The effects of MMPI-A T-score elevation on classification accuracy for normal and clinical adolescent samples

In this investigation we examined the ability of the Minnesota Multiphasic Personality Inventory-Adolescent (MMPI-A; Butcher et al., 1992) to classify accurately both clinical and normal adolescents using 2 different T-score elevation ranges, T > or = 60 and T > or = 65, and using 2 different clinical base rates for the occurrence of psychopathology. A clinical base rate of 50% and 20%, respectively, were created by comparing a clinical sample of 203 adolescent inpatients with cooccurring substance abuse and psychiatric disorders with 2 subsamples from the MMPI-A normative group. These subsamples consisted of 203 adolescents matched for sex and age, and a larger subsample of 1,015 adolescents proportionately matched for sex and age, with the clinical group. Classification accuracy analyses revealed that although clinical base rate did affect the accurate classification of cases, a T-score cutoff of 65 resulted in higher levels of accurate classification overall while minimizing the misclassification of both clinical and normal cases. Implications of these findings for the recommended use of the MMPI-A "gray zone" are presented, and the relative areas of strength and weakness of the MMPI-A are reviewed in the identification and description of psychopathology.