Abstract
This article reports scale-level factor analyses for two widely used self-report measures of psychopathology — the Symptom Checklist-90-R (Derogatis, 1983) and the Millon Clinical Multiaxial Inventory (Millon, 1983) — and compares the obtained factor structures with findings reported in the literature to determine whether each instrument possesses a reliable, meaningful dimensional structure underlying its scale scores. The sample was a heterogeneous group of psychiatric inpatients and outpatients (N = 130). The SCL-90 scale scores formed two highly correlated factors (anxious depression and paranoid thinking), although the scale intercorrelations could be adequately accounted for by a single factor (as in previous reports). The MCMI scale scores formed three factors (anxious depression and emotionality, paranoid and manic thinking, and schizoid thinking), of which the first and second were also highly intercorrelated. Supplementary analyses indicated that to a considerable degree the MCMI factor invariance was an artifact of item overlap among the scales. An interbattery factor analysis was then performed to determine whether any common factors could describe the variance shared among the two instruments' subscales. Two interbattery factors were obtained, representing anxious depression and emotionality and paranoid thinking respectively. The two measures, when used separately, appear to offer only limited interpretability of scale profiles, although their combined use appears to permit differentiation between two major symptom configurations.
Original language | English (US) |
---|---|
Pages (from-to) | 1-20 |
Number of pages | 20 |
Journal | Multivariate Behavioral Research |
Volume | 27 |
Issue number | 1 |
DOIs | |
State | Published - Jan 1 1992 |
ASJC Scopus subject areas
- Statistics and Probability
- Experimental and Cognitive Psychology
- Arts and Humanities (miscellaneous)