practice (Vol. 13, pp. 39-55).
Amityville, NY: Baywook.
McCrae, R R (1994a). The counterpoint of personality assessment: Self-reports and
observer ratings. Assessment, 1, 151-164.
McCrae, R R. (1994b). Openness to experience: Expanding the boundaries of fac-
tor V. European Journal of Personality, 8, 251-272.
McCrae, R. R. (1994c). Psychopathology from the perspective of the five-factor
model. In S. Strack & M. Lorr (Eds.), Differentiating normal and abnormal per-
sonality (pp. 26-39). New York: Springer.
 268 REFERENCES

McCrae, R R, & Costa, P. T., Jr. (1983a). Psychological maturity and subjective
well-being: Toward a new synthesis. Developmental Psychology, 19, 243-248.
McCrae, R R, & Costa, P. T., Jr. (1983b). Social desirability scales: More substance
than style. Journal of Consulting and Clinical Psychology, 51, 882-888.
McCrae, R R, & Costa, P. T., Jr. (1985a). Comparison of EPI and Psychoticism scales
with measures of the five-factor model of personality. Personality and Individ-
ual Differences, 6, 587-597.
McCrae, R. R, & Costa, P. T., Jr. (1985b). Openness to experience. In R Hogan & W. H.
Jones (Eds.), Perspectives in personality <Vol. I, pp. 145-172). Greenwich, CT: JAI.
McCrae, R R, & Costa, P. T., Jr. (1987). Validation of the five-factor model of per-
sonality across instruments and observers. Journal of Personality and Social Psy-
chology, 52, 81-90.
McCrae, R R, & Costa, P. T., Jr. (1988). Psychological resilience among widowed
men and women: A 10-year follow-up of a national sample. Journal of Social Is-
sues, 44,129-142.
McCrae, R R, & Costa, P. T., Jr. (1989a). Reinterpreting the Myers-Briggs Type In-
dicator from the perspective of the five-factor model of personality. Journal of
Personality, 57, 17-40.
McCrae R R, & Costa, P. T., Jr. (1989b). Rotation to maximize the construct valid-
ity of factors in the NED Personality Inventory. Multivariate Behavioral Re-
search, 57, 17-40.
McCrae, R R, & Costa, P. T., Jr. (1989c). The structure of interpersonal traits: Wig-
gins's circumplex and the five-factor model. Journal of Personality and Social
Psychology, 56, 586-595.
McCrae, R R, & Costa, P. T., Jr. (1992). Discriminant validity of the NED-PIR facet
scales. Educational and Psychological Measurement, 52, 229-237.
McCrae, R R, & Costa, P. T., Jr. (1994). Does Lorr's Interpersonal Style Inventory
measure the five-factor model? Personality and Individual Differences, 16,
195-197.
McCrae, R R, & Costa, P. T., Jr. (1995). Positive and negative valence within the
five-factor model. Journal of Research in Personality, 29, 443-460.
McCrae, R R., & Costa, P. T., Jr. (1996). Toward a new generation of personality the-
ories: Theoretical contexts for the five-factor model. In J. S. Wiggins (Ed.), The
five-factor model of personality: Theoretical perspectives (pp. 51-87). New York:
Guilford.
McCrae, R R, & Costa, P. T., Jr. (1997). Personality trait structure as a human uni-
versal. American Psychologist, 52, 509-516.
McCrae, R R, & John, 0. P. (1992). An introduction to the five-factor model and its
applications. Journal of Personality, 60, 175-215.
McCrae, R R, Costa, P. T., Jr., & Busch, C. M. (1986). Evaluating comprehensive-
ness in personality systems: The California Q-Set and the five factor model.
Journal of Personality, 54, 430-446.
McCrae, R R, Costa, P. T., Jr., Dahlstrom, W. G., Barefoot, J. c., Siegler, I. c., &
Williams, R B., Jr. (1989). A caution on the use of the MMPI K-correction in re-
search on psychosomatic medicine. Psychosomatic Bulletin, 51, 58-65.
 REFERENCES 269

McCrae, R. R., Costa, 1'. T., Jr., & Piedmont, R. L. (1993). Folk concepts, natural lan-
guage, and psychological constructs: The California Personality Inventory
and the five-factor model. Journal of Personality, 61, 1-26.
McCrae, R. R., Costa, 1'. T., Jr., Piedmont, R. L., Chae, J-H, Caprara, G. v., Barbaranelli,
c., Marusic, 1, & Bratko, D. (1996, November). Personality development from col-
lege to mid-lifo: A cross-cultural comparison. Paper presented at the 49th annual
meeting of the Gerontological Society of America, Washington, DC.
McCrae, R. R., Costa, 1'. T., Jr., Stone, S. v., & Fagan, P. J. (1996, August). Some reasons
for disagreement between spouses in personality assessment. Poster presented at
the 104th annual convention of the American Psychological Association,
Toronto, Ontario, Canada.
McCrae, R. R., Zonderman, A. B., Costa, P. T., Jr., Bond, M. H., & Paunonen, S. v.
(1996). Evaluating replicability of factors in the Revised NED Personality In-
ventory: Confirmatory factor analysis and Procrustes rotation. Journal of Per-
sonality and Social Psychology, 70, 552-566.
Meehl,1'. E. (1978). Theoretical risks and tabular asterisks: Sir Karl, Sir Ronald, and
the slow progress of soft psychology. Journal of Consulting and Clinical Psychol-
ogy, 46, 806-834.
Megargee, E. 1 (1972). The California Psychological Inventory handbook. San Francisco:
Jossey-Bass.
Megargee, E. I., & Parker, G. V. C. (1968). An exploration of the equivalent of Mur-
rayian needs as assessed by the Adjective Check List, the TAT, and Edwards
Personal Preference Schedule. Journal of Clinical Psychology, 24, 47-51.
Miller, T. (1991). The psychotherapeutic utility of the five-factor model of person-
ality: A clinician's experience. Journal of Personality Assessment, 57, 415-433.
Millon, T. (1990). Towards a new personology: An evolutionary model. New York: Wiley.
Mischel, W. (1968). Personality assessment. New York: Wiley.
Mooradian, T. A., & Nezlek, J. B. (1996). Comparing the NEO-FFI and Saucier's
mini-markers as measures of the Big Five. Personality and Individual Differ-
ences, 21, 213-215.
Mount, M. K., & Barrick, M. R. (1995). The big five personality dimensions: Impli-
cations for research and practice in human resources management. Research in
Personnel and Human Resources Management, 13, 153-200.
Mount, M. K., Barrick, M. R., & Strauss, J. 1'. (1994). Validity of observer ratings of
the Big Five personality factors. Journal of Applied Psychology, 79, 272-280.
Murray, H. A. (1938). Explorations in personality. New York: Oxford University
Press.
Muten, E. (1991). Self-reports, spouse ratings, and psychophysiological assessment
in a behavioral medicine program: An application of the five-factor model.
Journal ofPersonality Assessment, 57, 574-583.
Myers, I. B., & McCaulley, M. H. (1985). A guide to the development and use of the
Myers-Briggs Type Indicator. Palo Alto, CA: Consulting Psychologists Press.
Narayanan, L., Shanker, M., & Levine, E. L. (1995). Personality structure: A culture-
specific examination of the five-factor model. Journal of Personality Assessment,
64,51-62.
270 REFERENCES

Nicholson, R. A, & Hogan, R. (1990). The construct validity of social desirability.

American Psychologist, 45, 290-292.
Norman, W. T. (1963). Toward an adequate taxonomy of personality attributes:
Replicated factor structure in peer nomination personality ratings. Journal of
Abnormal and Social Psychology, 66, 574-583.
Ogles, B. M., Lambert, M. J.. & Masters, K. S. (1996). Assessing outcome in clinical
practice. Boston: Allyn & Bacon.
Ones, D. S. (1993). The construct validity of integrity tests. Unpublished doctoral dis-
sertation, University of Iowa, Iowa City.
Ormel, J., & Wohlfarth, T. (1991). How neuroticism, long-term difficulties, and life
situation change influence psychological distress: A longitudinal model. Jour-
nal of Personality and Social Psychology, 60, 744-755.
Ozer, D. J., & Riese, S. P. (1994). Personality assessment. Annual Review ofPsychol-
ogy, 45, 357-388.
Panter, A T., Tanaka, J. S., & Hoyle, R. H. (1994). Structural models for multimode
designs in personality and temperament research. In C. F. Halverson, G. A
Kohnstamm, & R. P. Martin (Eds.), The developing structure of temperament and
personality from infancy to adulthood (pp. 111-138). Hillsdale, NJ: Erlbaum.
Parker, J. D. A, Bagby, R. M., & Summerfeldt, L. J. (1993). Confirmatory factor
analysis of the Revised NEO Personality Inventory. Personality and Individual
Differences, 15,463-466.
Paunonen, S. v., Jackson, D. N., Trzebinski, J., & Forsterling, F. (1992). Personality
structure across cultures: Amulti-method evaluation. Journal of Personality and
Social Psychology, 62, 447-456.
Peabody, D., & Goldberg, L. R. (1989). Some determinants of factor structures from
personality-trait descriptors. Journal of Personality and Social Psychology, 57,
552-567.
Piedmont, R. L. (1988a). An interactional model of achievement motivation and
fear of success. Sex Roles, 19,467-490.
Piedmont, R. L. (1988b). The relationship between achievement motivation, anxi-
ety, and situational characteristics on performance on a cognitive task. Journal
of Research in Personality, 22, 177-187.
Piedmont, R. L. (1993). A longitudinal analysiS of burnout in the health care setting:
The role of personal dispositions. Journal ofPersonality Assessment, 61, 457-473.
Piedmont, R. L. (1994). Validation of the rater form of the NEO PI-R for college stu-
dents: Towards a paradigm for measuring personality development. Assess-
ment, 1,259-268.
Piedmont, R. L. (1995). Another look at fear of success, fear of failure, and test anxi-
ety: Amotivational analysis using the five-factor model. Sex Roles, 32, 139-158.
Piedmont, R. L. (1996a). Cracking the plaster cast: Big 5 personality change due to out-
patient drug rehabilitation counseling. Paper presented at the 67th annual meet-
ing of the Eastern Psychological Association, Philadelphia.
Piedmont, R. L. (1996b, August). Strategies for using the five-factor model in reli-
gious research. In R. L. Piedmont (Chair), The fivefactor model and its value for
religious research. Symposium presented at the annual convention of the Amer-
ican Psychological Association, Toronto, Ontario, Canada.
 REFERENCES 271

Piedmont, R L. (1997a, August). Development and validation of the Spiritual Tran-

scendence Scale: A measure of spiritual experience. In J. E. G. Williams
(Chair), Spiritual experience and the five-foetor model of personality. Symposium
presented at the annual convention of the American Psychological Associa-
tion, Chicago.
Piedmont, R L. (1997b). Test review: The NEO PI-R Newsnotes, 32(4), 3-4.
Piedmont, R L., & Chae, J-H. (1997). Cross-cultural generalizability of the five-fac-
tor model of personality: Development and validation of the NEO PI-R for
Koreans. Journal of Cross-Cultural Psychology, 28,131-155.
Piedmont, R. L., & Ciarrocchi, J. W. (in press). Psychometric utility of the NEO PI-R
in an outpatient, drug rehabilitation context. Psychology of Addictive Behavior.
Piedmont, R L., & McCrae, R R (1998). Are validity scales valid? Evidencefrom self-
reports and observer ratings in a volunteer sample. Manuscript under review. Un-
published manuscript, Baltimore, MD.
Piedmont, R L., & Piedmont, R I. (1996). Couples critical incidents check list, manual.
Baltimore: Author.
Piedmont, R L., & Weinstein, H. P. (1993). A psychometric evaluation of the new
NEO-PI-R facet scales for Agreeableness and Conscientiousness. Journal of
Personality Assessment, 60, 302-318.
Piedmont, R L., & Weinstein, H. P. (1994). Predicting supervisor ratings of job per-
formance using the NEO Personality Inventory. Journal of Psychology, 128,
255-265.
Piedmont, R L., DiPlacido, J., & Keller, W. (1989). Assessing gender-related differ-
ences in achievement orientation using two different achievement scales.
Journal of Personality Assessment, 53, 229-238.
Piedmont, R L., McCrae, R R, & Costa, P. T., Jr. (1991). Adjective check list scales
and the five-factor model. Journal of Personality and Social Psychology, 60,
630-637.
Piedmont, R L., McCrae, R R, & Costa, P. T., Jr. (1992). An assessment of the Ed-
wards Personal Preference Schedule from the perspective of the five-factor
model. Journal of Personality Assessment, 58, 67-78.
Pincus, A. L. (1994). The interpersonal circumplex and the interpersonal theory:
Perspectives on personality and its pathology. In S. Strack & M. Lorr (Eds.),
Differentiating normal and abnormal personality (pp. 114-136). New York:
Springer.
Pincus, A. L., & Wiggins, J. S. (1990). Interpersonal problems and conceptions of
personality disorders. Journal of Personality Disorders, 4, 342-352.
Plomin, R, & Daniels, D. (1987). Why are children in the same family so different
from one another? Behavioral and Brain Sciences, 10, 1-16.
Plomin, R, Chipuer, H. M., & Loehlin, J. C. (1990). Behavioral genetics and per-
sonality. In L. A. Pervin (Ed.), Handbook of personality: Theory and research (pp.
225-243). New York: Guilford.
Ptacek, J. T., & Dodge, K. L. (1995). Coping strategies and relationship satisfaction
in couples. Personality and Social Psychology Bulletin, 21, 76-84.
Reynolds, W. M., & Kobak, K. A. (1995). Hamilton Depression Inventory, manual.
Odessa, FL: Psychological Assessment Resources.
 272 REFERENCES

Rodgerson, T. E. (1994). The relation between situation, personality, and religious prob-
lem-solving in the prediction of burnout among American Baptist clergy. Unpub-
lished doctoral dissertation, Loyola College, Baltimore.
Russell, R. J. H, & Wells, P. A (1994). Predictors of happiness in married couples.
Personality and Individual Differences, 17,313-321.
Saklofske, D. H., Kelly, I. W., & Janzen, B. L. (1995). Neuroticism, depression, and
depression proneness. Personality and Individual Differences, 18, 27-3l.
Saucier, G. (1994). Mini-markers: A brief version of Goldberg's unipolar Big-Five
markers. Journal of Personality Assessment, 63, 506-516.
Scheier, M. E, Buss, A H, & Buss, D. M. (1976). Self-consciousness, self-report of
aggressiveness, and aggression. Journal of Research in Personality, 44, 637-644.
Schinka, J. A (1985). Personal Problems Check Listfor Adults. Odessa, FL: Psycholog-
ical Assessment Resources.
Schmeck, R. R., & Grove, E. (1979). Academic achievement and individual differ-
ences in the learning process. Applied Psychological Measurement, 3, 43-49.
Sch6nemann, P. H. (1966). A generalized solution of the orthogonal Procrustes
problem. Psychometrika, 31, 1-10.
Schroeder, D. H, & Costa, P. T., Jr. (1984). Influence of life event stress on physical
illness: Substantive effects or methodological flaws? Journal of Personality and
Social Psychology, 46, 853-863.
Schroeder, M. L., Wormworth, J. A, & Livesley, W. J. (1992). Dimensions of per-
sonality disorder and their relationships to the Big Five dimensions of per-
sonality. Psychological Assessment, 4, 47-53.
Schinka, J. A, Kinder, B. N., and Kremer, T. (1997). Research validity scales for the
NEO PI-R: Developmental and initial validation. Journal of Personality Assess-
ment, 68, 127-138.
Shock, N. W., Greulick, R. C, Andres, R., Arenberg, D., Costa, P. T., Jr., Lakatta,
E. G., & Tobin, J. D. (1984). Normal human aging: The Baltimore Longitudinal
Study of Aging (NIH Publication No. 84-2450). Bethesda, MD: National Insti-
tutes of Health.
Siegler, I. C, Zonderman, A B., Barefoot, J. C, Williams, R. B., Jr., Costa, P. T., Jr., &
McCrae, R. R. (1990). Predicting personality in adulthood from college MMPI
scores: Implications for follow-up studies in psychosomatic medicine. Psy-
chosomatic Medicine, 52, 644-652.
Silva, E, Avia, D., Sanz, J., Martinez-Arias, M. R., Grana, J. L., & Sanchez-Bemardos,
L. (1994). The five-factor model-I. Contributions to the structure of the NEO-
PI. Personality and Individual Differences, 17,741-753.
Silver, R. J., & Sines, L. K. (1962). Diagnostic efficiency of the MMPI with and with-
out the K correction. Journal of Clinical Psychology, 18,312-314.
Silverstein, A B. (1982). Factor structure of the Wechsler Adult Intelligence
Scale-Revised. Journal of Consulting and Clinical Psychology, 50, 661-664.
Snyder, D. K. (1982). Advances in marital assessment: Behavioral, communica-
tions, and psychometric approaches. In C D. Spielberger & J. N. Butcher
(Eds.), Advances in personality assessment (Vol. I, pp. 169-201). Hillsdale, NJ:
Erlbaum.
 REFERENCES 273

Snyder, M. (1983). The influence of individuals on situations: Implications for un-

derstanding the links between personality and social behavior. Journal of Per-
sonality, 51, 497-516.
Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970). State-Trait anxiety inven-
tory. Palo Alto, CA: Consulting Psychologists Press.
Spirrison, C. L. (1994). Factorial hue and cry: Comments on Jane Loevinger's "Has
psychology lost its conscience?" Journal of Personality Assessment, 63, 579-583.
Stagner, R. (1937). Psychology ofpersonality. New York: McGraw-Hill.
Steers, K. M. (1975). Task-goal attributes, nAchievement, and supervisory perfor-
mance. Organizational Behavior and Human Performance, 13, 392-403.
Stout, C. E. (1997). Psychological assessment in managed care. New York: Wiley.
Strack, S., & Lorr, M. (1994). Differentiating normal and abnormal personality. New
York: Springer.
Taylor, R. M., & Morrison, L.P. (1984). Taylor-Johnson Temperament Analysis manual.
Los Angeles, CA: Psychological Publications.
Tellegen, A (1982). Brief manual of the Multidimensional Personality Questionnaire.
Unpublished manuscript, University of Minnesota, Minneapolis.
Tellegen, A (1988). The analysis of consistency in personality assessment. Journal of
Personality, 56, 621-663.
Tellegen, A. (1993). Folk concepts and psychological concepts of personality and
personality disorder. Psychological Inquiry, 4,122-130.
Tellegen, A, & Waller, N. G. (in press). Exploring personality through test con-
struction: Development of the Multidimensional Personality Questionnaire.
In S. R. Briggs & J. M. Cheek (Eds.), Personality measures: Development and eval-
uation (Vol. 1). Greenwich, CT: JAI.
Tharp, R. G. (1963). Psychological patterning in marriage. Psychological Bulletin, 60,
97-117.
Thurstone, L. L. (1951). The dimensions of temperament. Psychometrika, 16, 11-20.
Tresemer, D. (1976). The cumulative record of research on fear of success. Sex Roles,
2,217-235.
Trull, T. J. (1992). DSM-III-R personality disorders and the five-factor model of per-
sonality: An empirical comparison. Journal of Abnormal Psychology, 101,553-560.
Trull, T. J., & Sher, K. J. (1994). Relationship between the five-factor model of per-
sonality and Axis I disorders in a nonc1inical sample. Journal of Abnormal Psy-
chology, 103, 350-360.
Trull, T. J., Useda, D., Costa, P. T., Jr., & McCrae, R. R. (1995). Comparison of the
MMPI-2 Personality Psychopathology Five (PSY-5), the NEO-PI, and the NEO
PI-R. Psychological Assessment, 7, 508-516.
Tupes, E. c., & Christal, R. E. (1961). Recurrent personalityfactors based on trait ratings
(USAF ASD Technical Report No. 61-97). Lackland Air Force Base, TX: U.S.
Air Force.
Vassend, 0., & Skrondal, A. (1995). Factor analytic studies of the NEO Personality
Inventory and the five-factor model: The problem of high structural com-
plexity and conceptual indeterminacy. Personality and Individual Differences, 19,
135-147.
274 REFERENCES

Viken, R. J., Rose, R. J., Kaprio, J., & Koskenvuo, M. (1994). A developmental ge-
netic analysis of adult personality: Extraversion and Neuroticism from 18 to
59 years of age. Journal of Personality and Social Psychology, 66, 722-730.
Wanous, J. P., Reichers, A E., & Hudy, M. J. (1997). Overall job satisfaction: How
good are single-item measures? Journal of Applied Psychology, 82, 247-252.
Watkins, C. E., & Campbell, V. L. (1989). Personality assessment and counseling
psychology. Journal of Personality Assessment, 53, 296-307.
Watson, D., & Clark, L. A (1991). Self- versus peer ratings of specific emotional
traits: Evidence of convergent and discriminant validity. Journal of Personality
and Social Psychology, 60, 927-940.
Watson, D., & Clark, L. A (1992). On traits and temperament: General and specific
factors of emotional experience and their relation to the five-factor model.
Journal ofPersonality, 60, 441-476.
Watson, D., Clark, L. A, & Harkness, A R. (1994). Structures of personality and
their relevance to psychopathology. Journal ofAbnormal Psychology, 103, 18-31.
Webb, E. (1915). Character and intelligence. British Journal of Psychology Mono-
graphs, 1(3), 1-99.
Welsh, G. S. (1975). Creativity and intelligence: A personality approach. Chapel Hill:
University of North Carolina, Institute for Research in Social Sciences.
Wernimont, P. F., & Campbell, J. P. (1968). Signs, samples, and criteria. Journal of Ap-
plied Psychology, 57, 120-128.
Widiger, T., & Corbitt, E. M. (1994). Normal versus abnormal personality from the
perspective of the DSM. In S. Strack & M. Lorr (Eds.), Differentiating normal
and abnormal personality (pp. 158-175). New York: Springer.
Widiger, T. A, & Costa, P. T., Jr. (1994). Personality and personality disorders. Jour-
nal of Abnormal Psychology, 103, 78-91.
Widiger, T. A, & Frances, A (1985). The DSM-III personality disorders: Perspec-
tives from psychology. Archives of General Psychiatry, 42, 615-623.
Widiger, T. A, & Frances, A (1994). Toward a dimensional model for the personal-
ity disorders. In P. T. Costa, Jr, & T. A Widiger (Eds.), Personality disorders and
the five-factor model ofpersonality (pp. 19-40). Washington, DC: American Psy-
chological Association.
Widiger, T. A, & Shea, T. (1991). Differentiation of Axis I and Axis II disorders. Jour-
nal of Abnormal Psychology, 100, 399-406.
Widiger, T. A, & Trull, T. J. (1992). Personality and psychopathology: An applica-
tion of the five-factor model. Journal ofPersonality, 60, 363-393.
Wiggins, J. S. (1979). Personality and prediction: Principles of personality assessment.
Reading, MA: Addison-Wesley.
Wiggins, J. S. (1980). Circumplex models of interpersonal behavior. In L. Wheeler
(Ed.), Review of personality and social psychology (Vol. 1, pp. 265-293). Beverly
Hills, CA: Sage.
Wiggins, J. S. (1995). Interpersonal Adjective Scales: Professional manual. Odessa, FL:
Psychological Assessment Resources.
Wiggins, J. S. (1996). An informal history of the Interpersonal Circumplex Tradi-
tion. Journal of Personality Assessment, 66, 217-233.
REFERENCES 275

Wiggins, J. 5., & Pincus, A L. (1989). Conceptions of personality disorders and di-
mensions of personality. Psychological Assessment: A Journal of Consulting and
Clinical Psychology, 1, 305-316.
Wiggins, J. 5., & Trapnell, P.D. (1996). A dyadic-interactional perspective on the
five-factor model. In J. S. Wiggins (Ed.), Thefive-foctor model ofpersonality: The-
oretical perspectives (pp. 88-162). New York: Guilford.
Winch, R. F. (1958). A study of complementary needs. New York: Harper.
Worthington, D. L., & Schlottman, R. S. (1986). The predictive validity of subtle and
obvious empirically derived psychological test items under faking condition.
Journal of Personality Assessment, 50, 171-181.
Wrobel, T. A, & Lachar, D. (1982). Validity of the Wiener subtle and obvious scales
for the MMPI: Another example of the importance of inventory-item content.
Journal of Consulting and Clinical Psychology, 50, 469-470.
Zonderman, A S., Costa, P. T., Jr., & McCrae, R. R. (1989). Depression as a risk for
cancer morbidity and mortality in a nationally representative sample. Journal
of the American Medical Association, 262,1191-1195.
Zonderman, A 8., Herbst, J. H., Schmidt, c., Jr., Costa, P. T., Jr., & McCrae, R. R.
(1993). Depressive symptoms as a non-specific, graded risk for psychiatric di-
agnoses. Journal of Abnormal Psychology, 102, 544-552.
 AUTHOR INDEX

Akamine, T.X., 73, 213, 266 Benet, v., 71, 219, 256 Butcher, J. N., 80, 114,257,
Aldenderfer, M.S., 114,255 Ben-Porath, YS., 69, 79, 80, 258,259,262,272
Al1port, G. W., 20, 21, 22, 255,256
23,24~26,27, 122,255 Benson, P. L., 234, 256 Campbell, I. M., 175, 260
Almagor, M., 219, 255 Bentler, P. M., 224, 256 Campbell. J. P., 206, 274
Andres, R., 272 Bergeman, e. 5., 39, 72, 256 Campbel\, V. L., 48, 274
Andrews, F. M., 117,255 Berne, S. L.. 49, 256 Cantor, N., 77, 257, 267
Angleitner, A., 22, 73, 265 Bigelow, G. E., 59, 257 Cantril, H., 117,257
Arbisi, P. A., 79, 255 Blashfield, R. K., 114, 255 Capara, G. v., 73, 213, 258
Arenberg, D., 272 Block, J., 7, 41, 46, 71, 80, Carmichael. e. M., 39, 258
Aston, M.e., 265 213 Caspi, A., 5, 258
Atkinson, J. w., 120, 267 Block, J. H., 134 Cattell, R. B., 15,17,24,25,
Attkisson, e. e., 242, 266 Bond, M. H., 60, 73, 256, 26,257,258
Avia, M.D., 73, 255, 272 269 Chae, J-H, 42, 43, 45, 73,
Bonnett, D. G., 224 213,228,258,269,271
Bagby, R.M., 40, 60, 223, Borgogni. L., 73, 258 Chen, M. e., 238, 258
246,255,270 Borkenau, P, 41. 54, 73, 223, Chen, P. Y, 215, 258
Bailey, K.D., 17, 113,255 256,257 Chipuer, H. M., 72, 256, 271
Bakan, D., 116, 255 Botwin, M., 31, 257 Christal, R. E., 25, 273
Balla, J.R., 224, 267 Bouchard, T. J., 47, 72, 257 Church, A. T., 73, 213, 224,
Bandura, A., 76, 255 Boyle, G. J.. 25, 257 225,258,266
Barbano, H.E., 260 Bradbury, T. M., 171, 172, Ciarrocchi, J. w., 59, 60, 61,
Barbaranelli, e., 73, 213, 265,266 62,245,248,271
258,269 Bratko, D., 269 Clance, P. R., 143,258
Barefoot, J.e., 268, 272 Brooner, R. K., 59, 257 Clark, L. A., 54, 63, 67, 118,
Barlow, D.H., 211, 256 Burke, P. J., 224, 225, 258 258,274
Barrick, M. R., 31, 46, 83, Buros, O. K., 47, 257 Clark, R. w., 120, 267
84,122,155,206,256, Busch, e. M., 61, 260, 268 Clayton, P., 264
269 Buss, A. H., 206, 272 Cloninger, e. R., 15,258
Barry, W.A., 173, 256 Buss, D. M., 72,110,172, Cohen, J., 180,237,259
Bern, D.J., 5, 258 173,178,206,257,267, Cohen, P., 180,237, 259
Bendig, A. w., 120, 256 272 Cohen, S., 118, 259

277
 278 AUTHOR INDEX

Colvin, C R., 54, 259, 262 Ellenberger, H., 20, 261 Gwaltney, J. M., 259
Comrey, A. L., 71, 73, 213, Ellison, C W, 234, 261
257,258,263 Entwisle, D. R., 48, 261 Haan, N., 7, 39, 263
Connors, J., 43, 263 Erickson, J. A., 234, 256 Hahn, R., 71, 263
Cook-Flannagan, C, 257 Estadt, B. K., 258 Haig, C, 172, 266
Corbitt, E. M., 57, 274 Eysenck, H. J., 219. 220, Hall, W B., 120,263
Cornoni-Huntley, J., 260 221,261 Hargreaves, W A., 242, 266
Costa, P. T., Jr.. 11. 30, 31, Eysenck, M. W, 219, 261 Harkness, A. R., 15,40, 63,
35,36,38,39.40.41,42, 67,246,255,263,274
43,45,49,51,53,56,59, Fagan, P. J., 62, 109, 260, Hartka, E., 7, 39, 263
60.61,62,63,67,68,72, 261,269 Hartmann. H., 76, 263
73,74,76,80.81,84.85, Fireman, P., 259 Hayes, S. C, 211, 256
86,87.92,97,109,116. First. M. B., 2,15,18,22.24, Heath, A. C, 39, 72, 263
117, 118, 119, 122, 123, 27,31,38.40,46.49,59. Heathers, L. B., 80, 261
157,180,213,214.217, 60.62,68,99,103.105, Heaven, P. C L., 43, 73,
220,221,222,229,244, 107,108.117,156, 166, 263
246,256,257,259,260, 168,175,177,180,182, Heilbrun, A. B., 47, 263
261,268,269,271,272, 192.203,204,206,209, Helmes, E., 63, 263
273,274,275 215,220.221.225,229, Helson, R., 7, 264
Craik, K. H., 110,257 232,234,240,246,249, Herbst, J. H., 59, 61, 84, 257,
Creamer, M., 175, 260 251, 262 275
Crowne, D. P., 81, 261 Fiske, D. W, 25, 262 Hershberger, S. L., 72, 264
Csarny, R. J.. 238. 261 Fleeson. W.. 257 Hervig, L. K., 43, 267
Forsterling, F.. 43, 270 Hirschfeld, R. M. A., 40,
Dahlstrom, W G., 114,257, Frances. A. J., 58, 63, 64, 264
268 262,274 Hofer, S. M., 25, 264
Dai, T., 215, 258 Friedman. M., 114,262 Hofstee, W K. B., 51, 67,
Daniels, D., 74, 271 Funder, D. C, 54, 259, 262 114,115,234,264
Davis, K. E., 55, 265 Furnham, A., 83, 262 Hogan, J., 71, 122,264
Davis, W. W., 262 Hogan, R., 71, 81, 83, 122,
de Raad, D., 51, 114,234, Gerbing, D. W, 262 205,206,264,268,270
264 Goldberg, L. R .. 25, 26, 27, Holden, R. R., 264
Deniston, W. M., 261 28,29,31,48,51,73,82, Holland, J. L., 260
Derogatis, L. R., 61, 70, 239, 114,220,221,234,262, Horn, J. L., 25, 264
261 264,265,270,272 Horner, M.S., 120,264
Diekstra, R. F. W, 63, 261 Goldman, S. L., 215, 263 Horney, K., 143, 264
Digman, J. M., 24, 25, 26, Goldstein, M. D., 215, 262 Horowitz, L. M., 116, 264
120,206,261, 262 Gorsuch, R. L., 4, 25, 59, Hough. L. M., 71, 83, 264
DiPlacido, J., 48. 271 220,225,233,234,235, Howard, K. I., 210, 267
Dodge, K. L., 175, 271 262.263,273 Hoyle, R. H., 223, 270
Donahue, M. J., 234, 256 Gottman, J. M., 185,263 Hsu, L. M., 81, 264
Doyle, W J., 259 Gough, H. G., 17.43,45,47, Hudy, M. J., 240, 274
Duijsens, I. J., 63, 261 49,120,263
Dunnette, M. D., 83, 264 Graham. J. R.. 114.257.263 lmes. S. A., 143, 258
Dye, D. A., 122,213.260 Grana, J. L., 255, 272 Isaka, H., 73, 264
Green, D. P., 215. 263
Eaton, N. K., 83, 264 Grove, E., 120,256,272 Jackson, D. N., 43, 63, 83,
Eaves, L. J., 39, 263 Gruber-Baldini, A. L., 173, 219,263,265,266.270
Eber, H. W, 25, 264 263 Jang, K. L., 63, 72, 266
Edwards, A. L., 47, 80, 261 Guilford, J. P., 24, 263 Janzen, B. L., 61, 272
Edwards, J. E., 120,261 Guilford, R. B., 24, 263 Jessor, R., 7, 265
 AUTHOR INDEX 279

Jex, S. M., 215, 258 Locke, B. Z., 260 Millon, T., 15,269
Joffe, R To, 40, 246, 255 Locke, H. J., 179, 267 Millsap, R, 7, 39, 263
John, O. P., 22, 24, 26, 31, Loehlin, J. C, 72, 271 Mischel, W, 52, 269
73,174,206,265,268 Loevinger, J., 213, 267, Moane, G., 7, 264
Jones, E. E., 55, 265, 268 273 Monello, R, 172, 266
Joreskog, J. G., 215, 223, 265 Lorr, M., 59, 114, 116,262, Montag, I., 43, 266
Juni, 5., 31, 265 263,267,268,271,273, Mooradian, T. A., 214, 269
274 Morrison, L. P., 174, 176,273
Kaemmer, B., 114,257 Lowell, E. L., 120,267 Mount, M. K., 31, 46, 83, 84,
Kaiser, R T., 8, 265 Lubin, B., 47, 267 122,155,206,207,256,
Kalemba, v., 40, 246, 255 Lushene, R. E., 4, 273 269
Kamp, J. D., 83, 264 Lyons, J. 5., 210, 211, 267 Murray, H. A., 15, 17,43,
Kaprio, J., 39, 274 47,49,259,269
Karney, B. R, 171, 172, 265, Marlowe, D., 81, 261 Muten, E., 83, 269
266 Marsh, H. W, 224, 267 Myers, I. B., 114,269
Katigbak, M.S., 73, 213, Marshall, G. N., 43, 267
224,225,266 Marshall, R. D., 261 Nakazato, H., 73, 256
Kaufman, A. 5., 237, 266 Martinez-Arias, M. R, 255, Narayanan, L., 73, 269
Kaufman, L., 114, 266 272 Neale, M. c., 39, 263
Keller, M. B., 173,264,266 Marusic, I., 269 Nelson, R 0., 211, 256
Keller, W, 48, 271 Masters, K. 5., 210, 270 Nesselroade, J. R., 256
Kelly, I. W, 61, 272 Matarazzo, J. D., 48, 267 Newsom, J. T., 259
Kendler, K. 5., 39, 263 McAdams, D. P., 71, 267 Nezlek, J. B., 214, 269
Kessler, R C, 39, 263 McCaulley, M. H., 114, 269 Nguyen, T. D., 242, 266
Klerman, G. L., 264 McClearn, G. E., 256 Nicholson, R. A., 81, 270
Kluth, C, 267 McClelland, D. C, 120, Nielsen, S. L., 80, 266
Kobak, K. A., 4, 271 267 Nisbett, R. E., 55, 265
Kosek, R B., 84, 172, 179, McCloy, R. A., 83, 264 Norem, J., 257
266 McCrae, R. R, 11,26,30, Norman, W T., 26, 270
Koskenvuo, M., 39, 274 35,36,38,39,40,41,42, Norris, A. H., 117,260
Kozma, A., 81, 266 43,45,48,49,51,53,56,
Kusulas, J. W., 43, 267 59,60,61, 68, 72, 73, 74, O'Mahoney, M. T., 210, 267
76,80,81,82,84,85,86, Odbert, H. 5., 23, 24, 26, 27,
Lachar, D., 83, 275 87,92,97,109,116, 117, 255
Lakatta, E. G., 272 118,119, 122,123,157, Ogles, B. M., 210, 211, 270
Lambert, M. J., 210, 270 165, 168, 180,213,214, Ones, D. 5., 122
Langston, C, 257 217,220,221,222,224, Ormel, J., 8, 270
Lanning, K., 266 225,226,229,232,244, Ostendorf, E, 22, 41, 73,
Larkin, 8., 264 246,256,259,260,267, 223, 257, 265
Larsen, D. L., 242, 266 268,269,271,272,273, Ozer, D. J., 8,19,48,51,232,
Larsen, R. M., 47, 267 275 265,270
Larson, R, 117, 266 McDonald, R. P., 224, 267
Leary, To, 116, 264, 266 McDonald-Scott, P., 264 Panter, A. T., 223, 270
Lester, D., 172,266 McGue, M., 39, 72, 257, Parker, G. V. C, 47, 269
Levin, J., 43, 266 258 Parker, J. D. A., 40, 223,
Levine, E. L., 73, 269 McNulty, J. L., 15, 263 255,270
Liebler, A., 54, 256 Meehl, P. E., 233, 269 Paunonen, S. v., 43, 60, 73,
Linehan, M. M., 80, 266 Megargee, E. I., 47, 174, 269 269,270
Lish, J. D., 210, 267 MieJ, G. M., 262 Peabody, D., 48, 270
Livesley, W. J., 63, 64, 72, Miller, T., 32, 59, 71, 92, 123, Pedersen, N. L., 72, 256,
265,266,272 126,247,269 264
280 AUTHOR INDEX

Perugini, M., 73, 258 Sher, K. J., 62, 69, 273 Tuley, M. R., 262
Piedmont, R. I., 178, 271 Shiraishi, D., 73, 256 Tupes, E. c., 25, 273
Piedmont, R. L., 29, 31, 40, Shock, N. W., 49, 272
41,42,43,46,47,48,49, Siegler, I. c., 7, 39, 268, 272 Useda, D., 40, 246, 273
51,54,56,59,60,61,62, Silva, E, 213, 255, 272
70,73,74,81,82,85, Silver, R. J., 81, 272 Vassend, 0., 213, 214, 218,
m, 120, 122, 155, 157, Silverstein, A. 8.,237,272 273
165,168,178,213,215, Sines, L. K., 81, 272 Venable, G. D., 234, 263
220,228,234,245,248, Skoner, D. P., 259 Vernon, P. A., 72, 265
258,260,269,270,271 Skrondal, A., 213, 214, 218, Vickers, R. R., 43, 267
Pincus, A. L., 63,116,262, 273 Viken, R. J., 39, 72, 274
271,275 Snyder, D. K., 185, 272
Plomin, R., 72, 74, 256, 264, Snyder, M., 76, 273 Wallace, K. M., 179, 182,
271 Sorbom, D., 215, 223, 265 183,267
Pontic as, Y., 62, 260, 261 Spector, P. E., 215, 258 Waller, N. G., 69, 71,80,82,
Przybeck, T. R., 15, 258 Spielberger, C. D., 258, 259, 219,255,256,273
Ptacek, J. T., 175,271 262,272,273 Wanous, J. P., 240, 274
Spirrison, C. L., 214, 273 Waters, L. K., 120,261
Ramanaiah, N. V., 261 Stagner, R., 122, 273 Watkins, C. E., 48, 274
Reichers, A E., 240, 274 Steers, K. M., 120,273 Watson, D., 54, 63, 118, 274
Reynolds, W. M., 4, 271 Stone, S. V., 109, 269 Webb, E., 24, 274
Riese, S. P., 19,48,51,232, Stones, C. R., 43, 263 Weinstein, H. P., 46, 85, 120,
270 Stones, M. J., 81, 266 122,155,271
Robbins, R. w., 206 Stout, C. E., 211, 273 Wells, P. A., 172, 272
Rodgerson, T. E., 238, 272 Strack,S., 59, 114,262,263, Welsh, G. 5., 49, 274
Rose, R. J., 39, 274 267,271,273,274 Wernimont, P. E, 206, 274
Rosenman, R. H., 114,262 Strauss, J. P., 46, 84, 206, West, M., 64, 223, 266
Rosolack, T. K., 221, 262 269 Wicks, R. J., 43, 258
Rouse, S. v., 80, 257 Strube, M. J., 215, 262 Widiger, T. A, 57, 58, 59, 63,
Rousseeuw, P. J., 114,266 Summerfeldt, L. J., 223, 270 64,67,84,260,262,274
Russell, R. J. H., 172, 272 Svrakic, D. M., 15,258 Wiggins, J. 5., 53, 63, 84, 116,
261,268,271,274,275
Saklofske, D. H., 61, 272 Takemoto-Chock, N. K., Williams, R. 8.,61,260,268,
Salovey, P., 215, 263 120,261 271,272
Sanchez-Bernard os, M. L., Tanaka, J. S., 223, 270 Willis, S. L., 173,263
255,272 Tanney, A., 64, 266 Wise, T. N., 62, 260, 261
Sanz, J., 255, 272 Taylor, R. M., 174, 176,273 Withey, S. B., 117,255
Saucier, G., 31, 269, 272 Tellegen, A., 71, 81, 82, 114, Wohlfarth, T., 8, 270
Schaie, K. w., 263 214,219.255.257,258, Wormworth. J. A., 63, 272
Scheier, M. E, 206, 272 273 Worthington, D. L., 83, 275
Schinka, J. A, 61, 82, 272 Tharp, R. Go, 173,273 Wortman. C. B.. 43. 267
Schlottman, R. S., 83, 275 Thiessen. D., 173, 266 Wrobel, T. A., 83, 275
Schmeck, R. R., 120, 272 Thurstone. L. L., 25, 273
Schmidt, C. w., 59, 61, 84, Tilly, S. M., 262 Young, R. K., 173, 266
257,261,275 Tobin, J. D., 272 Youniss, R. P.. 267
Schonemann, P. H., 225, 272 Tomes, J. L.. 219, 265
Schroeder, D. H., 30, 272 Trapnell, P. D., 116,275 Zirkel, 5., 257
Schroeder, M. L., 63, 266, Tresemer, D., 120, 273 Zonderman, A. B., 39, 60,
272 Trull, T. J., 40, 58, 59, 60, 62, 61,84,117,220,224,
Shanker, M., 269 63,65,69,246,273,274 225,226,232,260,269,
Shea, T., 63, 274 Trzebinski, J., 43, 270 272,275
 SUBJECT INDEX

AI: see Trust Agreeableness (cant.)

A2: see Straightforwardness in Axis II disorder, 66, 67, 135, 146
A3: see Altruism and Borderline personality disorder, 67
A4: see Compliance and CCICL. 181,188, 199,202
A5: see Modesty and Character, 122, 123, 124
A6: see Tendermindedness and Cooperativeness, 178
Abridged Big Five Dimensional and Interpersonal functioning, 116-117,
Circumplex, 115; see also 118
Circumplexes interpretation, 89, 92-93, 98,102
Accountability: see Treatment issues in substance abuse populations, 59
Achievement Striving (C4), 91, 96, 99, 120- Allport, Gordon, 21, 23
122,131-132,142,150,154,236 Alpha reliability, 37
Actions (04), 85, 94, 98, 99, 129, 132-133, Altruism (A3)' 89, 95,102,137,141-142,
137, 140, 142, 146, 153 146,154,236
Activity (E4)' 87, 93, 98, 123, 140 Angry Hostility (N2)' 59-85, 97, 131-132,
Adjective Check List (ACl), 44-45, 47 136, 139, 147
Adjectives, see also Five Factor Model; Antisocial personality disorder, 64
Lexigraphic hypothesis Anxiety (N]), 85, 97,131-132,136,139,145,
and FFM, 25-26 147
and folk concepts, 45-46 Assertiveness (E3), 41. 86, 93, 98, 101, 123,
interpretive variability, 29 141,146, 147, 150, 152
and 16 PF, 24 Assessment, 8-10, 11-14, 17,70,79-84,
uneven distribution in personality do- 249-251
main, 27 Avoidant personality disorder, 64-65,144-
versus sentences, 29-30 147
Adolescents, 74 Axis I disorders, 60-62, 69
Aesthetics (02), 88,102,140-141,145,148, Axis II disorders, 62, 67-68,150; see also
153 specific disorders
Agreeableness, 13, 22, 36-38, 89,129,133,137,
141,146,149,153;seealsospecij1cfacets Basic Personality Inventory, 45
and adolescents, 74 Borderline personality disorder, 64,131-135
and Anti-social personality disorder, 67 Brief Symptom Inventory, 61

281
282 SUBJECT INDEX

Cl: see Competence Convergent validity, 48, 53-54, 166, 184; see
C2: see Order also Validity
C3: see Dutifulness Coping, 32, 70,118,119,151,209
C4: see Achievement Striving Correlated error, 215-217
C5: see Self-Discipline Couples, see also Cross-observer agree-
C6: see Deliberation ment; Rater version
California Psychological Inventory, 44 Agreeableness and, 172
California Q-Set, 44 COA index, 182-184-185
Categorical model, 63 cross-observer differences, 175-185
Cattell, Raymond, 15, 16,24-26 manipulative tactics, 172
Character, 122-123, 124, 130-131, 134, 138, marital dissatisfaction, 172, 175
154 neuroticism and, 172, 192
Circumplexes, 61, 93-96, 97,114-115, 116- perceptions of spouse, 173-174, 182-183
117, 120-122, 130-134, 138, 140, 143- profile interpretaion, 100-105, 171-176
144,153-154,155; see also specific self-other congruence, 173-176
circumplexes Couples Critical Incidents Check List
Client-therapist relationship, 10, 13-14 (CClCL), 178-185
understanding the client, 9,11-12,70, Cooperativeness, 178, 181,195
249-251 correlations with NEO PI-R, 180, 183,
Clinical assessment, 12,62,66,69-71,207 184
Cloniger's Tridimensional Personality correlations with rater's personality, 180
Questionnaire, 44 Emotional, 178, 181, 187, 195, 196, 199
Cluster analysis, 114 Flexibility, 178, 181, 188, 195, 197
COA index: see Cross-observer agreement Interpersonal, 178, 181, 188, 195
Competence (Cn 90, 96, 102, 131-132, 137, Personal Reliability, 178, 181, 189, 195,
142,146,150,154,236 197
Competitive Orientation, 120-122, 130-131, Relationship Context, 178, 181
143-144, 154 Criss-cross testing, 174
Compliance (A4), 89-90, 102, 129, 142, 154 Cross-cultural generalizability, 73-74, 224-
Compulsive behaviors, 99 232
Computer reports, 149-151, 161-164 Cross-observer agreement (COA), 52-56,
Comrey Personality Scales, 45 73-74, 175,224-232; see also Couples;
Confirmatory factor analysis, 33, 215-218, Rater version
222-225 analyses, 176-177, 183-185
Congruence coefficients, 59-60, 220, 225 convergent validity, 53-54, 166, 184
Conscientiousness, 27, 36-38, 90,131-132, discriminant validity, 54
137, 142,146,149,150,154; see also index, 182
specific facets interfactor correlation, 214-215
and adolescents, 74 reliabi Ii ties, 180-181
in Axis II disorders, 66, 135 residualized rating scores, 182
and CClCL, 181, 188-189, 191, 193, 197, value of, 56, 83-84, 100, 204-205
199,202-203
and Character, 122-123, 124 Deliberation (C6), 91, 96, 97, 99, 102, 131-
and Competitive orientation, 120-122 132,135,137,142,146,154,236
interpretation, 90, 92-93, 95-96,101,102 Dependent personality disorder, 151
and Personal Reliability, 178 Depression (N3), 85, 131-132, 136, 139, 145,
in substance abuse popUlation, 59 147
and treatment selection, 126-127, 144 Differential diagnosis, ]],12,59-62,64-70,
Construct validity, 151, 233-237; see also 146-147
Validity Dimensional model, 63, 68
Discriminant validity, 54; see also Validity
 SUBJECT INDEX 283

Domains, 27, 31, 36-38, 40, 41, 213-218; see Factor analysis, 25, 26, 33, 35-41, 49-52,
also Five Factor Model; Personality; 114,212,215-218,222-225
specific domains Factor structure, 40-42, 59-60, 213-222
Dutifulness (C3), 91, 96,102,131-132,137, Faith Maturity Scale, 234-239
138, 142, 146, 149, 154 Fantasy (01). 87, 94, 102, 128, 132, 137-138,
140, 146, 153
El: see Warmth Feedback,9, 11, 12-13, 105-107, 108-110
E2: see Gregariousness Feelings (03), 41, 88, 95,102,128,137,140-
E3: see Assertiveness 141, 148
E4: see Activity Five Factor Model, 17-20, 24-27, 43-48, 58,
E5: see Excitement Seeking 71, 73-74, 212
E6: see Positive Emotions Basic Tendencies, 74-75
Edwards Personal Preference Schedule Character Adapatation, 74-75
(EPPS), 44, 47, 80 comprehensiveness, 43, 58
Emapthy, 11, 12 convergent validity, 48
Emotional Well-Being, 117-118, 130, 131- criticisms of, 71-72, 218-222
132, 133-134, 138, 140, 153 cross-observer convergence, 53, 73-74
English language: see Lexigraphic hypothe- Dynamic Processes, 75
sis External Influences, 74
Excitement Seeking (E5), 41, 59, 87, 93, 98, generalizability, 73-74, 212, 222-232
135, 147, 152 normal personality and, 58-59
Existential Well Being Scale, 234-239 Objective Biography, 74
Exploratory factor analysis, 33, 224-225 personality disorders, 67-68
Extraversion, 7, 9, 12, 13,25,27,36-38,86, robustness, 71
128,136,139,145,147,150,152;see Self Concept, 74
also specific facets taxonomy, 17-20,32
and adolescents, 74 theoretical foundations, 74-77
and Avoidant personality disorder, 65 vocational interests, 46
and Axis II disorder, 66, 136 Folk concepts, 45-46
and CClCL, 181, 187-188, 189, 190,201 Freud, Sigmund, IS, 20
and emotional we II-being, 117-119 Fundamental attribution error, 55
and Histrionic personality disorder, 65,
68 Gambling: see Impulsive behaviors
and interpersonal functioning, 116-117, Genotype, 4, 7-8, 38-39
118 Gough's Folk Concepts, 43, 45
interpretation 86, 92-93, 98, 101 Gregariousness <E2), 86, 98, 102, 128, 152
loading on A, 41, 86, 93, 98, 101-102, Guilford-Zimmerman Temperament Sur-
123 vey, 45
mean level change, 7
and Narcissistic personality disorder, High-order factors, 221-222
65 Histrionic personality disorder, 65, 68, 151
and Schizoid personality disorder, 12, Holden Psychological Screening Inventory,
65, 135 45
and Social phobia, 69 Holtzman Inkblot, 45
and treatment selection, 9,14,123-126 Hood Mysticism Scale, 234-235
Extrinsic Religiosity Scale, 234-235
Eysenck Personality Inventory, 44 Ideas (05), 88, 94, 102, 128, 132-133, 137,
Eysenck Personality Pro filer, 44, 222 145,153
Imposter phenomenon, 143
Facets, 30, 35, 36, 40-41, 84-92; see also spe- Impulsive behaviors, 131,133-135, 142, 146,
cific facets 147-149
284 SUBJECT INDEX

Impulsiveness (N5), 41, 85-86, 97,127-128, NEO PI-R (cont.)

131-132, 135, 137, 139, 147, 152 facets, 30, 35, 36, 40-41
Incremental validity, 69-71, 206-207, 212, factor analysis in clinical populations,
237-239 59-62,212-213
Insight analysis, 174 factor structure, 40-42, 59-60, 213-222
Internal consistency, 37, 52 incremental validity, 69-71, 234-239
Interpersonal Functioning, 36, 70, 116-117, interpretation, 84-112
118,130, 133, 138, 151, 155 measure of normal personality, 57-58,
Interpretation, 84-112 144, 156
Intrinsic Religiosity Scale, 234-235 orthogonality, 41, 213-218
and other personality measures, 44-45
jangle fallacy, 46-47 rater version, 35,116-171,206-207
jingle fallacy, 46-47 reliability, 36-40, 59-60
job selection, 121-122 replicability, 218-219, 222-228
joint factor analysis, 46-52; see also specific validimax factoring, 35
scales and validity scales, 79-84
jung, Carl, 14, 16, 19,20,43 Neuroticism, 36-38, 84-85, 131-132, 136,
139,145,147,152; see also specificfac-
K-Scale: see Validity scales ets
and adolescents, 74
Leary, Timothy, 116 in Axis I disorders, 135
Lexigraphic hypothesis, 20-32 in Axis II disorders, 66-67,131-132,135-
136, 146
Managed care environment, 9-10, 210, 211, and CCICL, 178, 181, 184, 187, 189, 190,
252 193,195,196,199,201
Mean level change, 7-8 and competitive orientation, 120-122
Millon Clinical Multiaxial Inventory, 45 and emotional well-being, 117-119
Minnesota Multiphasic Personality Inven- facets, 30, 35, 36, 40-41
tory,44,45,61,65,80-82,114 interpretation, 84-85, 92-93, 97, 100
Modesty (A5)' 90, 95, 133, 134, 137, 138, and mean level change, 7
141-142, 146,154 as predictor of psychological distress, 61,
Mood disorder: see Axis I disorders 93-96
Multidimensional Personality Question- secondary loading on C. 41
naire. 44, 81, 214-215 in substance abuse population, 59
Murray, Henry, 14, 15, 16,43,47 and treatment, 126-127
Myers-Briggs Type Indicator 44, 45, 114
01: see Fantasy
N1: see Anxiety 02: see Aesthetics
N2: see Angry Hositility 03: see Feelings
N3: see Depression 04: see Actions
N4: see Self-consciousness 05: see Ideas
N5: see Impulsiveness 06: see Values
N6: see Vulnerability Openness, 8, 14,27,36-38,87. 128, 137, 140,
Narcissistic personality disorder, 65 145, 148, 153; see also specific facets
NEOPI-R and adolescents, 74
cluster analysis, 114 and Axis II disorders, 13, 66-67
cross-cultural generalizability, 73-74, and CCICL, 181, 188, 189, 191, 192, 197,
224-232 199,202-202
development of, 27-32 and Flexibility, 178
and diagnosis. 57-70, 144, 156 and Histrionic personality disorder, 65,
domains,27,31, 36-38,40 66,68
 SUBJECT INDEX 285

Openness (conI.) Replicability, 218-219, 222-228

interpretation, 87, 94, 98, 102 Research
loadings on Extraversion, 41 construct validity, 233-237
and Narcissistic personality disorder, 65, incremental validity, 237-239
66 outcome studies, 9, 10,32, 138, 147, 210-
and Schizoid personality disorder, 65, 213,239-246
66,67,68 Residualized rating scores, 182
and treatment selection, 9, 14,123-126 Reverse acquaintanceship effect, 54-56
Order (C2), 90, 96, 131-132, 137, 142, 146,
154 Schizoid personality disorder, 64-65, 146-
Orthogonality: see Factor structure 147
Orthogonal Procrustean rotation, 225- Schizotypal personality disorder, 135-139
228 Self-Consciousness (N4), 12, 85, 120-122,
131-132, 136,139,145,147
Passive-aggressive behaviors, 103, 133, Self-Directed Search, 44
142, 146 Self-DiScipline (C5), 91, 96; 97, 99,102,131-
Personality 132, 135, 146, 149, 150, 154
adaptive function of, 3, 13, 156 Self-report data: see Rater version
assessment of, 20-22 Sixteen (16) PF Questionnaire, 24, 25, 45
change in, 5-8,39-40, 244-242 Social desirability, 80-81
cross-cultural generalizability. 73-74, Straightforwardness (A2), 89-95,141-142,
224--232 146,153,236
definition, 2-3 Substance abuse population, 59-60, 69-70,
developmental nature of, 3 135
dimensions, 26 Superego, 142
and genetics, 72-73
heritability, 72-73 T-scores, 150
identified by language, 21-22 Taxonomy, definition of, 19
models of, 14-17, 20, 42-43 Tender-Mindedness (A6), 90, 98, 99, 133,
normal, 1-5, 57-58, 156 137,141-142,154,236
stability of, 2, 5-8, 38-39, 72 Test-retest reliability, 37-38, 59-60
theory, 17 Theory, 17
Personality Assessment Inventory, 45 Therapies, 9-14, 62, 105-110, 125, 131, 135,
Personality Disorder Scale-R, 44 138,144; see also specific therapies
Personality disorder, NOS, 64 Treatment goals, 62
Personality disorders, 62-68,150; see also Treament issues, 9-14, 123-126, 131, 135,
specific disorders 138, 147, 210-213, 239-246, 247-
Personality Psychopathology 5, 45 249
Personality Research Forms, 44 anticipating the course of treatment, 11,
Phenotype, 4,7-8,38 13,70
Positive Emotions (E6), 87, 92, 101, 145, matching treatment to clients, 9,10,11,
147 13-14
Profile analysis, 113-114 TRIN, 81-82; see also Validity
Psychotherapy Treatment: see Treatment is- Trust (Al), 89,.95,133,137,141,146,153,
sues; specific therapies 236
Type A personality, 114
Rank order stability, 5-7 Typological analysis: see Profile analysis
Rater version, 166-167, 170, 171,206-207; Typologies, definitions of, 19, 113, 114; see
see also Cross-observer agreement also Circumplexes
Reliability, 5-7, 36-40, 59--60
Religious Well Being Scale, 234-235 Unidimensionality. 36
286 SUBJECT INDEX

Validimax factoring, 35, 41 Values (06), 88-89, 95, 98, 99, 132-133, 138,
Validity, 48, 53-54, 69-71, 79-84, 151,206- 140,142,148,154,236
207,212,233-237,237-239 VRIN, 81-82; see also Validity
construct validity, 151,233-237 Vulnerability (N6), 41,86,101,127,128,
content free, 81, 82 131-132,136-139,152
incremental validity, 151, 233-237
and self-report data, 83-84 Warmth (El), 41, 86, 102, 145, 153
and social desirability, 80-81 Wiggin's Interpersonal Adjective Scales, 45