Dr. Stephen Strack was introduced to personality theory and assessment as an undergraduate at U.C. Berkeley by such people as Jack and Jeanne Block, Harrison Gough, Kenneth Craik, and James Coyne. As a graduate student at the University of Miami he worked with Theodore Millon, the late Robert B. Meagher, Jr., and Catherine Green. Since receiving his PhD in clinical psychology (1983), Dr. Strack has continued to work in the area of personality, most notably with Maurice Lorr, with whom he co-edited the book Differentiating Normal and Abnormal Personality (Springer, 1994). He is a nationally recognized expert on Millon's personality theory and measures. He is a Fellow of the American Psychological Association and a Board Member of the Millon Institute for Advanced Studies in Personality and Psychopathology. Dr. Strack is currently a staff psychologist at the VA Outpatient Clinic, and an instructor at the California School of Professional Psychology and Fuller Graduate School of Psychology, all in the Los Angeles area.

The PACL is 153-item self-report and rating measure of Theodore Millon's (1969/1983) eight basic personality patterns for use with normal adults and counseling patients. It features a problem indicator (PI) scale that taps aspects of Millon's three severe schizoid, cycloid, and paranoid styles, and may be used a measure of personality disorder. Three response bias indices are available to detect protocols that have been answered in a random manner or with intent to give an overly favorable or unfavorable impression. PACL personality scales measure theoretically-derived, normal versions of the character types most frequently seen in clinical settings. Test results yield rich descriptions of respondents in a language that closely resembles that found in the DSM-IV (American Psychiatric Association, 1994). The PACL is exceptionally easy to administer and complete (5-10 mins.), and is readily accepted by resistant clients and those who find a questionnaire too demanding (e.g., the elderly and some disabled clients). It is frequently used by therapists and personnel psychologists who work with relatively high functioning individuals and who want to understand the strengths of their clients as well as their weaknesses. The measure has been used in numerous research studies that tested various propositions of Millon's theory and addressed the interface between normal and abnormal personality (e.g., Strack, 1991c, 1993, 1997).

DEVELOPMENT OF THE PACL

The PACL originated at the University of Miami in the early 1980s in a research group led by Theodore Millon, Catherine Green, and the late Robert Meagher, Jr. At that time very little empirical work had been accomplished using Millon's model of personality and we sought ways of changing this. Many theses and dissertations were spawned in the research group, including my second year (master's level) project, Development of the Personality Adjective Check List and Preliminary Validation in a Normal College Population (1981). The purpose of creating the PACL was to open the door for research on normal subjects. By developing a measure of Millon's personalities for normal individuals we hoped to capitalize on the large pool of nonclinical research subjects available to investigators in college and business settings. Our long-term goals included building an analogue model of personality disorders among normals, and demonstrating the inherent continuity between the normal and abnormal domains of personality functioning.

Adjective Check List Format

In deciding on a format for creating the measure, two factors were primary. First, Millon's empirical work had taught him that fast and simple tests were the ones most often selected for use by investigators and completed successfully by clients. Second, I wanted the measure to build on the interpersonal roots of Millon's theory, a strong attraction for me given my previous training and work at Berkeley. Therefore, in the tradition of LaForge and Suzcek's (1955) Interpersonal Check List, and Gough's (Gough & Heilbrun, 1983) Adjective Check List, we decided to develop a quick and easy-to-complete adjective measure that would allow for both self-reports and observer ratings.

Loevinger’s Model of Test Development

The check list was developed using a method outlined by Loevinger (1957), and which was used by Millon and his colleagues for creating his clinical measures. In this method test construction is theory-driven and follows a step-by-step process with development and validation occurring together.

In the first stage of development and validation, 405 theory-derived adjectives were selected to measure normal versions of Millon's (1969/1983) eight basic and three severe personality styles. Items were drawn from numerous sources, including Modern Psychopathology, and were selected based on rater judgments that each item had a clear best-fit for one style (see Strack, 1987, 1991c, for details).

The second, structural validity, phase of test construction involves creating scales that match the underlying theory. Toward this end, the 405-item experimental check list was given to 207 men and 252 women from colleges in Ohio and Florida. Preliminary scales were created from items that were endorsed by at least 5% and no more than 80% of subjects; had minimum item-scale correlations of .25; and maximum within-scale item-item correlations of .49 (to prevent redundancy; Strack, 1987, p. 577). Using these criteria, measures were created for each of Millon's eight basic styles that had satisfactory internal consistency and temporal reliability. Alpha coefficients ranged from .76 to .89 (new sample Median = .83; Strack, 1987, p. 578), while test-retest correlations over a three-month period ranged from .60 to .85 (Median = .72 across sexes; Strack, 1987, p. 578). Additional data showed the scales to be relatively free from social desirability bias (Strack, 1987, p. 581).

Unfortunately, measures could not be developed for the three severe schizoid, cycloid, and paranoid personalities because of extremely low endorsement rates (< 5%) for most keyed items. Rather than throw away the handful of good items that remained for these measures, they were combined into an experimental problem indicator scale, PI, which we thought might be useful in identifying persons with personality disorders.

Response Bias Indices

In addition to the personality and experimental scales, I developed three response bias indices to aid in the detection of faked protocols (Strack, 1991c), namely, Random (R), Favorable (F), and Unfavorable (UF). Separate groups of college students were asked to complete the PACL randomly, or with intent to give an overly favorable or overly unfavorable self-report. Discriminant function analyses were used to distinguish the faked tests from PACLs completed under the normal instructional set. Equations were derived from these analyses (separately for men and women) and were cross-validated with independent samples. The equations were able to correctly identify a large majority of faked (75%-91%) and normal tests (60%-94%).

External Validity

In accordance with the third stage of test development, extensive external validity data have been reported for the PACL by myself and a number of independent researchers in the form of correlations with tests of personality, mood, and dispositional variables, and reports from subjects about current and past behavior (Chung, 1993; Durff, 1994; Horton & Retzlaff, 1991; Pincus & Wiggins, 1990; Strack, 1987, 1991b, 1991c, 1994; Strack & Lorr, 1990b; Strack, Lorr, & Campbell, 1989; Wiggins & Pincus, 1989, 1994). My own research demonstrated that each PACL scale is in line with theoretical expectations and measures milder versions of Millon's (1969/1983) pathological styles. For example, the scale measuring the avoidant personality (Inhibited) was positively associated with measures of shyness, submissiveness, and social anxiety, and negatively associated with measures of sociability, dominance, and emotional well-being (Strack, 1991c). The scale measuring aggressive traits (Forceful) was positively linked to measures of arrogance, dominance, assertiveness, and autonomy, and negatively linked to measures of deference, submissiveness, and conscientiousness (Strack, 1991c). In a study comparing the PI scores of psychiatric patients (n = 124) and normal adults (n = 140) who completed the PACL using standard instructions, I (Strack, 1991a) found that 84% of the PI scores T = 60 and above were obtained by patients (only 16% of the normals had scores over 59).

Other investigators have reported expected relationships between PACL scales and a variety of measures. For example, Horton and Retzlaff (1991) correlated the PACL with Moos' Family Environment Scale in a sample of 65 undergraduates. They found that family cohesion and expressiveness were strongly associated with cooperative and sociable personality styles, while conflict was most prevalent in the families of sensitive and forceful persons. High scores on the Respectful scale were linked to family environments in which cohesion, organization, and religiosity were salient features.

Wiggins and Pincus (1989; Pincus & Wiggins, 1990) examined the PACL in the context of MMPI personality disorder scales, Big Five Interpersonal Adjective Scales (IAS-B5), the NEO-PI, and a circumplex version of Horowitz's Inventory of Interpersonal Problems. PACL scales exhibited anticipated relationships with each of the tests in correlational, canonical, and factor analyses. For example, PACL Introversive and Sociable were loaded (in opposite directions) on a factor that included the MMPI Schizoid and Histrionic scales, NEO-PI Extraversion, and IAS-B5 Dominance. PACL Forceful was correlated .59 with interpersonal problems associated with dominance behavior, while PACL Cooperative was correlated .48 with problems involving exploitation by others.

Norms

In keeping with the emphasis on normality, PACL scales were normed as T scores rather than BR scores. Normative data (Strack, 1991c) were obtained from 2507 normal adults between the ages of 16 and 72. Subjects were sampled between 1980-1986 with 90% coming from colleges and 10% from businesses. Men comprised 47.4% of sample and women 52.6%. Ethnic make-up was 65.2% non-Hispanic White, 17.3% Hispanic, 9.1% Black, 7.6% Asian, and 0.8% Native American Indian or Eskimo.

TEST AVAILABILITY

21st Century Assessment, located in South Pasadena, CA (USA), publishes the PACL as a paper-and-pencil measure that can be hand scored or entered into a computer file via optical scanner. Full-color, computerized versions of the check list for DOS (AUTOPACL; Robbins, 1991) and Windows (WinPACL; Robbins, 1994) are also available that permit computer administration of the test, scoring, and printing of profile plots of scores as well as narrative interpretations. These programs allow for unlimited uses on a single computer and, as an aid to researchers, can produce exportable files containing test data for multiple subjects. The narrative interpretations were written by me for use in counseling and personnel settings, and were based on Millon's writings, empirical information obtained during test construction and validation, and clinical experience with the test. Information on how to contact the publisher may be found in the last section of this essay.

COMPARISON WITH THE MCMII

The PACL was designed exclusively on the basis of Millon's (1969/1983) original model of personality and measures normal trait characteristics. This is in contrast to the three editions of the MCMI, which were designed to match DSM Axis II criteria for personality disorders. Additionally, Millon's original model differs somewhat from that found in his more recent writings (1986a, 1986b, 1990, 1994, 1996, 1997; Millon & Davis, 1994).

In accordance with Millon's (1969/1983, 1987, 1994, 1996) model and akin to the MCMI, PACL personality scales contain varying numbers of overlapping items, ranging from one for the Respectful scale to nine for the Sensitive scale. The percentage of overlapping items on PACL scales is substantially lower than that for MCMI scales, and ranges from 5% to 35%. As a result, scale intercorrelations for the PACL are somewhat lower than those for the MCMI (Median r = |.35| across sexes; Strack, 1987, p. 579). Also as a result, PACL scales containing only nonoverlapping items have been found to be quite reliable on their own, and to yield essentially the same factors as the overlapping scales (Pincus & Wiggins, 1990; Wiggins & Pincus, 1989; Strack, 1991c).

In practice, correspondence between the PACL and various versions of the MCMI is reduced by the dissimilar test formats (adjectives versus statements), models used, and focus on normality versus pathology. In spite of these differences, I (Strack, 1991b) found the eight PACL and ten MCMI-II basic personality scales to be correlated between .39 and .67 (Median = .52, using MCMI-II weighted raw scores) in a sample of 65 male and 75 female college students. The lowest values were found for PACL Sensitive/MCMI-II Self-Defeating (.39) and PACL Forceful/MCMI-II Antisocial (.41), suggesting that these MCMI-II scales are not strongly aligned with Millon's original (1969/1983) model. By comparison, the MCMI-II Aggressive scale was correlated .53 with PACL Forceful and the MCMI-II Passive-Aggressive scale was correlated .51 with PACL Sensitive.

Factor analyses of PACL, MCMI-I, and MCMI-II personality scales have revealed very similar results. The three higher-order dimensions found in the PACL (Strack, 1987), that is, Neuroticism, Assertiveness-Aggressiveness, and Social Extraversion-Introversion, correspond to the three factors found by Retzlaff and Gibertini (1987) for MCMI-I basic eight scales among psychiatric patients and normal adults, and by Strack, Lorr, Campbell, and Lamnin (1992) for the 13 MCMI-II personality scales with patients. A joint factor analysis of PACL and MCMI-II basic personality scales among college students also yielded three factors (using residual scores), with corresponding PACL and MCMI-II scales loading on the same dimensions (Strack, 1991b).

Strack, Lorr, and Campbell (1990) examined the circular ordering of MCMI-II personality disorder scales in a mixed group of psychiatric patients and compared results with those from the PACL among normal adults. Plotted against the orthogonally-rotated first two principal components, they found a reasonably good circle for MCMI-II scales (using residual scores) that, for the most part, followed Millon's (1987, p. 20) predictions. Ordering for the PACL scales was similar, although a less complete circle was noted: Sociable, Confident, and Forceful were loaded opposite Introversive, Inhibited, and Sensitive on one dimension, while Cooperative and Respectful defined one end of a second dimension but had no opposing scales.

MILLON'S PERSONALITIES AS MEASURED BY THE PACL

Correlational evidence demonstrates that normal versions of Millon's basic styles are milder variants of the personalities as disorders. Unfortunately, behavioral studies and side-by-side comparisons of matched groups of normals and patients on the PACL and MCMI have not yet been carried out. As a result, important data are still needed to address the precise nature of similarities and differences between normal and disordered forms of Millon's personalities.

With regard to the appearance of Millon's personalities in normal form, what can be offered at this point is a portrait of each style based on Millon's theory, empirical findings from studies associating PACL scales with other measures, and clinical experience with the test. Summaries of empirical findings can be found in Strack (1991c, 1993). The following descriptions represent normal prototypes of persons who obtain high scores on the individual scales. In practice, of course, people are seldom prototypical, instead exhibiting a mixture of traits from multiple styles. Nevertheless, these descriptions flesh-out various aspects of normal personality not readily grasped by extrapolations from Millon's writings on pathological character. Especially noteworthy among the normal styles are their positive dispositional features and interpersonal attitudes. Even less desirable traits are placed within a normal frame of reference.

Scale 1 - Introversive

Aloof, introverted, and solitary, these persons usually prefer distant or limited involvement with others and have little interest in social activities, which they find unrewarding. Appearing to others as nonchalant and untroubled, they are often judged to be easy-going, mild-mannered, quiet, and retiring. They frequently remain in the background of social life and work quietly and unobtrusively at a job. At school or in the workplace these people do well on their own, are typically dependable and reliable, are nondemanding, and are seldom bothered by noise or commotion around them. They are often viewed as level-headed and calm. However, these individuals may appear unaware of, or insensitive to, the feelings and thoughts of others. These characteristics are sometimes interpreted by others as signs of indifference or rejection, but reveal a sincere difficulty in being able to sense others' moods and needs. Introversive persons can be slow and methodical in demeanor, lack spontaneity and resonance, and be awkward or timid in social or group situations. They frequently view themselves as being simple and unsophisticated, and are usually modest in appraising their own skills and abilities. At the same time, their placid demeanor and ability to weather ups and downs without being ruffled are traits frequently prized by friends, family members, and co-workers.

Scale 2 - Inhibited

As with the introversive style, the inhibited personality is marked by a tendency toward social withdrawal. However, for inhibited individuals this pattern is motivated not by disinterest, but by a fear of negative consequences. Inhibited persons tend to be sensitive to their own feelings and to those of others. They often anticipate that others will be critical or rejecting of them, and because of this they frequently seem shy or skittish in unfamiliar surroundings. In this regard, family members and acquaintances may see them as being unnecessarily nervous, wary, and fearful. Although inhibited persons tend to get along reasonably well with others, they are often difficult to get to know on a personal level. These individuals usually wish that they could be at ease with others and tend to desire closeness, but they often are just too uncertain of the consequences of closeness and intimacy to let their guard down. As a result, they may experience feelings of loneliness, but be unable or unwilling to do anything about them. Because of their sensitivity to others, inhibited persons are often described as kind, considerate, and empathic by close acquaintances. Inhibited persons often prefer to work alone or in a small group with people they can come to know well. They do best in a stable work environment where stimulation and commotion are kept at low to moderate levels. Persons working with inhibited types need to appreciate their sensitivity to both positive and negative feedback, as well as their need to build trust over a long period of time.

Scale 3 - Cooperative

Cooperative persons can be identified by a need for approval and affection, and by a willingness to live in accord with the desires of others. They usually adapt their behavior to the standards of others but in the process may deny their own needs. Interpersonally, these individuals are often cooperative, reliable, considerate of others, and deferential. They may appear even-tempered, docile, obliging, self-effacing, ingratiating, or naive. Cooperative individuals often see themselves as being modestly endowed in terms of skills and abilities. They are often pleased when they can rely on others and may feel insecure when left on their own. Especially when faced with difficult or stressful situations, cooperative persons seek others to provide authority, leadership, and direction. They often prefer group work environments and will typically excel in them if given support and guidance. They are usually willing to follow directions and cooperate with co-workers in team efforts.

Scale 4 - Sociable

Like cooperative personalities, sociable individuals have a need for attention and approval. However, unlike cooperative persons sociable types take the initiative in assuring their reinforcements by being center-stage. They are characterized by an outgoing, talkative, and extroverted style of behavior and tend to be lively, dramatic, and colorful. These people are typically viewed by others as spontaneous, clever, enthusiastic, and vigorous. They can be quite sensitive to the needs and wants of others, at least to those aspects which will help them get the attention they seek. Sociable individuals may also be seen as fickle in their attachments. They may have quickly shifting moods and emotions, and may come across as shallow and ungenuine. These persons tend to prefer novelty and excitement, and are bored by ordinary or mundane activities. Like cooperative personalities, sociable individuals seem uncomfortable or deflated when left on their own. Not surprisingly, sociable types often excel in group work environments where they can excerise their showy style. They often do well interacting with the public, may be skilled and adept at rallying or motivating others, and will usually put their best side forward even in difficult circumstances.

Scale 5 - Confident

Aloof, calm, and confident, these personalities tend to be egocentric and self-reliant. They may have a keen sense of their own importance, uniqueness, or entitlement. Confident individuals enjoy others' attention and may be quite bold socially, although they are seldom garish. They can be self-centered to a fault and may become so preoccupied with themselves that they lack concern and empathy for others. These persons have a tendency to believe that others share, or should share, their sense of worth. As a result, they may expect others to submit to their wishes and desires, and to cater to them. Ironically, the confident individual's secure appearance may cover feelings of personal inadequacy and a sensitivity to criticism and rejection. Unfortunately, they usually do not permit others to see their vulnerable side. When feeling exposed or undermined these individuals are frequently disdainful, obstructive, or vindictive. In the workplace, confident persons like to take charge in an emphatic manner, often doing so in a way that instills confidence in others. Their self-assurance, wit, and charm often win them supervisory and leadership positions.

Scale 6 - Forceful

Like confident persons, forceful individuals can be identified by an inclination to turn toward the self as the primary source of gratification. However, instead of the confident personality's internalized sense of self-importance, forceful people seem driven to prove their worthiness. They are characterized by an assertive, dominant, and tough-minded personal style. They tend to be strong-willed, ambitious, competitive, and self-determined. Feeling that the world is a harsh place where exploitiveness is needed to assure success, forceful individuals are frequently gruff and insensitive in dealing with others. In contrast to their preferred, outwardly powerful appearance, these individuals may feel inwardly insecure and be afraid of letting down their guard. In work settings, these personalities are often driven to excel. They work hard to achieve their goals, are competitive, and do well where they can take control or work independently. In supervisory or leadership positions these persons usually take charge and see to it that a job gets done. However, they often need to temper an inclination to demand as much of others as they do of themselves.

Scale 7 - Respectful

Responsible, industrious, and respectful of authority, these individuals tend to be conforming and work hard to uphold rules and regulations. They have a need for order and are typically conventional in their interests. These individuals can be rule-abiding to a fault, however, and may be perfectionistic, inflexible, and judgmental. A formal interpersonal style and notable constriction of affect can make some respectful persons seem cold, aloof, and withholding. Underneath their social propriety there is often a fear of disapproval and rejection, or a sense of guilt over perceived shortcomings. Indecisiveness and an inability to take charge may be evident in some of these persons due to a fear of being wrong. However, among co-workers and friends, respectful personalities are best known for being well-organized, reliable, and diligent. They have a strong sense of duty and loyalty, are cooperative in group efforts, show persistence even in difficult circumstances, and work well under supervision.

Scale 8 - Sensitive

Sensitive personalities tend to be unconventional and individualistic in their response to the world. They march to the beat of a different drummer and are frequently unhappy with the status quo. They may be quick to challenge rules or authority deemed arbitrary and unjust. They may also harbor resentment without expressing it directly and may revert to passive-aggressive behavior to make their feelings known. Many sensitive people feel as if they don't fit in, and view themselves as lacking in interpersonal skills. In fact, to others they often appear awkward, nervous, or distracted, and seem angry or dissatisfied with themselves and others. They can be indecisive and have fluctuating moods and interests. An air of uncertainty and general dissatisfaction may reflect an underlying dependency and sense of personal inadequacy. With their best side forward, sensitive persons can be spontaneous, creative, and willing to speak out for what they believe in. These qualities make them especially suited to jobs that are not rule-bound, that give them a certain independence from supervision, and that require unusual duties or creative expression.

Scale 9 - Problem Indicator

Items for this scale were compiled from adjectives measuring the schizoid, cycloid, and paranoid personalities, for example, "chaotic," "fragmented," "depressed," and "suspicious." While the scale does not define a personality style, high scores are indicative of personality problems and the potential for disorder. High scorers possess personality disorder traits and symptoms such as low ego-strength and affective instability. They are likely to appear anxious, dysphoric, and fearful, exhibit strong self-doubt, and express dissatisfaction with themselves and others. They may have long-standing adjustment problems in major areas of life such as work, school, and relationships. Those who score high on this scale are not likely to fit the same picture of normality as are low scorers (e.g., by exhibiting interpersonal rigidity and maladaptiveness), but further assessment is advised before drawing conclusions regarding the presence of a disorder.

CLINICAL AND RESEARCH USES

Clinical Applications

The PACL is appropriate for use with persons 16 years-of-age and older who read at minimally the 8th-grade level. It has been successfully employed by therapists working in high school and college counseling centers and employee assistance programs; by vocational counselors, personnel psychologists, marriage and family counselors; by therapists doing custody and worker's compensation evaluations; and by general practitioners who work with a variety of clients. Because the PACL is quick and easy to administer (5-10 mins.), it is often given during initial screening visits to assess personality style and identify persons who may have more serious character problems. Clinicians have found it to be useful with people who can't or won't complete questionnaire measures, for example, some medical patients, teen-agers, and the elderly.

A number of clinicians use the PACL as a rating instrument to assess their clients' personality styles, and to have couples and family members assess each other. I have experimented with the PACL in these areas, utilizing the norms in the PACL Manual (Strack, 1991c) for scoring. Although the norms are based on self-reports, they worked remarkably well with a variety of ratings.

An important factor to keep in mind when using the PACL in clinical settings is that the test measures normal trait characteristics, not personality disorder features. High scores on the PACL indicate that an individual possesses more of the traits of a particular normal personality style than other adults in the general population. For example, the higher an individual's score is above T = 50 on any particular scale, the more likely it is that he or she will fit the prototype descriptions given earlier. The test won't assess disordered personality features beyond those measured by the PI scale.

Research

Because the PACL is fast and easy to administer, researchers have found that it can be easily added to a test battery without taxing subjects' time and attentional resources. Concerning important research uses, I encourage studies that further explicate Millon's personalities in normal form, and examine major premises of his model thought to differentiate normal and abnormal persons (Millon, 1969/1983; Millon & Davis, 1994; Strack, 1991c). In this regard, there may be some value in using the PACL with psychiatric samples and/or combining the PACL with the MCMI. PACL item responses may provide information about the normal characteristics of psychiatric patients not tapped by the MCMI, and the scale scores can show how subjects deviate from a normal mean, something MCMI BR scores cannot do.

There is still much to be learned about the appearance and function of Millon's basic personalities in normal form, for example, their vocational interests (Strack, 1994), work behavior, and coping styles. Correlational research employing measures pertinent to Millon's constructs will be helpful, as will experimental and rating studies that focus on real-life behavior. In addition to the regular PACL scale scores, investigators may use scores from nonoverlapping scales (Strack, 1991c, p. 68), factor scores derived from the scales (Strack, 1987) or items (Strack & Lorr, 1990a), and cluster profiles (Strack, 1992).

Millon's ideas about the differences between normal and abnormal personalities are important targets for research. At this point we simply don't know whether disordered styles are less interpersonally flexible and stable, and more pathogenic, than normal types. Likewise, there is no research information available concerning the relative position of normal and abnormal types on Millon's active-passive, pleasure-pain, and self-other polarities.

In describing personality development, Millon (e.g., 1969/1983, 1981, 1990, 1996) emphasized a number of individual difference and process elements thought to be influential in creating either normal or dysfunctional character, for example, biological predisposition (including temperament), early learning experiences, and parent-child relations. Many of these elements are central to his model and deserve careful scrutiny in both retrospective and longitudinal investigations.

Cross-cultural research is needed to assess the similarities and differences in trait structure of Millon's personalities based on cultural influences. As an interpersonal model, one would expect cultural differences to be evident among Millon's personalities at the nomothetic (normative) level of measurement. From an idiographic perspective, one would also expect test interpretation to differ somewhat by culture. For example, in mainstream America independence is valued more than dependence. This may be reflected in our assessment of an individual's strengths and weaknesses when, for example, we suggest that a confident and forceful style will help a person get ahead in business while a cooperative style may be a liability in the same environment. In other cultures dependence and cooperativeness are more highly valued than traits of independence, and changes in test interpretation would follow. The PACL is now available for research use in Chinese, Dutch, Italian, Portuguese, and Spanish languages, and I strongly support cross-cultural investigations of any kind.

FOR MORE INFORMATION OR TO ORDER THE PACL

Inquiries about the PACL are welcome, and may be directed to the publisher or the author. For current pricing of PACL test products and to place an order, please contact the publisher. VISA and MASTERCARD are accepted. Shipping is available to all international locations.

Publisher: 21st Century Assessment
P.O. Box 608
South Pasadena, CA 91031-0608

Telephone: (800) 374-2100
FAX: (626) 441-0614

Author: Stephen Strack, PhD
Psychology Service (116B)
VA Outpatient Clinic
351 E. Temple Street
Los Angeles, CA 90012

Telephone: (213) 253-2677 ext. 4752
FAX: (213) 253-5041
E-Mail: SNSTRACK@AOL.COM

REFERENCES

American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders (4th ed.). Washington, DC: Author.

Chung, Y. (1993). Relationship between types and degree of acculturation and Millon's personality types of Koreans in the United States. Unpublished doctoral dissertation, California School of Professional Psychology, Los Angeles.

Durff, T. (1994). The relationship between Millon's basic personality styles and the cognitive style of field dependence-independence. Unpublished doctoral dissertation, California School of Professional Psychology, Los Angeles.

Gough, H.G., & Heilbrun, A.B. (1983). The Adjective Check List manual (1983 edition). Palo Alto, CA: Consulting Psychologists Press.

Horton, A.D., & Retzlaff, P.D. (1991). Family assessment: Toward DSM-III-R relevancy. Journal of Clinical Psychology, 47, 94-100.

LaForge, R., & Suczek, R.F. (1955). The interpersonal dimensions of personality: III. An interpersonal check list. Journal of Personality, 24, 94-112.

Loevinger, J. (1957). Objective tests as instruments of psychological theory. Psychological Reports, 3, 635-694.

Millon, T. (1981). Disorders of personality. New York: Wiley.

Millon, T. (1983). Modern psychopathology. Prospect Heights, IL: Waveland Press. (Original work published 1969).

Millon, T. (1986a). A theoretical derivation of pathological personalities. In T. Millon & G.L. Klerman (Eds.), Contemporary directions in psychopathology: Toward the DSM-IV (pp. 639-670). New York: Guilford.

Millon, T. (1986b). Personality prototypes and their diagnostic criteria. In T. Millon & G.L. Klerman (Eds.), Contemporary directions in psychopathology: Toward the DSM-IV (pp. 639-670). New York: Guilford.

Millon, T. (1987). Manual for the MCMI-II (2nd Ed.). Minneapolis, MN: National Computer Systems.

Millon, T. (1990). Toward a new personology. New York: Wiley.

Millon, T. (1994). Millon index of personality styles manual. San Antonio, TX: Psychological Corporation.

Millon, T. (1996). Disorders of personality (2nd ed). New York: Wiley.

Millon, T. (1997). MCMI-III manual (2nd ed). Minneapolis, MN: National Computer Systems.

Millon, T., & Davis, R.D. (1994). Millon's evolutionary model of normal and abnormal personality: Theory and measures. In S. Strack & M. Lorr (Eds.), Differentiating normal and abnormal personality (pp. 79-113). New York: Springer.

Pincus, A.L., & Wiggins, J.S. (1990). Interpersonal problems and conceptions of personality disorders. Journal of Personality Disorders, 4, 342-352.

Retzlaff, P.D., & Gibertini, M. (1987). Factor structure of the MCMI basic personality scales and common-item artifact. Journal of Personality Assessment, 51, 588-594.

Robbins, B. (1991). AUTOPACL user's guide. So. Pasadena, CA: 21st Century Assessment.

Robbins, B. (1994). WinPACL user's guide. So. Pasadena, CA: 21st Century Assessment.

Strack, S. (1981). Development and preliminary validation of the personality adjective check list. Unpublished manuscript.

Strack, S. (1987). Development and validation of an adjective check list to assess the Millon personality types in a normal population. Journal of Personality Assessment, 51, 572-587.

Strack, S. (1991a). Comparison of PACL PI scale elevations in samples of psychiatric patients and normal adults. Unpublished manuscript.

Strack, S. (1991b). Factor analysis of MCMI-II and PACL basic personality scales in a college sample. Journal of Personality Assessment, 57, 345-355.

Strack, S. (1991c). Manual for the personality adjective check list (PACL) (Rev.). So. Pasadena, CA: 21st Century Assessment.

Strack, S. (1992). Profile clusters for men and women on the Personality Adjective Check List. Journal of Personality Assessment, 59, 204-217.

Strack, S. (1993). Measuring Millon's personality styles in normal adults. In R.J. Craig (Ed.), The Millon clinical cultiaxial inventory: A clinical research information synthesis (pp. 253-278). Hillsdale, NJ: Lawrence Erlbaum Associates.

Strack, S. (1994). Relating Millon's basic personality styles and Holland's occupational types. Journal of Vocational Behavior, 45, 41-54.

Strack, S. (1997). The PACL: Gauging normal personality styles. In T. Millon (Ed.), The Millon inventories: Clinical and personality assessment, pp. 477-497. New York: Guilford.

Strack, S., & Lorr, M. (1990a). Item factor structure of the personality adjective check list (PACL). Journal of Personality Assessment, 55, 86-94.

Strack, S., & Lorr, M. (1990b). Three approaches to interpersonal behavior and their common factors. Journal of Personality Assessment, 54, 782-790.

Strack, S., Lorr, M., & Campbell, L. (1989, August). Similarities in Millon personality styles among normals and psychiatric patients. Paper presented at the annual convention of the American Psychological Association, New Orleans, LA.

Strack, S., Lorr, M., & Campbell, L. (1990). An evaluation of Millon's circular model of personality disorders. Journal of Personality Disorders, 4, 353-361.

Strack, S., Lorr, M., Campbell, L., & Lamnin, A. (1992). Personality and clinical syndrome factors of MCMI-II scales. Journal of Personality Disorders, 6, 40-52.

Wiggins, J.S., & Pincus, A.L. (1989). Conceptions of personality disorders and dimensions of personality. JCCP: Psychological Assessment, 1, 305-316.

Wiggins, J.S., & Pincus, A.L. (1994). Personality structure and the structure of personality disorders. In P.T. Costa & T.A. Widiger (Eds.), Personality disorders and the five-factor model of personality (pp. 73-93). Washington, D.C.: American Psychological Association.