MMPI adolescent norms: patterns and trends across 4 decades

This article examines central issues involving the accuracy of Minnesota Multiphasic Personality Inventory (MMPI) adolescent norms through examination of findings from eight investigations (n = 17,286), reporting MMPI mean values for samples of normal adolescents conducted between 1949 and 1964, and four studies (n = 1,758), based on adolescent samples collected since 1975. Issues related to sex and racial differences in adolescent samples are also examined. In addition, MMPI data from 1,315 normal adolescents, collected at the Mayo Foundation, and from 217 normal adolescents, collected in Norfolk, Virginia, are evaluated in relation to adult normative values on the Harris-Lingoes (Harris & Lingoes, 1955) content subscales to identify unique characteristics of adolescents' response patterns. Results support the adequacy of the adolescent norms, developed by Marks, Seeman, and Haller (1974), for evaluation of samples of normal adolescents collected between 1947 and 1965. MMPI patterns, produced by contemporary samples of adolescents collected since 1975, are elevated above the Marks et al. norms on most clinical scales. This latter finding indicated the need for the development of new adolescent norms that more precisely reflect current adolescent response frequencies and patterns. Finally, results from Harris-Lingoes subscales are reviewed to highlight unique aspects of adolescent developmental experience and to underscore the necessity of development of age appropriate adolescent norms for MMPI special scales.     

MMPI response-set characteristics among adolescents

This study examined the Minnesota Multiphasic Personality Inventory (MPPI) profile characteristics for five different response sets in adolescent populations employing adolescent MMPI norms. In general, findings for "all-true" and "all-false" adolescent profiles were highly similar in shape and elevation to those reported in the adult literature (Graham, 1977; Lachar, 1974), whereas random profiles on adolescent norms differed substantially from adult-norm random profiles in shape but not elevation. The characteristics of "fake good" or "fake bad" response sets were investigated by the use of adolescent subjects in two settings. Ninety-four public high-school students were administered the MMPI with standard and "fake bad" instructions, and 24 adolescents in inpatient psychiatric treatment were administered the MMPI under standard and "fake good" instructions. The MMPI profiles generated by special instructional sets were analyzed in relation to the subject's age, sex, race, and actual MMPI profile features. Findings indicated that although sex and race effects were evident in normal adolescents' attempts to simulate psychopathology on the MMPI, teenagers generally produce profiles containing grossly exaggerated symptom patterns that are relatively easy to detect as invalid. In contrast, it was found that a substantial number of psychiatrically disturbed adolescents may effectively simulate normal profiles and that effectiveness in these attempts was related to greater age and lower actual MMPI T-score values on the Hs and Hy scales.     

Clinical Correlates of MMPI Depression Scales

We investigated whether Scale 2 (Depression [D]) and the Wiggins Content Scale of Depression (DEP) of the Minnesota Multiphasic Personality Inventory (MMPI) have different clinical correlates when only one of these two scales is elevated. According, a group of patients who elevated DEP higher than Scale 2 (DEP > 2) were compared with a group of patients who elevated Scale 2 higher than DEP (2 > DEP). The patients with DEP > 2 were rated as being less severe than the patients with 2 > DEP on the following Brief Psychiatric Rating Scales: Somatic Concern (SOM), Emotional Withdrawal (WDRA), Depressive Mood (DEP), and Blunted Affect (AFF). The patients with DEP > 2 were rated as more severe on Excitement (EXC). The patients with DEP > 2 were more likely to receive the Axis I diagnoses of: bipolar disorder, manic, and alcohol abuse. Schizophrenia was equally probable for patients in the two groups. It appears that these two MMPI scales of depression have different clinical correlates when either one scale or the other is elevated.     

You Can Get Here From There

We introduce 3 studies of the empirical correlates of the MMPI–2 Restructured Clinical (RC) Scales: a study by Arbisi, Sellbom, and Ben-Porath (2008/this issue) that expands the range of correlates of the RC scale in mental health settings; an investigation by Sellbom, Ben-Porath, Baum, Erez, and Gregory (2008/this issue) of the empirical correlates of the scales in a forensic setting; and a study by Forbey and Ben-Porath (2008/this issue) examining convergent and discriminant validities of the RC scale in a nonclinical setting. To serve as a general introduction to the 3 new empirical studies presented in this special section, we begin with a review of the existing literature on the RC scales and briefly describe these studies.     

Subtypes of substance abusers: personality differences associated with MacAndrew scores

Several earlier studies have reported clinically relevant personality correlates of high vs low scores on the MacAndrew Scale (Mac) of the MMPI. Unfortunately, these projects have not adjusted for age or nature of abuse. Also, most have assumed that the personality correlates are the same for female patients as for male patients. This study attempts to address these deficiencies. Even after correcting for age and diagnosis, high Mac patients differ from low Mac patients on major scales of the MMPI. The pattern of such differences varies considerably between men and women. Substance abuse treatment implications of these differences are discussed.     

MMPI scale development methodology

A multitude of special MMPI scales have been developed without sufficient consideration for the methodology of empirical scale construction. Suggestions are presented in this article regarding the selection of items for a new MMPI scale and the tasks needed for the evaluation of a new scale. An investigator who develops a new MMPI scale should consider (a) the need for K corrections of scale scores, and (b) the possibility that the information provided by the special scale can also be obtained from the standard MMPI scales.     

MMPI Adolescent Norms: Patterns and Trends Across 4 Decades

This article examines central issues involving the accuracy of Minnesota Multiphasic Personality Inventory (MMPI) adolescent norms through examination of findings from eight investigations (n = 17,286), reporting MMPI mean values for samples of normal adolescents conducted between 1949 and 1964, and four studies (n = 1,758), based on adolescent samples collected since 1975. issues related to sex and racial differences in adolescent samples are also examined. In addition, MMPI data from 1,315 normal adolescents, collected at the Mayo Foundation, and from 217 normal adolescents, collected in Norfolk, Virginia, are evaluated in relation to adult normative values on the Harris-Lingoes (Harris & Lingoes, 1955) content subscales to identify unique characteristics of adolescents' response patterns. Results support the adequacy of the adolescent norms, developed by Marks, Seeman, and Hailer (1974), for evaluation of samples of normal adolescents collected between 1947 and 1965. MMPI patterns, produced by contemporary samples of adolescents collected since 1975, are elevated above the Marks et al. norms on most clinical scales. This latter finding indicated the need for the development of new adolescent norms that more precisely reflect current adolescent response frequencies and patterns. Finally, results from Harris-Lingoes subscales are reviewed to highlight unique aspects of adolescent developmental experience and to underscore the necessity of development of age appropriate adolescent norms for MMPI special scales.     

The relationship of MMPI and Sensation Seeking Scales to adolescent drug use

This study examined the relationship of Minnesota Multiphasic Personality Inventory (MMPI) measures, including the MacAndrew alcoholism (MAC) scale, and the Sensation Seeking Scales (SSS) to adolescents' drug use across nine drug categories. Subjects were 51 male and 72 female high school students between the ages of 14 and 18 (mean age = 16 years, 5 months). The drug use/abuse measure consisted of adolescents' self-reports on the Segal (1973) Alcohol-Drug Use Research Survey. Drug categories included for investigation were alcohol, amphetamines, barbiturates, caffeine, cocaine, hallucinogens, marijuana, narcotics, and tobacco. Scores from standard MMPI scales, the MAC scale, and the SSS were examined in relation to individual drug use outcomes, and multivariate procedures were used to predict polydrug versus single drug use patterns. Results demonstrated significant and meaningful relationships between personality measures and drug use among adolescents, with consistently strong findings for the SSS.     

Characteristics of the frequent nightmare sufferer.

Previous research has found that persons who experience frequent nightmares score highly on scales that measure psychotic symptomatology. Neurotic symptoms have also been implicated as correlates of nightmare frequency. In this study, 30 adult lifelong nightmare sufferers were compared with 30 control subjects, matched for age, sex, and socioeconomic status. Subjects were asked to record all dreams for 1 month and to complete the Minnesota Multiphasic Personality Inventory (MMPI) and the Eysenck Personality Questionnaire (EPQ). Nightmare subjects scored significantly higher on the EPQ Neuroticism scale and on 8 MMPI clinical scales than did the control group. These scales also best discriminated between the groups in a direct discriminant analysis. The results are interpreted as a reflection of global maladjustment rather than of specific psychotic symptomatology.     

Empirical Correlates of the MMPI–2 Restructured Clinical (RC) Scales in Psychiatric Inpatients

We identified empirical correlates of the MMPI–2 Restructured Clinical (RC) scales in 1,872 male and 498 female psychiatric inpatients drawn from 2 large tertiary care medical centers. We generated clinical criteria from a systematic review of the patients' intake and discharge medical records. We report zero order correlations between RC scales and clinical criteria and relative risk ratios for dichotomous variables. We found the RC scales to be correlated with conceptually relevant criteria such that for each scale, we identified significant increases in the risk for a variety of emotional, cognitive, and behavioral problems among individuals whose T score exceed 64.     

Psychometric Stability of Rewritten MMPI Items

The current form of the Minnesota Multiphasic Personality Inventory (MMPI) has been in use for over 50 years with no revisions in its item content or wording. As part of the MMPI Restandardization Project (Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer 1989), 82 of the original items were rewritten. Subjects in this study were administered either the original booklet twice to serve as a baseline or the original and the experimental form (MMPI-AX) in a counterbalanced order. Comparisons were made between the percentage of subjects who responded similarly to the two administrations of the original items and the percentage of subjects responding similarly to the original and rewritten versions of the items. Of the 82 items, only 9 changed significantly and only for one gender. None of the correlations between these 9 items and the MMPI validity, clinical, or special scales they are scored on changed significantly for the rewritten items as compared with the original items.     

Social desirability and Wiggin's MMPI content scales.

Using the Minnesota Multiphasic Personality Inventory (MMPI) item pool, Wiggins (1966) developed 13 scales, each with a homogeneous content. The 13 scales, along with marker scales for the 1st 3 MMPI factors, Edward's social desirability (SD), Welsh's repression (R), and Wiggins's social desirability (Sd) respectively were scored in the MMPI. The same scales were scored in an Experimental Multiphasic Personality Inventory (EMPI). A principal-components analysis of the 16 scales when scored in the MMPI resulted in 4 factors. A principal-components analysis of these same scales when scored in the MMPI and when scored in the EMPI were found to be highly congruent. The SD, R, and Sd scales proved to be excellent markers for the 1st 3 factors of the MMPI and also for the 1st 3 factors of the EMPI. Results provide further evidence that the 1st MMPI factor is a social desirability factor rather than a content factor.     

The Utility of MMPI–2–RF Scale Scores in the Differential Diagnosis of Schizophrenia and Major Depressive Disorder

This study was designed to determine whether scores on selected Minnesota Multiphasic Personality Inventory–2–Restructured Form (MMPI–2–RF) scales could be used to differentiate between individuals diagnosed with schizophrenia (SCZ) and major depressive disorder (MDD). The sample was drawn from 2 psychiatric inpatient hospitals and included data from 199 individuals with SCZ and 808 individuals with MDD. A series of multivariate analyses of variance, analyses of variance, and odds ratios were calculated to determine which MMPI–2–RF scales provide the best differentiation between individuals presenting with these 2 disorders. Results indicated scales assessing internalizing dysfunction, including Emotional/Internalizing Dysfunction (EID), Restructured Clinical Scales Demoralization (RCd), Low Positive Emotions (RC2), Suicidal/Death Ideation (SUI), and Self Doubt (SFD) best discriminated MDD from SCZ. Scales assessing thought dysfunction, incluidng Thought Dysfunction (THD), Restructured Clinical Scales Ideas of Persecution (RC6) and Aberrant Experiences (RC8), and Psychoticism-Revised (PSYC-r) were demonstrated to best identify SCZ. Comparisons of the examined MMPI–2–RF scales to MMPI–2 scales assessing similar constructs suggested scales from the MMPI–2–RF perform similarly to their MMPI–2 counterparts in detecting MDD or SCZ, but might have increased ability to discriminate SCZ from other conditions. Overall, results of this study suggest that scores on the examined MMPI–2–RF scales provide important information about the differential diagnosis of MDD and SCZ to clinicians working in inpatient settings.     

Using IRT to separate measurement bias from true group differences on homogeneous and heterogeneous scales: an illustration with the MMPI

The authors present a didactic illustration of how item response theory (IRT) can be used to separate measurement bias from true group differences on homogeneous and heterogeneous scales. Several bias detection methods are illustrated with 12 unidimensional Minnesota Multiphasic Personality Inventory (MMPI) factor scales (Waller, 1999) and the 13 multidimensional MMPI validity and clinical scales. The article begins with a brief review of MMPI bias research and nontechnical reviews of the 2-parameter logistic model (2-PLM) and several IRT-based methods for bias detection. A goal of this article is to demonstrate that homogeneous and heterogeneous scales that are composed of biased items do not necessarily yield biased test scores. To that end, the authors perform differential item- and test-functioning analyses on the MMPI factor, validity, and clinical scales using data from 511 Blacks and 1,277 Whites from the California Youth Authority.     

Relations between the family environment scale (FES) and the MMPI

This study investigated the relations between FES scales and MMPI scales and the relation of social desirability and endorsement of FES items in a sample of 185 college students. FES scales were found generally not to be highly redundant with MMPI variables although MMPI Scale K (defensiveness) was significantly related to several FES scales. The social desirability and the endorsement rates of FES items were found to correlate approximately .80. Implications of these findings for the research and clinical use of the FES are discussed.     

Oral-dependence and gender: factors in help-seeking response set and self-reported psychopathology in psychiatric inpatients

Psychoanalytic theory suggests that orality and being female are positively related to adopting a help-seeking stance in the world. This may result in inflated scores on self-report psychological tests. These hypotheses were investigated using Rorschach-based, oral-dependence scores (Masling, 1986), the Minnesota Multiphasic Personality Inventory (MMPI) help-seeking F minus K index (Gough, 1950; Meehl, 1951), and the MMPI clinical and supplementary scales. High orals are found to be more likely than low orals to adopt a help-seeking response set which in turn results in inflated scores on a number of MMPI clinical scales. Contrary to stereotypes, women are no more likely than men to be help seeking or oral dependent. Several other significant findings, including that orality is positively related to MMPI Scale 2 (Depression) scores for women, not men, are also discussed.     

Oral-Dependence and Gender: Factors in Help-Seeking Response Set and Self-Reported Psychopathology in Psychiatric Inpatients

Psychoanalytic theory suggests that orality and being female are positively related to adopting a help-seeking stance in the world. This may result in inflated scores on self-report psychological tests. These hypotheses were investigated using Rorschach-based, oral-dependence scores (Masling, 1986), the Minnesota Multiphasic Personality Inventory (MMPI) help-seeking F minus K index (Gough, 1950; Meehl, 1951), and the MMPI clinical and supplementary scales. High orals are found to be more likely than low orals to adopt a help-seeking response set which in turn results in inflated scores on a number of MMPI clinical scales. Contrary to stereotypes, women are no more likely than men to be help seeking or oral dependent. Several other significant findings, including that orality is positively related to MMPI Scale 2 (Depression) scores for women, not men, are also discussed.