Comparisons between the Minnesota Multiphasic Personality Inventory-Adolescent-Restructured Form (MMPI-A RF) and MMPI-A in adolescent psychiatric inpatients.

Abstract

This study explored the association between the Minnesota Multiphasic Personality Inventory (MMPI)-Adolescent-Restructured Form (MMPI-A-RF) and the MMPI-Adolescent (MMPI-A) form in a sample of 3,516 adolescents receiving inpatient psychiatric treatment, including 2,798 adolescents meeting validity inclusion cutoffs for both measures. There was 92.5% agreement rate with respect to global identification of cases as valid or invalid and some empirical support for lowering interpretive cutoffs for validity scales on the MMPI-A-RF. The MMPI-A-RF Demoralization Scale (RCd) was shown to correlate significantly less strongly with Restructured Clinical (RC) scales than with MMPI-A clinical scales. RC scales also demonstrated significantly lower mean interscale correlations than MMPI-A clinical scales. As expected, this greater level of scale independence resulted in significantly fewer profiles with multiple scale elevations. As was anticipated, with the exception of RC1 predicting MMPI-A hypochondriasis, correlational and classification agreement analyses suggested moderate associations between the RC and MMPI-A clinical scales, but somewhat stronger agreement between comparable PSY-5 scales. Changes in interpretive cutoff procedures for the RC scales, including RCd, also resulted in 5.5% fewer "within normal limits" profiles than the use of MMPI-A with all 10 clinical scales. Finally, stepwise linear regression analyses indicated that MMPI-A-RF Higher-Order scales were best predicted by those MMPI-A clinical scale combinations that they are purported to be linked with in the MMPI-A-RF manual. (PsycINFO Database Record