Korean Cultural Influences on the Millon Clinical Multiaxial Inventory III.

The purpose of this study was to investigate the effect of Korean culture on the results of the Millon Clinical Multiaxial Inventory-III by comparing profiles of 147 Korean and 132 American college students. The Korean and American college student samples differed significantly on seven scales: Schizoid, Avoidant, Depressive, Dependent, Histrionic, Aggressive, and Self-Defeating. The results indicate that MCMI-III personality profile differences exist between Korean and American college students. Treatment implications for mental health counselors are discussed.

Counselors conducting personality assessments of clients from diverse cultural backgrounds are faced with the formidable task of evaluating the influence of culture and ethnicity on client personality structure and functioning (Okazaki & Sue, 1995). Characteristics such as dependence and external locus of control that are seen as maladaptive within Western culture may be adaptive and appropriate within a different culture (Sue & Sue, 1987). There is increasing awareness of cultural and ethnic differences in the assessments and treatments of clients (Glass, Bieber, & Tkachuck, 1996; Munley, Vacha-Haase, Busby, & Paul, 1998; Okazaki & Sue, 1995).

Culture is the accumulation of behavioral patterns, attitudes, and symbols of individuals over time; it determines behavioral patterns, thoughts, and attitudes (Escovar, 1997). According to Kluckhohn (1954), culture is to society as personality is to the person. Cultural patterns are shared whereas personality traits are private. People are born into cultures that provide tried and proven formulas. The unfolding of personality traits and their relative salience is determined partially by cultural norms and practices. Therefore, culture is a critical factor influencing the development of individual personality.

The Millon Clinical Multiaxial Inventory III (MCMI-III; Millon, 1997) has become a popular tool for personality assessment; it was included among the five assessment instruments selected for review in the Journal of Personality Assessment (Schinka & Green, 1997). Recent survey data suggest that the MCMI ranks among the most popular tests in use by mental health professionals (Pitrowski, 1997). Furthermore, the MCMI has been one of the three most widely researched personality instruments in recent years (Choca & Van Denburg, 1997; Pitrowski & Keller, 1989; Pitrowski & Lubin, 1990). However, little research has been conducted to examine the effect of culture on the MCMI-III (Choca & Van Denburg, 1997). It is an ethical (American Mental Health Counselors Association, 2000) and professional (American Educational Research Association, American Psychological Association, & National Council on Measurement in Education, 1999; Prediger, 1993) imperative to learn more about the validity of the MCMI-III for diverse cultural groups.

This study focused on the influence of Korean culture on the MCMI-III. According to census data, there were approximately 900,000 Koreans or Korean-Americans living in the United States in 1990 (Bureau of the Census, 1995). This population has grown to well over one million during the past decade. Therefore, it is likely that mental health counselors will work with clients of Korean descent. It is important to learn more about the influence of Korean culture on personality assessment in general and the MCMI-III in particular.

EMPIRICAL FINDINGS

The authors were able to identify only two studies regarding Korean personality assessment. Eysenck and Lee (1985) compared Korean and English adults using the Eysenck Personality Questionnaire. They found that Korean adults scored higher than English adults on the Psychotism, Neuroticism, and Lie scales and lower on the Extroversion scale. In the second study, Han (1996) compared the Minnesota Multiphasic Personality Inventory-2 (Hathaway & McKinley, 1989) profiles of Korean, American, and Japanese college student samples. Han reported that all Korean mean scores, except the Lie and infrequency scales, fell above the means for the American normative sample. The T scores for Korean men on the Depression and Schizophrenia scales were 10 or more points higher than those of the American and-Japanese samples. In the Korean male sample, the Schizophrenia scale was most elevated with a T score of 70. The findings for the female sample were similar to those of the male sample. These studies indicate that Korean scores are elevated on these two measures of personality pathology, which raises the concern that normal Korean behavior may be interpreted as pathological if counselors are not sensitive to the limitations of use of psychological assessments with culturally diverse populations. These findings also suggest that Koreans can be expected to demonstrate MCMI-III profile patterns that differ from those of Americans.

Recent studies also indicate that significant racial differences occur in MCMI profiles (Choca et al., 1992). Further, some MMP-2 scales seem to be less effective in distinguishing psychiatric from nonpsychiatric samples in ethnic groups (Walters, Greene, Jeffrey, Kruzich, & Haskin, 1983). It appears that mental health counselors have to better understand the role of culture in personality assessment. This can be accomplished by developing normative data for popular personality assessments with diverse cultural groups. The purpose of this study was to investigate the effect of Korean culture on the results of the MCMI-III. The MCMI-III is an apt choice because there is so little information regarding the influence of culture on this assessment tool and because Millon's (1990) theory can be used to develop a priori hypotheses regarding the influence of culture. Such studies should ultimately enable counselors to provide clients with more accurate assessment interpretations (Butcher, Narikiyo, & Vitousek, 1992; Han, 1996) and more effective treatment plans. Millon's personality theory will be reviewed next to establish the rationale for our hypotheses.

OVERVIEW OF MILLON'S PERSONALITY THEORY

In Millon's system, personality types can be conceptualized along three bipolar dimensions: Self-other, passive-active, and pleasure-pain (Strack, 1994). The self-other dimension reflects a tendency to find gratification in the self or from external sources. The pleasure-pain dimension reflects orientation toward experiences that are attractive and positively reinforcing or toward those that are aversive and negatively reinforcing. The passive-active dimension reflects the tendency either to take initiative in shaping or reacting to life events.

Millon and Davis (1996) used this framework to derive the various personality patterns. For example, the Schizoid prototype has a strong passive tendency and an average self-other tendency. Dependent and Compulsive prototypes have strong passive and other tendencies and average pleasure-pain tendencies. The Histrionic prototype has strong active and other tendencies and an average pleasure-pain tendency. The Narcissistic prototype has strong passive and self tendencies and an average pleasure-pain tendency. Each personality type has a distinct combination of the three bipolar dimensions.

EXTENSION OF MILLON'S THEORY TO THE SOCIOCULTURAL LEVEL

Escovar (1997) extended Millon's personality theory to examine the functional mechanisms that underlie human behavior within culture. Escovar's (1997) sociocultural analysis of Millon's bipolar dimensions were used to formulate the hypotheses.

Self Versus Other

One of the well-known distinctions between European-American and Asian cultures is individualism-collectivism (Triandis, 1994). According to Escovar (1997), individualism-collectivism parallels Millon's self-other dimension in the extent to which cultural groups give priority to collective or individual goals. Collectivist cultures emphasize the views, needs, and goals of the group; and individualistic cultures give priority to the self. In addition, collectivists are other-oriented while individualists are self-oriented. Therefore, collectivists tend to focus more on the person than on the content of social communication; they are more concerned with the other's emotional reaction than to the content of the communication. Because collectivists give high priority to maintaining good relationships, they are likely to suppress negative communications and to tell others what they want to hear in order to avoid negative feelings and conflicts (Triandis, 1994). Collectivists, for example, are less likely to say no, which leads them to be less assertive than individualists. In sum, collectivists tend to be perceived as more emotionally dependent on other group members, lacking in social skills, and more other-oriented than individualists. These different priorities result in different behavioral patterns and personality development.

Passive Versus Active

The second bipolar dimension in Millon's model pertains to the manner in which people adapt to their surroundings. The passive dimension refers to an accommodating style that emphasizes fitting into the environment. The active dimension refers to a more energetic style that emphasizes actively modifying existing conditions. Some cultural groups have a passive orientation and prefer to preserve their environment and change slowly, while other groups have an active orientation that leads to a more dynamic and rapidly changing society.

Pain Versus Pleasure

Millon's third bipolar dimension refers to a preference for enhancing life and experiencing pleasure or for preserving life and avoiding pain. Escovar (1997) differentiated cultures as malevolent versus benevolent to parallel Millon's pain-pleasure dimension. Malevolence refers to the practice of using pain as the primary mechanism to control, suppress, and limit behavior. Other cultural groups implement benevolent practices to foster pleasure-seeking behaviors by their members.

MILLON'S PERSONALITY THEORY AND KOREANS

Escovar's (1997) extension of Millon's personality theory was used to categorize Korean culture on two of these three bipolar dimensions. There have been numerous studies indicating that African, Asian, and Latin American cultures emphasize collectivism (Escovar, 1997). A recent study by Liem and Lim (2000) found that Asian Americans exhibit more other-focused qualities in their emotional life than European Americans. Gudykunst, Yoon, and Nishida (1987) found that Koreans were highly collectivist, that Japanese were somewhat collectivist, and Americans were highly individualistic. The authors expected to see a collectivist orientation reflected in the personality profiles of Koreans. Regarding the passive-active dimension, Asians value harmony and fitting into one's surroundings whereas European-Americans value a controlling nature (Katz, 1985; Pedersen, 1988; Sue & Sue, 1990).This tendency to adapt to the environment suggests a more passive orientation for Asian cultures. The authors expected the Korean personality profile to reflect a passive orientation. The empirical findings regarding the pleasure-pain dimension in Korean culture are limited. Therefore, the pleasure-pain dimension was not incorporated into the hypotheses.

These cultural orientations should be reflected in MCMI-III profile scores for Koreans that differ from those of Americans. It was specifically hypothesized that the Korean sample would score higher than the American sample on the MCMI-III Dependent and Compulsive scales because they reflect other-oriented and passive tendencies.

METHOD

Participants

Participants were 147 (90 female, 57 male) students from a major university in Seoul, Korea. Mean ages for female and male participants were 20.93 and 22.37, respectively. Because the MCMI-III has not been translated into Korean, these participants were able to read English. All students majored in either English or English Education. This university is well known for the high quality of its English training system.

Some 132 (81 female, 51 male) American college student volunteers were administered the MCMI-III. Of 132, 126 were European-American, one was African American, three were Asian Americans, and two were Latin Americans. Mean ages for American female and male participants were 20.19 and 20.98, respectively.

Procedures

The Korean participants were recruited from English reading, laboratory, and phonology classes in 1998. The American participants were recruited from career development and group counseling classes. The American students participated voluntarily in November, 1999, and in April, 2000, respectively. All students recorded their ages, year of college, and gender.

Measure

The MCMI-III has 24 scales, including 11 clinical personality patterns (Schizoid, Avoidant, Depressive, Dependent, Histrionic, Narcissistic, Antisocial, Aggressive, Compulsive, Passive-Aggressive, Self-Defeating), three severe personality pathology (Schizotypal, Borderline, and Paranoid), seven clinical syndrome (Anxiety, Somatoform, Bipolar: Manic, Dysthymia, Alcohol Dependence, Drug Dependence, Post-Traumatic Stress Disorder), and three severe clinical syndromes (Thought Disorder, Major Depression, and Delusional Disorder) scales.

According to Millon (1997), the clinical personality pattern scales measure neurosis; and the severe personality pathology, clinical syndrome, and severe clinical syndrome scales are more relevant for inpatient population and are not appropriate for use with outpatient populations. Choca et al. (1992) stated that the original clinical personality patterns measure personality styles rather than disorders. This study focused on the 11 clinical personality patterns as dependent variables because the participants were all college students who are considered a normal population. Raw scores were chosen for analysis rather than the more commonly applied base rate (BR) scores because BR scores are calculated with prevalence rates for personality disorders found in clinical settings that do not apply to normal adults.

The MCMI-III consists of 175 items that assess a wide range of clinically relevant behaviors (Millon, 1997). The level of reading is eighth grade; the majority of respondents can complete the MCMI-III in 20 to 30 minutes. The retest reliabilities for each scale range from .82 to .96, and the median stability coefficient is .91.

The MCMI-I and MCMI-II were developed to assess and differentiate personality disorders (Axis II), as described by the DSM-III-R, (1987) from clinical (Axis I) disorders (Millon, 1987). Millon (1983,1987,1994,1997) argued that the MCMI should only be used with a clinical population, because it was standardized on a psychiatric population. However, there has been considerable discussion as to whether the MCMI measures a personality disorder or style, because nonclinical populations often obtain clinically significant scores (Choca et al., 1992). Several researchers have used the MCMI with normal populations (e.g., Strack, 1991, 1994). Millon (1992) views personality style and functioning on a continuum ranging from normal to abnormal and states that it is difficult to determine where a style becomes a disorder. Therefore, some experts have suggested that the MCMI can be used with a nonpsychiatric population as long as the counselor is mindful of the group with which the client is being compared (Choca et al., 1993; Choca & Van Denburg, 1997).

Independent Variable

Cultural group, Korean or American, was the sole independent variable.

RESULTS

A one-way multivariate analysis of variance (MANOVA) was conducted in order to compare 11 MCMI-III clinical personality pattern scales for the Korean and American samples. The MANOVA was used to reduce Type I error. This analysis resulted in a significant multivariate effect for Culture, F(13, 265) = 18.934, p [is less than] 001. Subsequent univariate analyses of variance resulted in significant Culture effects for seven of the 11 dependent variables: Schizoid, Avoidant, Depressive, Dependent, Histrionic, Aggressive, and Self-Defeating scales. See Table for the means and standard deviations for the two groups and F values for each group comparison.

Table 1. Means, Standard Deviations, and F-values for MCMI-III
Scales

                      Americans       Koreans
                      (n = 132)      (n = 147)

MCMI-III Scale         M     SD      M      SD        F

Schizoid             3.37   3.05    7.43    4.17    84.60(***)
Avoidant             3.99   3.94    7.69    5.25    43.55(***)
Depressive           4.12   4.17    5.61    5.23     6.82(**)
Dependent            7.17   4.53    8.35    4.93     4.30(*)
Histrionic           18.14  4.50   11.41    4.78   145.97(***)
Narcissistic         15.06  4.15   14.05    4.97     3.38
Antisocial           5.78   4.14    5.86    3.33     0.04
Aggressive           6.38   4.68    8.23    4.75    10.72(***)
Compulsive           14.73  4.72   14.87    4.11     0.07
Passive-Aggressive   6.65   4.53    7.71    5.18     3.29
Self-Defeating       2.73   3.47    5.20    4.40    26.54(***)

(*) p < .05,

(**) p < .01,

(***) p < .001

The results supported the first hypothesis, as the Korean students scored higher on the Dependent scale. The second hypothesis was not supported as there was no difference between the two groups on the Compulsive scale.

DISCUSSION

The results showed general support for the expectation that Korean and American students would differ in the MCMI-III profiles; the two groups differed in seven of the 11 scales. The first hypothesis was substantiated, as the Korean sample scored significantly higher on the Dependent scale that reflects other and passive personality orientations. Contrary to expectations, there was no difference in the Compulsive scale. The Korean group also scored higher on the Schizoid, Avoidant, Depressive, Aggressive, and Self-Defeating scales. They scored much lower than the American group on the Histrionic scale. These findings were generally consistent with our expectation that Korean personality would reflect orientation towards others and passivity. The Avoidant, Depressive, Dependent, and Self-Defeating scales all reflect orientation towards others. The Schizoid, Dependent, and Depressive scales reflect a passive orientation; Koreans also scored significantly lower on the Histrionic scale, which reflects an active orientation to life.

Three of the differences were inconsistent with the general predictions derived from Millon's theory. First, the Korean group scored higher on Avoidant, which reflects an active orientation. However, the desire to avoid conflict and strife appears to be consistent with the general emphasis on harmony in Asian culture. The significantly higher Aggressive and Self-Defeating scores are more difficult to explain. Both reflect an active and pain-oriented approach to life; those with aggressive personalities focus their pain on others and those with self-defeating personalities direct the experience of pain to the self. These findings suggest that Korean culture allows for the resolution of psychological conflict through inflicting pain on self or others. This finding was unanticipated and should be the focus of subsequent research on personality functioning in the Korean culture.

When trying to understand and interpret differences in personality across cultures, it is important to consider how much the MCMI-III measures cultural versus individual differences. A universal model of mental health cannot be applied when different cultural norms exist. It is important to remember that these cultural differences are relative and not descriptive of levels of pathology (Glass et al., 1996). The authors interpret the findings as indicating that there are significant differences in Korean and American culture that are reflected in response to the MCMI-III.

These results indicate that counselors have to be extremely cautious in interpreting MCMI-III results for Koreans who have immigrated to the United States and for first- and second-generation Korean Americans. The MCMI-III should not be used as the sole or primary evidence for diagnosis of a personality disorder with this population. Moreover, counselors need to understand that orientation towards others and passivity are important parts of the Korean culture. Therefore, counselors should not interpret these characteristics as maladaptive or pathological until they have a full appreciation of the ways in which these orientations to life are healthy and adaptive in Korean culture. Further, counselors should consider using counseling approaches such as Morita therapy (Kondo, 1953; Morita, 1976) that employ a more passive and other-oriented approach to problem solving.

This study had three major limitations. First, convenience samples were used to form both the Korean and American groups. These samples were not systematically stratified for pertinent variables such as socioeconomic status and religiosity. Second, the Koreans may have interpreted MCMI-III items differently than native English speakers, even though they were English or English Education majors. Future research should employ a translated version of the MCMI-III derived with appropriate back-translation procedures. Finally, these Korean students may not have been representative of Koreans in general; there may have been some idiosyncratic characteristics of students highly motivated to speak English that was reflected in their MCMI-III responses. However, the current study is instructive, raising awareness that counselors need to use extreme caution in interpreting MCMI-III results in counseling Korean clients.

Ignorance of culture and lack of training in cross-cultural assessment can lead to misdiagnosis and ineffective counseling treatment planning. Therefore, continuous efforts should be made to explore cultural influences on all psychological assessments, including the MCMI-III.

REFERENCES

American Educational Research Association, American Psychological Association, & National Council on Measurement in Education. (1999). Standards for educational and psychological testing (2nd ed.). Washington, DC: American Educational Research Association.

American Mental Health Counselors Association. (2000). Code of ethics of the American Mental Health Counselors Association (2000 rev.). Journal of Mental Health Counseling, 23, 1-18.

American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed. rev.). Washington, DC: Author.

Bureau of the Census. (1995). Population: Race and Hispanic origin for the United States, 1990 and 1980. Washington, DC: United States Department of Commerce.

Butcher, J. N., Narikiyo, T., & Vitousek, K. B. (1992). Understanding abnormal behavior in cultural context. In P. B. Sutker & H. E. Adams (Eds.), Comprehensive handbook of psychopathology (2nd ed.; pp. 83-105). New York: Plenum Press.

Choca, J. P., Shanley, L. A., Van Denburg, E. V., Agresti, A:, Mouton, A., & Vidger, L. U. (1992). Personality disorder or personality style: That is the question. Journal of Counseling and Development, 70, 429-431.

Choca, J. P., & Van Denburg, E. V. (1997). Interpretive guide to the Millon Clinical Multiaxial Inventory (2nd ed.). Washington, DC: American Psychological Association.

Escovar, L. A. (1997). The Millon inventories: Clinical and personality assessment. In T. Millon (Ed.), The Millon inventories: Sociocultural considerations (pp. 264-285). New York, NY: Guilford Press.

Eysenck, S. B., & Lee, H. (1985). Cross-cultural study on the characteristics of personality dimensions: Korean and English. Korean Journal of Psychology, 5, 51-66.

Glass, M. H., Bieber, S. L., & Tkachuck, M. J. (1996). Personality styles and dynamics of Alaska Native and nonnative incarcerated men. Journal of Personality Assessment, 66, 583-603.

Gudykunst, W. C., Yoon, Y., & Nishida, T. (1987). The influence of individualism-collectivism on perceptions of communication in ingroup and outgroups relationships. Communication Monographs, 54, 295-306.

Han, K. (1996). International adaptations of the MMPI-2: Research and clinical applications. In J. N. Butcher (Ed.), The Korean MMPI-2 (pp. 88-135). Minneapolis, MN: University of Minnesota.

Hathaway, S. R., & McKinley, J. C. (1989). Minnesota Multiphasic Personality Inventory-2. Minneapolis: The University of Minnesota Press.

Katz, J. (1985). The sociopolitical nature of counseling. The Counseling Psychologist, 13, 615-624.

Kluckhohn, F. (1954). Culture and behavior. In G. Lindsay (Ed.), Handbook of social psychology (Vol. 2; pp. 921-976). Reading, MA: Addison-Wesley.

Kondo, A. (1953). Morita therapy: A Japanese therapy for neurosis. American Journal of Psychoanalysis, 13, 31-37.

Liem, R., & Lim, B. A. (2000). Acculturation and emotion among Asian Americans. Cultural Diversity and Ethnic Minority Psychology, 6, 13-31.

Millon, T. (1983). Millon Clinical Multiaxial Inventory manual (3rd ed.). Minneapolis, MN: National Computer Systems.

Millon, T. (1987). Millon Clinical Multiaxial Inventory: II manual. Minneapolis, MN: National Computer Systems.

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

Millon, T. (1992). An appreciative rejoiner. Journal of Counseling and Development, 70, 432-433.

Millon, T. (1994). Manual for the MCMI-II (2nd ed.). Minneapolis, MN: National Computer Systems.

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

Millon, T., & Davis, R. D. (1996). Disorders of personality-DSM-IV and beyond (2nd ed.). New York, NY: John Wiley.

Munley, P. H., Vacha-Haase, T., Busby, R. M., & Paul, B. D. (1998). The MCMI-II and race. Journal of Personality Assessment, 70, 183-189.

Morita, S. (1976). Morita therapy: Zen Buddhism applied to psychotherapy. In W. S. Sahakian (Ed.), Psychotherapy and counseling: Techniques in intervention (2nd ed.; pp. 303-323). Chicago: Rand McNally College Publishing.

Okazaki, S., & Sue, S. (1995). Methodological issues in assessment research with ethnic minorities. Psychological Assessment, 7, 367-375.

Pedersen, P. (1988). Handbook for developing multicultural awareness. Alexandria, VA: AACD Press.

Pitrowski, C. (1997). Use of the Millon Clinical Multiaxial Inventory in clinical practice. Perceptual and Motor Skills, 84, 1185-1186.

Pitrowski, C., & Keller, J. W. (1989). Psychological testing in outpatient mental health facilities: A national study. Professional Psychology: Research and Practice, 20, 423-425.

Pitrowski, C., & Lubin, B. (1990). Assessment practices of health psychologists: Survey of APA Division 38 clinicians. Professional Psychology: Research and Practice, 21, 99-106.

Prediger, D. (Ed.). (1993). Multicultural assessment standards: A compilation. Alexandria, VA: Association for Assessment in Counseling.

Schinka, J. A., & Greene, R. J. (1997). Introduction to the special series-personality assessment instruments: Current status and future directions. Journal of Personality Assessment, 68, 2-4.

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

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

Sue, D., & Sue, S. (1987). Cultural factors in the clinical assessment of Asian Americans. Journal of Consulting and Clinical Psychology, 55, 479-487.

Sue, D. W., & Sue, D. (1990). Counseling the culturally different (2nd ed.). New York: Wiley.

Triandis, H. C. (1994). Major cultural syndromes and emotion. In S. Kitayama & H. R. Markus (Eds.), Emotion and culture: Empirical studies of mutual influences (pp. 285-306). Washington, DC: American Psychological Association.

Walters, G. D., Greene, R. L., Jeffrey, T. B., Kruzich, D. J., & Haskin, J. J. (1983). Racial variations on the MacAndrew Alcoholism scale of the MMPI. Journal of Consulting and Clinical Psychology, 51, 947-948.

Ae-Jung Chang Gunsalus is a counseling psychology doctoral candidate, and Kevin R. Kelly, Ph.D., is a professor. Both are in the Department of Educational Studies at Purdue University, West Lafayette, IN. Correspondence should be addressed to Kevin R. Kelly, Purdue University, Department of Educational Studies, 1446 LAEB, West Lafayette, IN 47907-1446. Email [e-mail]