MMPI T scores: Linear versus normalized.

Discusses the recommendations of R. C. Colligan et al (see record 1980-11023-001), who advised that raw scores on MMPI scales be transformed into normalized T scores (rather than the more traditional linear T scores) for clinical interpretation. The present author questions the advisability of this recommendation, considering the incomparability of normalized scores from distributions with different extremes and the possibility of statistically skewed normalized T distributions. It is concluded that the benefits of the recommended switch would not outweigh its costs. (11 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Normalized transformations and the interpretation of MMPI T scores: A reply to Hsu.

Replies to L. M. Hsu's (see record 1985-02859-001) criticisms of the present authors' (see record 1980-11023-001) recommendation to use normalized T scores for MMPI scales. The present reply addresses Hsu's objections from a theoretical standpoint and in the context of MMPI responses from a new reference sample of 1,408 normal adults. These new data support the present authors' previous recommendation and represent a step in stimulating new research and revitalizing the MMPI for clinical relevance. (6 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Uniform versus linear T scores on the MMPI2/MMPI in an outpatient psychiatric sample: Differential contributions.

Hypothesized that the uniform T-score procedure was one possible source of observed differences between the MMPI-2 T scores and the MMPI T scores. MMPI-2 linear T scores were computed for a sample of 200 psychiatric outpatients (mean age 38.1 yrs) whose MMPI-2 and MMPI T scores had been determined at the same point in time. Differences created by the uniform technique were slight and did not exceed 1.2 T-score points. Contrary to hypothesis, the uniform T-score procedure reduced the differences between mean MMPI-2 T-scores and the MMPI T scores on 12 of the 16 scales. These data suggest that the differences between the MMPI and MMPI-2 T scores are mainly due to differences between the normative samples. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The new uniform T scores for the MMPI—2: Rationale, derivation, and appraisal.

Although normative distributions of Minnesota Multiphasic Personality Inventory—2 (MMPI—2) clinical and content scales tend to be positively skewed, individual scales differ in degree of skewness. The linear T scores in use with the original MMPI preserve these skewness differences. For the MMPI—2, a Uniform T (UT score) transformation has been adopted. Its target distribution is the positively skewed composite (or prototypical) linear T score distribution of MMPI—2 clinical scales. UT scores are percentile comparable, yet, unlike normalized T scores, depart minimally from the familiar linear T scores. The authors describe the UT score derivation, report degree of percentile comparability achieved, and argue that the positive skewness of the UT-score distributions, besides minimizing discontinuity, is conceptually meaningful. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Chronological age differences in MMPI scores of male chronic alcoholics seeking treatment.

M. Gynther's (1979) general hypothesis that older Ss self-report better adjustment on the MMPI than do younger Ss was tested among 616 male chronic alcoholics classified into 5 age groups: 30 or younger, 31–40, 41–50, 51–60, and 61 or older. Multivariate and univariate analyses of variance and covariance supported Gynther's general hypothesis as well as 3 specific corollary hypotheses predicting that older alcoholics, relative to younger alcoholics, are lower in impulsivity-control difficulties and higher in health concerns, but are not different in social introversion. (8 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Driver behavior and scores on the MMPI.

Are auto accidents related to driver personality? Using a paper and pencil personality inventory (MMPI), the driver behavior and MMPI scores of 993 college students were compared. Some slight relationship was found. "Knowledge of the kind of personality organization and motivation of a driver may be useful for purposes of both licensing and training drivers." (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Prediction of mean scores on MMPI scales.

The accuracy of an equation for predicting raw score means on personality scales, based upon social desirability consideration, was tested using 57 MMPI scales. It was found that 25% of the predicted means were in error by less than 1 point, 51% by 2 points or less, and 70% by 3 points or less. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Age, intelligence, and MMPI F scores.

This study was designed to determine the effects of age, intelligence, and other variables on the F scale, which is critical for Minnesota Multiphasic Personality Inventory (MMPI) interpretation and research. Data consisted of MMPIs and Wechslers obtained from 407 white hospitalized psychiatric patients over a 7-year period. No significant sex differences on MMPI F scores were found. The major result was that F scores decrease with increasing age for low- and high-IQ subjects, but remain relatively constant for average-IQ subjects. Neither diagnosis nor educational level was found to affect F scores. The use of unrestricted distributions of age, intelligence, and F scores may account for the fact that these findings are not consistent with those reported by previous investigators. (21 ref.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

MMPI scores for field-dependent and field-independent psychiatric outpatients.

Investigated the relationship of field dependency to personality traits and styles of psychopathological adjustment. 3 shortcomings of previous research were avoided by using a larger sample (N = 414), a nonprojective measure of personality, and a clinical population. Validity scales, clinical scales, and 13 experimental Minnesota Multiphasic Personality Inventory (MMPI) scales were compared for field-dependent and field-independent psychiatric outpatients. Separate analyses were made for males and females. As predicted, field-dependent males scored higher on social desirability and acquiescence but lower on social introversion. Field-dependent females scored higher on hysteria, repression, and hypochondriasis but lower on ego strength and control. Overall results partially support the proposed relationship of field dependency to personality traits and modes of psychopathological adjustment. The sex of the S is an important variable in predicting specific relationships. Correlational data indicate that the strength of the relationship between field dependency and MMPI scales should not be overestimated. (16 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Education and MMPI Mf scores in homosexual and heterosexual males.

Investigated the relationship between MMPI Mf, Hsx, and education in 64 heterosexual and 49 homosexual men. Ss were obtained from a variety of sources, were generally older than college students, and averaged between 2 and 3 yr. of college education. The groups differed significantly on Mf, Hsx, and Kinsey self-ratings. MMPI scale differences were found on F, K, D, Pd, Pa, Pt, Sc, and Si. Correlations of .20 and .27 were found between education and the Mf and Hsx scores. Data indicate that Mf was a better discriminator between the homosexual and heterosexual groups than Hsx, and was less contaminated by educational bias. It is concluded that the relationship between educational level and Mf appears to be more complex than generally assumed. (23 ref.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Differences in MMPI scores of Black and White compulsive heroin users.

Investigated the question of racial bias in the MMPI by comparing responses of 252 Black and 120 White compulsive heroin users on both (a) empirically derived validity and clinical scales and (b) intuitively constructed content scales. Ss were male veterans admitted to a drug treatment program at a VA Hospital. In both univariate and multivariate ANOVAs and covariance of individual scales and overall profiles, Black users differed significantly from White users; the direction of differences was opposite to that reported for Blacks in previous studies of normal and psychiatric samples. Blacks scored significantly lower on 6 clinical scales (i.e., Infrequency, Depression, Psychopathic Deviate, Psychasthenia, Schizophrenia, and Social Introversion) when profile validity and the covariates age, intelligence, and socioeconomic status were both controlled and uncontrolled. On content scales, Whites displayed greater personality disturbance in Social Maladjustment and Family Problems, whereas Blacks obtained higher scores on Feminine Interests and Phobias. Results show that methodologically the question of a Black bias in the MMPI awaits additional information about the interaction of ethnicity with clinical samples; substantively, the results refute the assumption of personality trait communality among compulsive heroin users and suggest that ethnicity is an influential S background characteristic by which subgroups of heroin users might be identified. (28 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Comparison of MMPI scores of White and Japanese-American medical patients.

Compared the MMPI scores of 164 patients (mean age 40.3 yrs) classified by sex, diagnostic group (somatization disorder or organic brain syndrome), and race (White or Japanese-American). MANOVA revealed effects for sex and diagnostic group, as well as a sex by race interaction. Male Ss scored significantly higher than females on 9 clinical scales; thus, factors other than race (e.g., sex) appear to be critical in the interpretation of MMPI scores. Implications for the use of the MMPI with non-White patients are discussed. (9 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Dimensions of control: Correlations between MMPI and I-E scores.

Administered Rotter's Internal versus External Locus of Control (I-E) scale and the MMPI to 25 volunteer members of a suburban rescue squad. I-E scores correlated with 3 MMPI scales: negatively with the K and Hy scales and positively with the F scale. A comparison of these relationships with those found in a sample of alcoholics is made. It is suggested that the correlations observed indicate "that a sense of control over external events is related to self-control and competence in handling internal events." (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Normative data for the F(p) Scale of the MMPI–2: Implications for clinical and forensic assessment of malingering.

Data from several clinical samples and the Minnesota Multiphasic Personality Inventory–2 standardization group are presented to familiarize the reader with response patterns of different groups on a new validity scale designed to assist in the identification of exaggeration or fabrication of psychological disturbance. Sensitivity-specificity analyses are included along with suggestions for use of the F(p) Scale with other validity scales. Cautions about setting single cutoff scores are also discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A comparison of MMPI scores of Anglo- and Mexican-American psychiatric patients.

Reports MMPI profiles of 40 bilingual Mexican-American and 109 Anglo-American psychiatric patients. There were controls for the major variables of age, sex, ethnicity, socioeconomic status, and presenting problem. With K -corrected T scores, Mexican-American Ss scored higher on the L and K scales and significantly lower on the Pd, Pt, and Ma scales. Anglo-American males scored higher than Mexican-American males on the Mf scale. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Usefulness of the K correction in MMPI profiles of patients and nonpatients.

This study tested whether the MMPI K correction would reduce false negatives among patients and increase false positives among nonpatients. Ss were 214 psychiatric inpatients and 289 nonpatients. As L. M. Hsu (see record 1986-28964-001) predicted, nonpatients' profiles violated statistical normality more often than did patients'. However, kappa agreement rates between corrected and uncorrected profiles were .76 for patients, .86 for nonpatients, and .89 overall, showing that the K correction had little effect on the frequencies of primed profiles. Because profile shape was often changed by the K correction, it was concluded that its routine use may not always be helpful. An unexpected finding, in support of research by L. N. Robins et al (see PA, Vol 73:1363) on lifetime prevalence of psychiatric disorders, was that 30% of nonpatients' uncorrected profiles were primed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

MMPI scores of counselors on alcoholism prior to and after training.

Compared the MMPI scores of male alcoholism counselors before and after a 9-mo counselor training program. All Ss were recovered alcoholics or drug addicts. Few significant standard score differences were found, although raw score data reveal considerable shifting in individual scores. Additional data indicate that MMPI Pd and Ma scores measure rather enduring traits of addictive personalities. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Moderators of racial differences on the MMPI.

In a study with 104 Black and 191 White male prison inmates, multiple discriminant analysis of MMPI scores between high-IQ White, high-IQ Black, low-IQ White, and low-IQ Black Ss yielded 2 significant canonical variates. Results suggest that racial differences on the MMPI do not occur in all racial comparisons but instead are restricted to low-IQ groups. (8 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Prediction of mean scores on MMPI scales: A reply.

The practical utility of Edwards' prediction equation was re-examined as well as other psychometric issues raised by him. It was concluded again that the SD scale is not an effective substitute for the MMPI. (PsycINFO Database Record (c) 2010 APA, all rights reserved)