Background: Most research on the Personality Inventory for DSM-5 (PID-5) was conducted with self-reports. One of the specific areas for which a multimethod design has yet to be implemented is for the PID-5's associations with aggression. The main objectives of this study were to (a) compare the PID-5 associations with self-reported and file-rated aggression, (b) compare these associations between women and men, and (c) identify the relative importance of PID-5 facet predictors.
Methods: A sample of outpatients with personality disorder (N = 285) was recruited in a specialized public clinic to complete questionnaires, and a subsample was assessed for file-rated aggression (n = 227). Multiple regression analyses were performed with PID-5 facets as statistical predictors but using distinct operationalizations of aggression (self-reported vs. file-rated). Moderation analyses were performed to identify the moderating effect of biological sex. Dominance analyses were computed to identify the relative importance of predictors.
Results: PID-5 facet predictors of self-reported and file-rated aggression were very consistent in both conditions. However, the amount of explained variance was reduced in the latter case (from 39% to 14%), especially for women (from 40% to 2%). The most important predictors were Hostility, Risk Taking, and Callousness.
Conclusion: Pertaining to the statistically significant facets associated with aggression, strong evidence of multimethod replication was found. The women-men discrepancies were not most obvious in their specific associations with aggression, but rather in their amount of explained variance, maybe reflecting examiners' or patients' implicit biases, and/or different manifestations of aggression between women and men.
Keywords: Alternative Model for Personality Disorders; Personality Inventory for DSM-5; aggression; dominance analysis; multimethod assessment; personality disorders.
Copyright © 2022 The Authors. Published by Elsevier Inc. All rights reserved.