MMPI-2 (Minnesota Multiphase Personality Inventory – 2)

(Tsepho Mosupye)

 

Developed by Hathaway and McKinley (1989) the MMPI-2 is a broadband test designed to assess several of the major patterns of personality and emotional disorders.It can be administered to an individual or a group.Eighth grade school level of reading and comprehension is required. It was first developed in hospital setting on patients and non-patients in the 1930s.The items are written in the form of a statement of some personal experience, beliefs, attitude, or concern.There are 504 items in the set, each worded idiomatically to avoid the impression of formal examination or integration.Other items have been added to cover gender-role characteristics and a defensive style of self-representation thus increasing the item total to 550. Cross-validation studies have been done separating patients from non-patients (e.g. hypochondriac, neurotic patients, psychosis, manic depression, paranoia, and schizophrenia).There are also validity indicators: L scale (for defensive role-play); I scale (measure of extreme deviant or random responding); R scale (answered with subtle but pervasive tendency either to cover up (high score) or to exaggerate (low score) the problem). Data from the non-patients group (men and women) were used to convert the raw score into linear T scores for the validity of the clinical scales.These test norms have been used in the standard MMPI profiles throughout the US and in many countries. Alternative sets of norms were developed for adolescents, students, and elderly subjects.

Restandardisation of the Inventory

Concerns about some items that deal with sexual adjustment, bodily functions and religion led to the re-standardisation with contemporary national norms.82 items were modified and 154 provisional items added to make the scale to consist of 704 items.The items were revised with psychometrically sound items/wording.Additional were forms created to gather biographical and supplementary (changes in lives) information about the sample of adults used for new test norms.The sample was drawn from communities in seven US states; 1138 (males) and 1462 (females).It constituted of Asians, Blacks, Hispanic, Native Americans, Whites and others.Occupational income and educational level reflected the skew towards upper socio-economic level in the normative group.

Testability of the Subjects

Ability of the subjects to understand test instructions and to comply with the requirements and to comprehend and interpret items in reliable way is the criteria for administering the MMPI-2.The item must be meaningful to the subject given his/her life experience (i.e. culture).If not, it is suggested that more manageable forms of the inventory are used, or the discontinuance of the session.

Test formats

üBooklets:for groups with 567 items (basic scale = 370 items) and separate answer sheets.

üCassettes: for individual and answer sheet (for 1 & half hour)

üComputer-based administration (for 1 & half hour)

 

MMPI-2 (Minnesota Multiphase Personality Inventory – 2):

 

A Discussion

Introduction

The MMPI is undoubtedly the most widely used and researched of all inventories (Kline, 1993).There are more than 10, 000 published articles, books and chapters about this test.As mentioned above, there has been a substantial revision of this test – the MMPI-2.The following discussion is going to focus on the critical evaluation of the test.The discussion will also look at some of the published articles on the MMPI and finally a brief discussion about the applicability of the instrument in South Africa will be given.The latter discussion will include a discussion with Dr Chris Nunns of the SA Breweries.

A critical psychometric Evaluation of the MMPI 

The scales of this test were constructed by the criterion keyed-method, i.e. items were selected if they could discriminate between different clinical groups (Kline, 1993).This itself presents a psychometric problem.According to Kline (1993) the criterion-keyed method of test construction leads to scales of little psychological meaning since the criterion grouping may differ on a number of variables.The method also allows the same items to appear in different scales.When this occurs, correlational and factor analyses are difficult to interpret.The other psychometric problem with the MMPI is the low reliability of scales (Kline, 1993).This phenomenon is not surprising given the criterion-keyed test of the MMPI.This thus highlights the lack of homogeneity and meaning of the scales.

Another important difficulty with the MMPI is the fact that it was developed as a clinical instrument for use with abnormal subjects (Kline, 1993).While the test discriminates among abnormal groups, mainly because many of the items refer to neurotic symptoms, when it is administered to normal subjects there is far less variance and scores on the scales tend to be low.Most normal individuals have fewer symptoms than those receiving psychiatric treatment.The lack of variance affects the clarity of the emerging factors.This makes MMPI research difficult and the results of such research have to be treated with caution (Kline, 1993).

Finally, the factors derived from the MMPI items are of two kinds, normal and abnormal.Several studies have indicated that the normal factors are similar to the big five (e.g. Costa, & McCrae, 1986; Johnson, 1984), the abnormal factors, on the other hand, require external validation (Kline, 1993).

The new version of the MMPI, MMPI-2 did not address the above limitation.The only things improved were the samples from which the original scales were derived and the modification of items, some of which were obsolete in terms of content.Thus in summary; the deficiency in relation to the lack of clear factor structure; item overlap; and the lack of external validation of abnormal factors, brings a psychometric limitation to the Inventory.

Discussion of some of the studies conducted on the Instrument

Literature research on the MMPI in South Africa did not yield results.It appears that, although used in certain settings (including industry), there seems to be a lack of published research done on the test in South Africa.Conversely, many studies have been conducted internationally.We can extrapolate from these studies about the applicability of the instrument in the South African context.

The table below consists of a number of studies, which were identified as relevant pertaining to the question of the validity and reliability of the instrument. The table consists of brief abstracts and references of these studies.These studies address issues such as racial/ethnic and demographic characteristics issues, translation of the MMPI to other languages, construct and content validity issues, methodological issues, and so on.The issues addressed in these articles are very relevant especially when considering the applicability of the instrument to the South African context.They can also serve as a basis for discussion or debate about the applicability of the MMPI in South Africa.The reference information provided directs the reader to the articles and their locations.

Issues to consider when testing in South African context

Several issues need consideration when using psychometric tests in a South African context.In their study about psychometric testing in South Africa, Sehlapelo and TerreBlanche (1996) interviewed a range of stakeholders including, human resource executives, personnel managers, political spokespersons, trade unionists, workers and academics.They identified several issues pertaining to psychological testing in South Africa. These are discussed below.

One of the pertinent arguments around the use of personality tests such as the MMPI in industry is the fact that many people perceive them as irrelevant to the actual skills required in jobs.Workers do not see the sense in being judged on their performance in pencil and paper tests that have no bearing on day-to-day activities.Following from this argument is job experience, which is considered more important than testing.Tests are perceived as giving undue emphasis on formal qualifications rather than proven competence.Alternatively, interviewing is considered as having more legitimacy than tests.Perhaps because interviews provide the overt personal interaction and preliminary feedback, that tests fail to provide (Sehlapelo & TerreBlanche, 1996).

Psychological tests such as personality inventories are also accused of having dehumanising and traumatic effects.Not only this, such tests particularly in South Africa are distrusted as promoting inequality between the races.They are seen as a smoke screen designed to keep blacks out (Sehlapelo & TerreBlanche, 1996).One of the contributory factors to the absence of trust is the lack of adequate feedback and the fact that someone possesses personal information and this information is kept away from the testees themselves.

Discussion with Dr Chris Nunns of the SA Breweries

Although a lot of these issues are relevant to testing in South Africa in general, some of these issues were addressed that relate to the MMPI usage in South Africa.An interview with one of the proponents of the MMPI-2, Dr Chris Nunns looked at some of these issues.According to Dr Chris Nunns, the MMPI-2 is a useful tool in industry especially looking at very large organisations such as the South Africa Breweries.He however notes that the use of personality inventories such as the MMPI is suitable for positions that bear more responsibility to the organisation such as management level positions.Not only should we look at the person’s competence but also whether the person can handle things like stress-related and emotionally charging situations associated with such positions, Dr Nunns said.The MMPI is therefore relevant in industry because it provides additional information about the applicants’ psychological ability to handle particular situations.Furthermore, it is helpful to the applicant or testees themselves in that when provided with feedback, they can learn about themselves and perhaps some of their limitations.

The SAB approach to the MMPI is that it is only used on management level employees, including supervisors, sales representatives and so on.That is, those who can understand the instrument and its value.The MMPI is also a part of test batteries and is thus used in collaboration with other tests that are more related to the job.This approach, according to Dr Nunns, has produced predictive validity in terms of the association between salary and level of performance. SAB has employed a number of psychologists, both internal and externally based.The MMPI can also help in identifying some of the employees who may actually need therapeutic attention.Those employees will thus use the psychological services provided by the organisation.

Responding to the question of the validity and reliability of the instrument given the fact that it was developed from a different country and hence consists of different norms, Dr Nunns said that he believes that the instrument is universally applicable and not necessarily culturally biased.He however, thinks that the translation of the instrument may be sufficient do deal with some of the psychometric issues such as item bias effect.In fact, he mentioned that there has been an initiative to translate the MMPI into Xhosa. 

Conclusion

The study conducted by Sehlapelo and TerreBlanche indicated that some of the views and perceptions of psychometric testing in South Africa are not positive.These perceptions may not necessarily reflect the objective truth about the psychological test, nevertheless, they are shared among stakeholders, particularly workers.In order for these instruments to be used effectively, such perceptions should be challenged and eliminated.One of the ways of doing that is to address some of the issues mentioned above, such as restoring/building trust for the testing process, providing adequate and useful feedback, and generally demonstrating the link between the tests and the job.Knowledge about psychological tests should not be a privilege of a few practitioners, but testees themselves should also be equipped with the knowledge about the tests and their importance.

The MMPI-2 also falls within the above argument, more so because, unlike other tests that are more directly related to the industry, the MMPI-2 was developed for clinical objectives.It is for this reason that the instrument should be used with caution and in collaboration with other testing instruments.When used appropriately, the instrument can be useful in industry as the above discussion has revealed.Except for SAB, no other organisations were identified as using the MMPI-2.Following from this report it is hoped that a follow-up to establish who else uses this instrument is imperative.Below is list of studies that address some of the issues discussed above.The reference information will direct the reader to the full articles.

International Research on the MMPI-2


 

ABSTRACTS

REFERENCES

 TITLES

Ss were 75 African American and 725 White men and 65 African American and 743 White women who were part of the MMPI restandardization sample (J.N. Butcher, 1989).Mean differences on the MMPI-2 validity and clinical scale between ethnic groups were small for both genders; however, any differences could be because of variation in demographics.When Ss were matched by age, education and income, fewer MMPI-2 scale means differed.In a follow-up study, the accuracy of the MMPI-2 clinical scales in prediction partner ratings was calculated.The predictions, which were based on a regression equation, were not significantly different between the ethnic groups.
Timbrook, R.E., & Graham, J.R. (1994) Psychological Assessment, 6(3), 212-217.
Ethnic differences on the MMPI-2
Developed and tested a Spanish (Mexican) translated version of the MMPI-2.This version was developed using the concept of transliteration rather than literal translation of individual scale items.The instrument was evaluated by comparing profiles from 929 male and 1,245 female Mexican undergraduate student profiles to MMI-2 profiles from data from the US college normative sample consisting of 515 men and 797 women.Results indicate that the Spanish MMPI-2 was appropriate for use in the college population.Standardised profiles for Mexican males and female college students are presented.

Lucio, R., Reyes-Lagunes, L & Scott, R.L. (1994) Journal of Personality Assessment. 63 (1), 105-116.
MMPI-2 for Mexico: Translation and Adaptation.
Examined the differential criterion-related validity of the MMPI and MMPI-2 and their stability over 4 months among 137 undergraduates by correlating the clinical scales with their counterpart SCL-90-R Factors.Fair to moderate correlations were found on all 8 paired MMPI scales and SCL-90-R factors, while only 2 of 8 MMPI-2-SCL-90-R pairings correlated significantly.Further analyses, however found no significant differences between these MMPI-SCL-90-R and MMPI-2 D scale with the SCL-90-R depression factor. Several issues related to the equivalence between between the MMPI and the MMPI- are discussed. 

Rodjev, R., Nelson, W.M., Hart, K.J., & Fercho, M.C. (1994) Journal of Clinical Psychology, 50(3), 361-367.
Critrion-related validity and stability: Equivalence of the MMPI and the MMPI-2
Examined a sample of 289 MMPI-2 results from worker’s compensation and personal injury cases to determine the relationship of various indicators of exaggeration. Intercorrelations of the F, F-K the MMPI Dissimulation Scale-revised, total of obvious minus subtle scales, Fake Bad Scale, VRIN, and TRIN were computed and the relative sensitivity of each score calculated using various cut-offs. Factor analysis suggests that malingering may take the form of inconsistent responding as will as symptom exaggeration.Patients evaluated at the request of plaintiff attorneys showed a seemingly greater degree of symptoms exaggeration and inconsistent responding than did those referred by defence counsel.
Fox, D.D., Gerson, A. & Lees-Haley, P.R. (1995). Journal of Clinical Psychology 51(1) 42-47.
Interrelationship of MMPI-2 validity scales in personal injury claims.
Minnesota Multiphasic Personality Inventories (MMPIs) were administered to an urban mixed-race sample of adolescent psychiatric inpatients and outpatients (N=218).Racial differences were noted on 4 of the 13 raw Wiggins scale scores, with scores for African–American patients more elevated on Poor Health, Feminine Interests, phobias, and psychoticism.A backward multiple-regression was used to predict raw Wiggins scale scores from parent descriptions as measured by the Personality Inventory for Children clinical scales.Predictors supported the validity of the Organic Symptoms, Social Maladjustment, Family Problems, Authority Conflict, and Hypomania scales, and to some extent, Poor Morale, Depression, Poor Health, and Phobias, Predictors differed across racial groups for 9 of the scales, with particular discrepancies in predictive meaning for poor health, hypomania, family problems, and authority conflict.Implications for MMPI-A content scale are discussed.
Wrobel, N.W., Lachar D. (1995). Psychological Assessment. 7(2), 140-147.
Racial differences in adolescent self-report: A comparative validity study using homogeneous MMPI content measure.
The comparability and validity of a computerised adaptive (CA) MMPI-2 were assessed in a sample of 237 male and 334 female undergraduate college students (aged 17-44 yrs).The CA MMPI-2 administered adaptive Scale L, F, the 10 clinical scales, and the 15 content scales, utilising the countdown method.All Ss completed the MMPI-2 twice, with 3 experimental conditions: booklet test-retest, booklet-CA, and conventional computerised (CC)-CA.Profiles, across administration modalities show a high degree of similarity, providing evidence for comparability of the 3 forms.Correlations between MMPI-2 scales and other psychometric measures (Beck Depression Inventory, SCL-90-R; State Trait Anxiety and Anger scale, and the Anger Expression Scale) supported the validity of the CA MMPI-2.Substantial item saving may be realised with the implementation of the countdown procedure.
Roper, B.L., Ben-Porath, Y.S. & Butcher, J.N. (1995) Journal of personality Assessment. 65(2), 358-371
Comparability and Validity of Computerised adaptive testing with the MMPI-2.
Investigated the concurrent validity and internal consistency of the Feminine Gender Role (GF) and Masculine Gender Role (GM) scales of the MMPI-2.GF and GM scores were obtained from 173 female and 90 male college students (aged 18-57 yrs) and were correlated with scores from the Bem Sex-Role Inventory, the Sex Role Behaviour Scale, and the Sex Role Identity Scale.Results show that the GF and GM scales have low internal consistency and low concurrent validity, their patterns of correlation with personality measures suggest that GF and GM are more related to personality traits of interpersonal potency and sensitivity than to masculinity and femininity.
Johnson, M.E., Jones, E., & Brems, C. (1996) Journal of Personality Assessment. l66(1), 153-168.
Concurrent Validity of the MMPI-2 Feminine gender role (GF) and Masculine gender Role (GM) scales.
Personality questionnaires are among the most versatile and user-friendly approaches to personality assessment. The article focuses on the methodological consideration in conducting research on the MMPI-2, the most widely used clinically personality instrument.The article addresses ways of identifying methodological problems in research and alerts researchers to potential pitfalls in conducting personality assessment research.The topics addressed include the following: methodological factors addressing the continuity of the MMPI-2 and the original MMPI; sample selection in MMPI-2 research; issues concerning test administration; the application of exclusionary criteria in developing research samples; methodological factors in processing, reporting, and analysing data; developing and evaluating new MMPI-2 scales; and assessing test bias in personality research. 
Butcher, J.N., Graham, J.R. & Ben-Porath, Y.S. (1995) Psychology Assessment. 7(3), 320-329.
Methodological problems and issues in MMPI, MMPI-2, and MMPI-A research.
Investigated the degree to which racial identity and levels of acculturation influenced 51 undergraduate Mexican-Americans’ performance on the L, K, and Mf scales of the MMPI-2.Ss completed and acculturation scale, a racial identity attitude scale, and the MMPI-2.Performance of the L and K scales was influenced by racial identity attitudes, and performance on the L scale was influenced by level of acculturation.No evidence was found to suggest a relationship between cultural variables and performance on the Mf scale.
Canul, G.D., & Cross, H.J. (1994) Journal of Clinical Psychology. 50(5), 736-745.
The influence of Acculturation and racial identity attitudes on Mexican-Americans’ MMPI-2 performance.
Psychometrics played a central role in the development of psychology in South Africa and continues to affect more South Africans than any other branch of psychology.For this reason the transformation of psychology in South Africa will not be complete until psychometric testing practices have been transformed.Interviews with personnel practitioners, political spokespersons, trade unionists, workers and academics, revealed that certain convergences and divergences between the views from above and below.Workers tended to have little trust in tests and the testing process, and to see the solution in the formulation the explicit testing policies arrived at through a process of consultation.Personnel practitioners, while acknowledging cultural bias in testing, tended to think that it could be overcome through construction of non-bias tests.Views from above and below coincide with regard to the popularity of interviewing as an alternative to testing, the idea of learning potential assessment, and the high degree of respect for the technical expertise of psychometricians.This places a heavy burden of responsibility on experts in psychological testing to help redeem psychological testing and psychology, in the eyes of South Africans.
Sehlapelo, M. & Terre Blanche, M. (1996) Psychology in Society Vol 21, 49-59
Psychometric testing in South Africa: Views from above and below

Additional articles of the MMPI-2 were compiled from the Internet and they were selected based on their relevance to the validation issue.These are presented below.

 

International Developments with the MMPI-2


James N. Butcher
University of Minnesota


It took almost a decade for the original MMPI to begin its international odyssey, with the first translation in Italy in 1948. MMPI-2 began its cross-cultural adaptation immediately after its publication in 1989, and in just 3 years, there have been 15 translation projects of the MMPI-2 completed or in progress.

International interest in MMPI-2 and MMPI-A was highlighted at the recent MMPI-2/MMPI-A workshops held in Bruges, Belgium sponsored by the Free University of Brussels, Department of Psychology, University of Minnesota, The University of Antwerp, and the University of Nijmegan (the Netherlands). The workshop was presented as a satellite meeting of the International Congress of Psychology.Psychologists from 19 countries and all continents (except Antarctica) participated n the program. An interesting feature of the Bruges workshop included a session in which psychologists did clinical interpretation of the same MMPI-2 profiles from different countries. The exercise demonstrated that MMPI-2 interpretation possesses similar features and comparable conclusions regardless of national perspectives from which the interpretation is done.

Thegeneralisability of MMPI-2 constructs across national boundaries was extensively discussed in an international research forum on adaptation of the MMPI-2. Psychologists from several countries shared their experiences in translating and adapting the MMPI-2 in different languages. Psychologist translators who worked with the original MMPI are in agreement that the revised MMPI-2 items are easier to adapt than the original version of the MMPI. Another interesting finding, of those who have collected normative data, is that contemporary normal samples in other countries appear closer to the new MMPI-2 Norms than earlier international samples were to the original MMPI norms. The re-norming of the MMPI appears to have "brought normal samples from different countries closer together."

A survey of current translation projects on the MMPI-2 is informative. The broadening interest in adapting the MMPI-2 in other countries is reflected in the fact that in addition to those countries that have had established programs on the original MMPI, there are several new frontiers (i.e. Russia) where testing has been traditionally discouraged.
Five Spanish translations of the MMPI-2 have been completed or are in progress: Chile (Fernando Rissetti), Mexico (Ernilia Lucio GM and Isabel Reyes-Lagunes), Spain (Alejando Avila-Espade), Argentina (Maria Casullo), and Puerto Rico (an adaptation of the Chilean version by Cecilia Colon).

Several European translations are in process: Italy (Saolo Sirigatti &Paolo Pancheri), Russia (Vladimir Martens & Victor S. Koscheyev), Norway (Bjorn Ellertsen & Halgrim Klove), Iceland (Solvina Konraos), France (Mireille Simon), Belgium and Holland (Hedwig Sloore & Jan Dirksen). Two Turkish psychologists have initiated a translation of the MMPI-2 into Turkish (Isik Savacir and Merel Chula). Dr. Savacir has a great deal of experience with the original MMPI in Turkey since she worked on the original translation and conducted a major study (with Nese Erol) validating the MMPI there in 1989. Anna Kokkevi is translating the MMPI-2 and MMPI-A in Athens, Greece.

There are two translation programs of the MMPI-2 underway in the Middle East. Moshe Almagor, University of Haifa, is leading a team of Israeli psychologists in adapting the MMPI-2 and MMPI-A in Israel. Abdalla Soliman, an Egyptian psychologist at the United Arab Emirates University, is developing an Arabic version of the MMPI-2. Several Asian language translations have been completed and are the subject of ongoing research projects: China and Hong Kong (Fanny Cheung & Song Weizhen), Japan (Noriko Shiota), Korea (Kyunghee Han), Thailand (La-Or Pong Panich) and Vietnamese (Pauline Tran). 

Are Unisex Norms for the MMPI-2 Needed? Would They Work?


Auke Tellegen, James Butcher & Tawni Hoeglund

University of Minnesota

Traditionally, MMPI scales have been normed separately for men and women, in large part because McKinley and Hathaway (1940) found a somewhat different distribution of scores for men and women on some scales. They observed:

Likewise, the scores for females are, without exception, higher than for the corresponding males. One might indulge in considerable speculation on these findings, but since the validity of the test within the normal group is at present in process of study, and the differences themselves are slight, we are not prepared to draw any conclusions as to the meaning of such differences.(p. 266).

However, some scales, notably Si, were based on a single norm for men and women. Observed gender differences in personality scale responding even prompted some authors to recommend different items (in separate booklets) for men and women in order to refine interpretation on certain personality scales (Block, 1965). For a current evaluation of gender differences, see recent discussions by Deaux (1985) and Deaux and Major (1987). Recent civil rights legislation (Adler, 1993) has called into question the use of separate norms for men and women, especially when employers use the tests in employment selection situations. The implications of changing the basic normative approach on which more than fifty years of research has accumulated could be great and need to be examined. Consequently, Tellegen, Butcher, and Hoeglund initiated a first study to evaluate the potential impact of employing "unisex norms for the MMPI-2 clinical scales. 

Results
First, Tellegen, Butcher, and Hoeglund evaluated possible differences between men and women in their responses to the items on MMPI-2 by examining the endorsement percentages for each item by men and women. They found that most MMPI-2 items are endorsed by similar percentages of men and women. Furthermore, most items that did show substantial differences belonged to the Mf scale, not to items reflecting psychopathology.

Next, to evaluate the impact of different norming procedures on the MMPI-2scales, Tellegen et al. (1993) developed a set of "unisex norms "for the MMPI-2 clinical scales. Using a combined sample of 1,138 men and to avoid confusion 1,139 women from the MMPI-2 restandardization sample, they computed uniform T scores following the same procedures employed in the development of the MMPI-2 norms (Butcher, Dahlstrom, Graham, Tellegen,& Kaemmer,1989). They found that the frequency distributions for theMMPI-2 clinical scales for the combined (unisex) sample closely resembled the frequency distributions of the separate gender norms. As a result, the unisex norms appeared to operate in a manner similar to the correct gender-specificMMPI-2 norms. Relatively small T score changes occur for either men or women when unisex norms were employed instead of gender-specific norms. Their scores for the cases have been plotted on both the MMPI-2 gender-specific and unisex norms. It is apparent that interpretations of the profiles would be the same, regardless of which set of norms were used. Such differences as do occur between the MMPI-2 clinical scale distributions tend to be small and differ somewhat at different ages. That is, for some age groups the differences are greater than for others.

Conclusions
Item response differences between men and women were relatively small for most items. The Mf scale accounted for most of the items that showed substantial differences between men and women. Applying the MMPI-2 normative data to the clinical scales, unisex norms appear to operate in a manner similar to that of traditional gender-specific norms. Unisex norms do not appear to "disadvantage" either gender.On the other hand, the traditional normative approach initiated by Hathaway and McKinley, which is represented by MMPI-2 T scores (based on gender specific distributions), likewise does not appear to "disadvantage" either gender. Similar distributions and T scores are obtained irrespective of which normative procedure is followed.

Relatively small differences were found on a few scales in some elevation ranges. The differences appeared to be somewhat influenced by age. That is, at some age levels the normative gender differences are larger than at others, but no extreme differences were found.

This study did not examine the relative validities of one set of norms overran other. An important next step would be for interested researchers to evaluate the extent to which either unisex norms or traditional gender specific norms actually perform better in the prediction of behavior.

References

Adler, T. (1993). Separate gender norms on tests raise questions. Monitor,24, 6.

Block, J. (1965). Challenge of response sets. New York: Appleton-Century-Crofts.

Butcher, J. N., Dahlstrom, W.G., Graham, J. R., Tellegen, A., & Kaemmer, B. (1989). Minnesota Multiphasic Personality Inventory-2 (MMPI-2): Manual for administration and scoring. Minneapolis: University of Minnesota Press.

Deaux, K. (1985). Sex and gender. Annual Review of Psychology, 36, 46-92.

Deaux, K., & Major, B. (1987). Putting gender into context: An interactive model of gender related behavior. Psychological Review, 94, 369-389.

McKinley, J. C., & Hathaway, S. R. (1940). A multiphasic schedule (Minnesota):II. A differential study of hypochondriases. Journal of Psychology, 10,255-268.

Tellegen, A., Butcher, J. N., Hoeglund, T. (1993, March). Are unisex norms for the MMPI-2 needed? Would they work? Paper given at the 28th Annual Symposium on Recent Developments in the Use of the MMPI/MMPI-2/MMPI-A. St. Petersburg, Florida.

International MMPI-2 Update

James N. Butcher
University of Minnesota


In October 1991, two international MMPI-'/MMPI-A programs were presented in Asia by James Butcher, Fanny Cheung, Song Wei Zhen and Carolyn Williams.The first program, a two day workshop co-organized by James Butcher and Fanny Cheung, was held in Hong Kong at the invitation of the Chinese University of Hong Kong to address the soon-to-be completed Chinese version of theMMPI-2. The second program, a three-day training session on the MMPI-2/MMPI-A, was held in Guanzhou, China and was co-sponsored by the University of Minnesota, the Zongshan (Sun Yat Sen) University, and the Chinese Academy of Science in Beijing. The organizing committee for this program was James Butcher, Fanny Cheung, Song Wei Zhen, and Peiguan Wu. Sixty psychologists from throughout China attended the workshop in the People’s Republic of China.Psychologists traveled great distances to attend the program--the travel record for getting to MMPI meetings was broken by two psychologists from Harrbing and JiLing who were en route for seven days by train to reach the conference in Guanzhou!

There is a great deal of interest in the MMPI-2 in Asia with MMPI-2 translations in Chinese, Japanese, Korean, and Thai. Currently there are several research programs devoted to development and standardization in these countries.Probably the most extensive international research on the MMPI-2 has been in China. The extensive research and clinical application of the Chinese MMPI is due, in large part, to the efforts of Fanny Cheung from Hong Kong, who translated the original MMPI into Chinese, and Song Wei Zhen from Beijing, who adapted and standardized the original instrument in the People's Republic of China in the early 1980s. Cheung and Song have stimulated a great deal of research on the MMPI in China as evidenced by the numerous publications on its use. The past decade of research has been recently summarized in a article: Cheung & Song (1989) entitled "A review on the clinical applications of the Chinese MMPI" in Psychological Assessment, 1, 230-237.

The MMPI-2/MMPI-A workshop in Guanzhou provided an opportunity for the Chinese psychologists to meet and establish a national research network to collect normative data on the MMPI-2 with representation from all the provinces in China. Continuity and Changes in MMPI-2 Validity Indicators: Points of Clarification

Note: How readable is the MMPI-2 anyway?


James N. Butcher
University of Minnesota


An interesting problem has emerged since the publication of the MMPI-2 manual concerning the reading level required for the MMPI-2. In order to determine the reading level of the MMPI-2 items, the MMPI Committee checked the reading level of the revised booklet by computing lexile values according to the Metametrics Corporation's the lexile framework. This program indicated that items were at the 8th grade reading level. Surprisingly, this was two years higher than what we have traditionally assumed for the original MMPI. What happened? Has our educational system slipped so far in teaching people how to read? It might appear so but there is another explanation. It is likely that we were victimized by technology and assumed that everything coming out of the computer is correct!


When the overall reading difficulty of the MMPI-2 items was evaluated using other reading level indices such as Grammatik Mac, which provides the Flesch-Kincaid Grade Level Index, the MMPI-2's readability was found to be at the 5th or6th grade level. Thus, it appears likely that neither the reading level of MMPI items nor the educational system's preparation of Americans to read has drastically changed as far as understanding MMPI items go.