TOC Electronic Journal: To print this article select pages 90-94.

The Millon Adolescent Personality Inventory and

the Millon Adolescent Clinical Inventory
Theodore Millon and Roger D. Davis

The theoretical grounds, purposes, and features of the Millon Adolescent Personality Inventory
(MAPI; Millon, Green, & Meagher, 1982) and its forthcoming replacement the Millon Adolescent
Clinical Inventory (MACI; Millon, in press) are reviewed. The rationale and procedure for the
construction of the component scales are briefly explained, and the logic of configural or profile
interpretation is examined and recommended. Uses and limitations of the MAPI and MAC1 are
considered. Evaluative research, although limited in scope, points to the general utility of the MAPI
and MACI as clinical tools, and to areas where further study may enhance their applicability in a
counseling context.

I
n a mature counseling science, theory, nosology, instrumentation, and Furthermore, the scales of the MAPI and MACI were also designed
intervention form a conceptually unified whole (Millon, 1990). Un- to coordinate with, and at least in part represent, the categorical diagno-
fortunately, as many have noted (e.g., Butcher, 1972), measurement ses of the official nosology. With the advent of the Diagnostic and
techniques and personality theory have developed almost indepen- Statistical Manual of Mental Disorders (3rd edition; DSM-III; Ameri-
dently. As a result, few diagnostic measures either have been based on can Psychiatric Association, 1980) and its evolution through DSM-III-R
or have evolved from clinical theory. The Millon inventories are a (American Psychiatric Association, 1987) and now DSM-IV (American
notable exception to this trend. Psychiatric Association, in press), diagnostic categories, and their cri-
teria have been reasonably well specified and refined. Despite many
CONCEPTUAL AND DEVELOPMENTAL BACKGROUND minor divergences, few diagnostic instruments currently available are
as fully consonant with these official nosologies as are the MAPI and
Within a context of physical and social changes, the behavior of ado- MCMI. Separate scales have been constructed for the MAPI and MACI
lescents often appears as some strange language undecipherable to the to help distinguish between more established personality patterns and
adult world. A variety of psychological tests for adolescents have been other clinical disorders requiring attention, a distinction judged to be of
devised through the years for decoding the strange “Rosetta Stone” of considerable utility by both test developers and clinicians (Dahlstrom,
adolescent behavior. Although these instruments provide useful infor- 1972). This should assist the counselor in disentangling the complex
mation, they tend not to be attuned to current issues and behaviors and interaction of personality patterns with other troublesome aspects of
lack grounding in a unifying theoretical model. adolescent functioning. The “expressed concern” scales address the
The Millon Adolescent Personality Inventory (MAPI) and Millon phenomenological attitudes that teenagers have regarding significant
Adolescent Clinical Inventory (MACI) are “junior” members, if you developmental problems (e.g., family problems, peer relations). The
will, of the Millon inventories, related to the Millon Clinical Multiaxial recently added and empirically derived clinical scales in the new MACI
Inventories (MCMI-I and MCMI-II; Choca, Shanley, van Denburg, (e.g., eating disorders, suicidal ideation) address behaviors that may
1991; Craig, 1992), but constructed specifically with the adolescent signify serious difficulties for the adolescent.
population in mind. Counselors and clinical psychologists were in- Development of the MAPI and MACI have been informed by
volved with the MAPI and MACI throughout all phases of their devel- several post-MMPI (Minnesota Multiphasic Personality Inventory) psy-
opment. Both inventories are relatively short, the first at 150 items and chometric developments. Item selection and scale development for both
the second at 165 items, and adapted to a sixth-grade reading level. older and more recent forms progressed through three sequential vali-
Questions are presented in a language that teenagers use, and questions dation stages (Loevinger, 1857): (a) theoretical-substantive, (b) inter-
deal with matters teenagers can understand and find relevant to their nal-structural, and (c) external-criterion. Rather than attempt to establish
concerns and experiences. As a result, the short attention span and validity after instrument construction, this approach builds validity into
opposition of adolescents constitute much less of a barrier than with the instrument from the beginning, upholding standards of developers
longer inventories designed originally for use with a broad and largely committed to diverse construction and validation methods (Hase &
adult population. Goldberg, 1967). This multistage process culminates in scales that are
Unlike most diagnostic instruments, the personality scales of the theoretically, statistically, and empirically valid. Because each item
Millon adolescent inventories were designed to be an operational mea- must survive each stage of refinement, the chance that any item will
sure of a set of personality patterns and syndromes derived from a theory prove “rationally surprising,” “ structurally unsound,” or “empirically
of personality and psychopathology (Millon, 1969, 1977, 1981, 1990). impotent” is greatly diminished. As might be expected, the likelihood
The presence of a comprehensive theoretical system undergirding a of instrument generalizability is greatly increased.
diagnostic instrument, in conjunction with the three-stage process of In the theoretical-substantive stage, a large item pool is derived
inventory construction described in this article, significantly increases according to an explicit theoretical framework, here Millon’s biosocial
the instrument’s counseling and clinical utility, suggesting dynamics (1969) and evolutionary (1990) theories of personality and psychopa-
and developmental hypotheses and possible interrelationships between thology. For construction of the original MAPI personality scales, more
the two. than 1,000 items were gathered from numerous sources, including other

570 Journal of Counseling & Development l May/June 1993 l Volume 71

 The Millon Adolescent Personality Inventory and the Millon Adolescent Clinical lnventory

psychological tests and abnormal and personality texts. Many other should have at least a master’s-level degree in a relevant field of mental
items were generated to ensure full content coverage. The entire pool health, meeting membership qualifications for their appropriate profes-
was then sorted by eight professionals into their appropriate theoretical sional organization—American Psychological Association, American
categories. Only those items sorted correctly by six of eight judges were Counseling Association, American Psychiatric Association, American
retained. The internal structural stage concerns the homogeneity of the Association for Marriage and Family Therapy, or the National Associ-
individual scales and their relationship to each other. A total of 64 items ation of Social Workers. The credentials of test users are monitored by
were retained that contributed to the homogeneity of their respective National Computer Systems in an effort to prevent misuse of the
“personality” scales. Items that correlated less than .30 with their instruments.
respective scale were dropped, thereby increasing the fidelity of the The interpretive report service is considered a professional-to-
scale to its underlying theoretical construct. Items for the “expressed professional consultation. As Cansler (1986) noted, how computerized
concerns” scales were then developed on theoretical-substantive reports are to be used is a professional issue of some weight. Counselors
grounds and on the basis of counselors’ and therapists’ impressions of are encouraged to use this information in developing guidance or
their relevance to adolescent problem areas. These items were then clinical programs and in making ongoing management decisions (direct
evaluated further by counselors and psychologists with extensive expe- sharing of information contained in the report with either the family or
rience; those items placed correctly by 75% or more of professional the adolescent is not recommended). Furthermore, the interpretive
judges were retained. A total of 80 items were added by this procedure, report should not be appropriated in its totality. Rather, an interpretive
plus 6 for correcting distortions, resulting in a final MAPI form of 150 report should serve as one component of the evaluation of the adolescent
items. and should be viewed by the consultee as a series of probabilistic
The MAPI was normed on groups of both nonclinical and clinical judgments, as one facet of a total evaluation, rather than as a set of
participants. The “normal” group consisted of more than 2,000 boys definitive statements. It is doubtful that a professional can truly
and girls 13 to 18 years of age. The clinical group consisted of 325 “understand” the patient unless an iterative process of clinical inference
outpatient and 105 inpatient adolescents. Separate norms are provided and hypothesis testing, using all available extra-test information, has
for boys and for girls, divided into age groups 13 to 15 and 16 to 18. been personally engaged. The report is a supplement, not a substitute,
Essentially identical procedural steps were taken in the revision leading for counseling and clinical judgment.
to the MACI. Because the MACI is more specifically oriented to the The MAPI and MACI are well suited for scientific research. Objec-
assessment of “clinical” populations, the normative groups consisted tive and largely theory-grounded scale scores and profile patterns can
of approximately 700 patients seen in outpatient, residential, and other be used to test a variety of clinical, experimental, and demographic
mental health centers. Norms are now provided for those up until 19 hypotheses. As noted, the samples used for such purposes are best drawn
years of age. Actuarial base rate data, rather than normalized standard from clinical populations. An attempt to apply the MAPI or MACI to a
score transformations, were used to calculate and quantify scale mea- wider range of problems or participants (e.g., to identify neurologic
sures. These data provide a basis for selecting optimal cutting lines and lesions or to assess personality traits among normal school students) is
help ensure that the obtained frequency of MAPI- and MACI-generated to apply the instrument to settings and populations for which they are
diagnoses and profile patterns are comparable to representative diag- neither intended nor appropriate.
nostic prevalence rates.
ADMINISTRATION, SCORING, AND INTERPRETATION
USES AND LIMITATIONS OF THE MAPI AND MACI
A principal goal in constructing the MAPI and MACI was to keep the
The MAPI and MACI were designed to be used by counselors, clini- total number of items small enough to encourage use with adolescents
cians, and other mental health service professionals as an aid in identi- in a wide range of diagnostic and treatment settings, yet large enough
fying, predicting, and understanding a wide range of psychological to allow the assessment of a wide range of clinically relevant content
attributes characteristic of adolescents. The new MACI, in contrast with areas. The MAPI consists of 150 items, the MACI of 165. Both are much
the MAPI, which was routinely used as one component of testing for shorter than many comparable instruments. As a result, they can be
vocational, academic, and psychological advising, was designed specif- completed by most patients in 20 to 30 minutes. Although usually
ically for use in mental health service settings as an instrument administered individually, there are no special conditions or instructions
for adolescent clinical assessment. The wide set of empirically validated required to achieve reliable results beyond those printed on the test
scales constituting the MACI not only seeks to identify personal diffi- booklet itself. Thus, the MAPI and MACI lend themselves well to group
culties, such as peer alienation, identity confusion, and family conflicts, settings.
but also should assist counselors and clinicians in characterizing young- The MAPI and MACI are intended for use with all adolescents who
sters engaged in acting-out behaviors, depressive moods, and borderline have at least a sixth-grade reading level. Optimally, the participant
personality features. Assessing strengths and weaknesses, both inven- should be comfortable and free of excessive fatigue, severe confusional
tories aid the counseling and clinical professional to maximize poten- states, drug intoxication, or sedation, each of which significantly alters
tials by reflecting the full scope of personality attributes in addition test results. If the test must be administered under these conditions, later
to specific problem areas. The forthcoming MACI will be especially retesting would be wise. Procedures for administering these instruments
useful with adolescents involved in clinical contexts. Norms for both are similar to those of other self-report inventories. Test directions are
instruments are divided to maximize sensitivity to both age and sex printed on the front page where client codes and relevant information
characteristics. can be filled in, while the test items themselves and spaces for true-false
Individuals responsible for supervising the use of both the MAPI responding constitute the remaining three pages. The words true and
and MACI and their computer-based interpretive reports must have a false are printed next to each item, minimizing patient confusion,
sufficient background in test logic, psychometric methods, and relevant maximizing response accuracy, and allowing the clinician to easily scan
professional practice. With the exception of graduate students conduct- the form for completeness immediately after it is handed in, and later
ing research under supervision, all individuals using these inventories generate hypotheses at the item level, if desired. Machine scoring is the

Journal of Counseling & Development l May/June 1993 l Volume 71

571
Millon and Davis

simplest method for obtaining MAPI profiles and reports; hand-scoring Such an interpretive procedure seeks to break the pattern of labeling
templates, however will be available shortly. patients and fitting them into procrustean categories. By stressing the
Modifying indices (e.g., self-disclosure, desirability, debasement) network of interlocking personality traits and clinical symptoms, con-
and a validity index have been included in the MACI to detect adoles- figural analysis allows the clinician to form a conceptually unified
cents for whom test results may have been unduly distorted because understanding of each patient’s pathology, a formulation which includes
of their test-taking attitudes (e.g., lack of conviction regarding their not only diagnostic, dynamic, and developmental considerations, but
responses). also suggests unique factors may sustain the pathology, and possible
Profile interpretation is a useful method of evaluating MAPI and avenues for effective intervention.
MACI results. In general, the goal of any assessment is to provide the In addition to undertaking one’s own interpretation, rapid and con-
professional with information relevant to accurate characterization and venient computer-generated MAPI and MACI narratives are available
treatment of the individual patient. Considering all the evidence, includ- that integrate the personological, phenomenological, and behavioral
ing biographic data and data from other tests, the professional should features of the adolescent. Arranged in a style similar to those prepared
use an iterative process of hypothesis generation to formulate a satisfac- by professional psychologists, the individualized Interpretive Report
tory conception of the patient’s difficulties, as well as a plan for printout is provided by the publisher, National Computer Systems (also
counseling intervention. No matter how useful individual MACI scales publishers of the MCMI and MMPI). These reports synthesize data from
may be in identifying specific personality patterns or clinical difficul- both scale score elevations and profile configurations and are based on
ties, their value in these regards will be enhanced appreciably if they are actuarial research findings and Millon’s theoretical schema (Millon,
seen in a configuration of several interrelated scales (e.g., an eating 1969, 1981, 1990). Beyond giving a complex description of patient
disorder in a compulsive adolescent means something different than in dynamics, the reports summarize findings along several dimensions.
a borderline youngster). The theoretical framework and clinical charac- Note, however, that the report is intended to serve as a rich source of
terizations associated with each personality style are available in Mod- clinical hypotheses. What is to be selected, rejected, emphasized, or
ern Psychopathology (Millon, 1969), Disorders of Personality (Millon, deemphasized in the report depends on the individual case and the
1981), and Toward a New Personology (Millon, 1990). counselor’s experience and judgment. As noted by Wetzler and Mar-
The following discussion recommends some of the steps to be lowe (1992), “The test is only as good as its user” (p. 428).
followed in making a configural interpretation. Before this process can
be engaged, however, the personality and clinical features characteriz- EVALUATIVE RESEARCH
ing each of the separate scales should be thoroughly reviewed. This must Research with the MAPI is still somewhat scarce (a complete list of
be done prior to undertaking an analysis of the profile configurations, references is available from the test’s publisher, National Computer
because the accuracy of such interpretations is dependent on the mean- Systems). In part, this reflects the fact that both the MAPI and MACI
ing and significance of each of the individual scales of which the profile are still new tests relative to such established veterans as the MMPI. The
is composed. Configural interpretation is a deductive synthesis achieved paucity of research, however, may also, in part, reflect the widespread
by refining, blending, and integrating the separate characteristics tapped belief that adolescence is a time of rapid changes, of developmental
by each scale. chaos, a time dominated by variance rather than continuity, and thus, in
A basic separation should be made in the initial phase of interpreta- general, is difficult to manage with sustainable hypotheses.
tion between those scales that pertain to the teenager’s basic personality Because the MAPI and MACI were developed using essentially the
style, those that address the adolescent’s expressed attitudes, and those same techniques as the MCMI (e.g., Loevinger’s [1957] three-stage
that signify the presence of clinical syndromes. Each of these sections approach to test construction and validation), many of the issues and
reflects different and important dimensions of the diagnostic picture. concerns that have been raised in connection with the MCMI are also
For this reason, the configural interpretation should begin by dividing relevant to these inventories. Indeed, one of the advantages of a family
the profile into these subsections, focusing first on the significance of of instruments, similarly constructed, is that increases in knowledge
scale elevations and profile patterns within each section. Only then can regarding one test tends to inform an understanding of the others.
a synthesis of the separate scales be undertaken. For example, first the Psychometrically, many researchers have questioned the wisdom of
counselor should note that a youngster shows, say, unruly (antisocial) having a high level of item overlap among diagnostic scales in regard
and forceful (sadistic) personality patterns (Section l), expressed con- to the MCMI (e.g., Gibertini, Brandenburg, & Retzlaff, 1986; McCann,
cerns regarding self-devaluation and family discord (Section 2), with 1990; Wiggins, 1982). Related to this is the fact that the scales are
clinical syndromes revealing impulsive propensity and suicidal ten- intercorrelated, perhaps reflecting fewer underlying dimensions than are
dency (Section 3). ostensibly present. Variable success has been achieved in efforts to
In addition, the richness and accuracy of all self-report measures are validate the MCMI as a gauge of diverse clinical syndrome (Axis I)
enhanced when their findings are viewed in the context of other clinical groups, and some have suggested that the MAPI tends to underdiagnose
sources and data, such as demographic background, biographic history, psychopathology in clinical populations as well (Tracy, 1986), although
and other clinical features. “Blind” interpretations may be useful the relationship of MAPI code patterns to depression is somewhat better
during graduate training, but are unwise in professional counseling and understood (Pantle, Evert, & Trennery, 1990, Trenerry, Pantle, &
clinical settings. It is here in the sphere of auxiliary data that the clinician Zimbelman, 1988; Watchman, 1987). Others, although acknowledging
is in a superior position to the automated computer report. Not only does the acceptable reliability of individual MAPI scales, have questioned
the combination of various gauges from diverse settings provide the data the stability of the 2-point codes (Reidy & Carstens, 1990), an area that
aggregates (Epstein, 1979, 1983) that increase the likelihood of drawing has not yet been adequately researched with most personality invento-
correct inferences, but also multimethod approaches (Campbell & ries.
Fiske, 1959) provide both the experienced and novice counselor with What must be kept in mind when evaluating the MAPI and the
an optimal base for deciphering those special, if not unique, features that forthcoming MACI for differential diagnosis is the organic unity of the
characterize each patient. individual and the questionable validity of drawing sharp distinctions

572 Journal of Counseling & Development l May/June 1993 l Volume 71

 The Millon Adolescent Personality Inventory and the Millon Adolescent Clinical Inventory

between clinical syndromes and personality patterns. Certain personal- would no longer be as well coordinated with the official nosology as
ity disorder scales are sensitive to the patient’s current affective state. before. Moreover, the senior author’s own personality theory has moved
Despite methodologic and psychometric procedures designed to tease forward, from a theory concerning sources of reinforcement (Millon,
apart the enduring characteristics of personality and the more transient 1969, 1981) to one grounded in evolutionary principles (Millon, 1990).
features of clinical conditions, every scale reflects a mix of predisposing These theoretical developments have seen the explication of several new
and generalizing attributes, as well as those of a more situational or acute personality patterns, notably the doleful or depressive (MACI scale 2B),
nature. These results stem in part from shared scale items, but the level the forceful or sadistic (MACI scale 6B), and the sensitive or self-
of covariation is appreciably greater than what would be suggested by defeating (MACI scale 8B). New clinical problems regarding adoles-
item overlap alone (Lumsden, 1987). In other words, although person- cents have also moved to the forefront since the MAPI was constructed,
ality measures are influenced by the state characteristics of clinical such as disordered eating behavior, increasing suicides, borderline
conditions, measures of clinical conditions are influenced by personality tendencies, and alcohol and other drug problems. Finally, rapidly chang-
characteristics as well. Although presenting a challange for counselors, ing cultural forces and professional vocabulary also suggested that a
this is perhaps as it should be, as it points to the nature of persons as revision might be in order. Appendix A compares the MAPI and MACI
synthetic beings rather than as a conglomeration of noninteracting data by clinical scale domains. The reader is advised to note specifically the
domains. addition of several new personality scales and a substantial increase in
the number of explicitly clinical scales. We hope that the addition of
CONCLUDING COMMENTS AND FUTURE DIRECTIONS such scales will facilitate its use in both counseling and clinical contexts.

Bypassing for the moment psychometric issues, the influence of mod- REFERENCES
erator variables, sample representativeness, concerns regarding national
versus local base rates, the unreliability of psychiatric diagnoses as well American Psychiatric Association. (1980). Diagnostic and statistical manual of
as their suitability as a criterion from which to evaluate self-report mental disorders (3rd ed.). Washington, DC: Author.
American Psychiatric Association. (1987). Diagnostic and statistical manual of
instruments, and a score of other concerns that enter into evaluating and
mental disorders (3rd ed., rev.). Washington, DC: Author.
constructing a given test, perhaps the best index of the value of an
American Psychiatric Association. (in press). Diagnostic and statistical manual
instrument is its acceptance and utility in the counseling and clinical of mental disorders (4th ed.). Washington, DC: Author.
community. Simply put, the MAPI has become the most frequently used Archer, R. P., Maruish, M., Imhof, E. A., & Piotrowski, C. (1991). Psychological
adolescent-oriented personality-clinical inventory in recent years test usage with adolescent clients: 1990 survey findings. Professional Psychol-
(Archer, Maruish, Imhof, & Piotrowski, 1991; Craig & Horowitz, 1990; ogy: Research and Practice, 22, 247-252.
Piotrowski & Keller, 1984). Butcher, J. N. (Ed.). (1972). Objective personality assessment. New York:
Nevertheless, counselors must be aware of the inherent limitations Academic Press.
of the self-report method as a data source. The psychometric character- Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation
istics of particular instruments, the tendency of similar patients to by the multitrait-multimethod matrix. Psychological Bulletin, 56, 81-105.
Cansler, D. (1986). Review of the Millon Adolescent Personality Inventory.
interpret questions differently, the effect of current affective states, as
Journal of Counseling and Development, 64, 469-471.
noted earlier, and the effort of patients to effect certain appearances and
Choca, J. P., Shanley, L. A., & van Denburg, E. (1991). Interpretive guide to the
convey certain impressions all narrow the boundaries of this method’s Millon Clinical Multiaxial Inventory. Washington, DC: American Psycholog-
accuracy. Moreover, neither the MAPI nor the MACI is a perfectly ical Association.
precise gauge of their underlying theoretical constructs or of the clinical Craig, R. J. (Ed.). (1992). The Millon Clinical Multiaxial Inventory: A clinical
characteristics measured. Systematic construction through accepted research information synthesis. New York: Erlbaum.
techniques of validation (Loevinger, 1957), however, should allow an Craig, R. J., & Horowitz, M. (1990). Current utilization of psychological tests at
inventory to approach the upper boundaries of clinical utility. Experi- diagnostic practicum sites. The Clinical Psychologist, 43, 29-36.
ence with the MCMI, the adult form of the MAPI and MACI, has shown Dahlstrom, W. G. (1972). Whither the MMPI? In J. N. Butcher (Ed.), Objective
that it is on the mark with about 55% to 65% of patients to whom it is personality assessment (pp. 85-116). New York: Academic Press.
administered, is appraised as both useful and generally valid, but with Epstein, S. (1979). The stability of behavior: I. On predicting most of the people
most of the time. Journal of Personality and Social Psychology, 37, 1097-
partial misjudgments in about another 25% to 30% of cases, and is
1126.
significantly off-target (i.e., appreciably in error) about 10% to 15% of
Epstein, S. (1983). Aggregation and beyond: Some basic issues on the prediction
the time. These statistics are probably true of the MAPI and MACI as of behavior. Journal of Personality, 51, 360-392.
well, but are in the quantitative range of five to six times greater than Gibertini, M., Brandenburg, N. A., & Retzlaff, P. D. (1986). The operating
random diagnostic assignments or chance (Millon, 1987). characteristics of the Millon Clinical Multiaxial Inventory. Journal of Person-
Beyond the limitations of data sources and instruments lie the ality Assessment, 48, 308-311.
tendencies and limitations of individual test consumers themselves. Hase, H. D., & Goldberg, L. R. (1967). Comparative validity of different
Many of these have been addressed in previous sections. The ease and strategies of constructing personality inventory scales. Psychological Bulletin,
availability of computer-generated reports threaten to lull consumers 67, 231-248.
into an uncritical acceptance of test results. Test users have an ethical Loevinger, J. (1957). Objective tests as instruments of psychological theory.
Psychological Reports, 3, 635-694.
obligation to compare such assessments against independently gener-
Lumsden, E. A. (1987). The impact of shared items on the internal-structural
ated clinical evidence.
validity of the MCMI. In C. Green (Ed.), Conference on the Millon clinical
Throughout this article we have referred to the revision of the MAPI, inventories (MCMI, MBHI, MAPI) (pp. 325-333). Minneapolis, MN: National
the MACI. Why was the MAPI revised? The last years have seen several Computer Systems.
significant changes in the personality landscape. The revision of the McCann, J. T. (1990). A multitrait-multimethod analysis of the MCMI-II clinical
DSM (American Psychiatric Association, 1980) through DSM-III-R syndrome scales. Journal of Personality Assessment, 55, 465-476.
(American Psychiatric Association, 1987) and now DSM-IV (American Millon, T. (1969). Modern psychopathology. Philadelphia: Saunders (reprinted;
Psychiatric Association, in press) raised the possibility that the MAPI Prospect Heights, IL: Waveland Press).

Journal of Counseling & Development l May/June 1993 l Volume 71 573

Millon and Davis

Millon, T. (1977). Millon Clinical Multiaxial Inventory: (MCMI-I) manual. C Body Comfort
Minneapolis, MN: National Computer Systems. D Sexual Acceptance
Millon, T. (1981). Disorders of personality: DSM-III, Axis II. New York: Wiley. E Peer Security
Millon, T. (1987). Millon Clinical Multiaxial Inventory: II (MCMI-II) manual. F Social Tolerance
Minneapolis, MN: National Computer Systems.
G Family Rapport
Millon, T. (1990). Toward a new personology: An evolutionary model. New
York: Wiley-Interscience. H Academic Confidence
Millon, T. (in press). Millon Adolescent Clinical Inventory manual. Behavioral Correlates
Millon, T., Green, C. J., & Meagher, R. B., Jr. (1982). Millon Adolescent SS Impulse Control
Personality Inventory manual. Minneapolis, MN: National Computer Sys- TT Societal Conformity
tems. UU Scholastic Achievement
Pantle, M. L., Evert, J. M., & Trennery, M. R. (1990). The utility of the MAPI WW Attendance Consistency
in the assessment of depression. Journal of Personality Assessment, 55(3&4),
673-682.
Piotrowski, C., & Keller, J. W. (1984). Psychodiagnostic testing in APA- MACI
approved clinical psychology programs. Professional Psychology: Research
and Practice, 15, 450-456.
Personality Scales
Reidy, T. J., & Carstens, C. (1990). Stability of the Millon Adolescent Personality 1 Introversive (Schizoid)
Inventory in an incarcerated delinquent population. Journal of Personality 2A Inhibited (Avoidant)
assessment, 55(3&4), 692-697. 2B Doleful (Depressive)
Tracy, H. M. (1986, March). The clinical use of the Millon Adolescent Person- 3 Cooperative (Dependent)
ality Inventory. Paper presented at the Millon Workshop, Conference, and 4 Sociable (Histrionic)
Training Institute, Miami, FL. 5 Confident (Narcissistic)
Trenerry, M. R., Pantle, M. L., & Zimbelman, K. K. (1988, March). Relationships Unruly (Antisocial)
6A
between the Rorschach and the Millon Adolescent Personality Inventory.
6B Forceful (Sadistic)
Paper presented at the meeting of the Society for Personality Assessment, New
Orleans, LA. 7 Respectful (Compulsive)
Watchman, B. (1987). The impact of personality style on depressive subtypes 8A Negative (Negativistic)
[Abstract]. In C. Green (Ed.), Conference on the Millon Clinical Inventories 8B Sensitive (Self-Defeating)
(MCMI, MBHI, MAPI) (p. 115). Minneapolis, MN: National Computer Sys- 9 Borderline Tendency
tems. Expressed Concerns
Wetzler, S., & Marlowe, D. (1992). What they don’t tell you in the test manual: A Identify Diffusion
A response to Millon. Journal of Counseling & Development, 70, 427-428.
B Self-Devaluation
Wiggins, J. (1982). Circumplex models of interpersonal behavior in clinical
C Body Disapproval
psychology. In P. Kendall & J. Butcher (Eds.), Handbook of research methods
in clinical psychology (pp. 183-222). New York: Wiley. D Sexual Discomfort
E Peer Insecurity
APPENDIX A F Social Insensitivity
G Family Discord
Comparison of Millon Adolescent Personality Inventory H Childhood Abuse
(MAPI) and Millon Adolescent Clinical Inventory (MACI) Clinical Indexes
by Assessment Scales AA Eating Dysfunction
BB Academic Noncompliance
MAPI CD Chemical Dependency
Personality Scales EE Delinquent Disposition
1 Introversive (Schizoid) FF Impulsive Propensity
2 Inhibited (Avoidant) GG Anxious Feelings
3 Cooperative (Dependent) HH Depressive Affect
4 Sociable (Histrionic) II Suicidal Tendency
5 Confident (Narcissistic)
6 Forceful (Antisocial) Theodore Millon is a professor in the Department of Psychology at the Univer-
7 Respectful (Compulsive) sity of Miami, Coral Gables, Florida, and is a professor in the Department of
8 Sensitive (Passive-Aggressive) Psychiatry at Harvard Medical School. Roger D. Davis is a university fellow in
Expressed Concerns the Department of Psychology at the University of Miami, Coral Gables, Florida.
A Self-Concept Correspondence regarding this article should be sent to Theodore Millon, 5400
B Persona1 Esteem SW 99 Terrace, Coral Gables, FL 33156.

574 Journal of Counseling & Development l May/June 1993 l Volume 71