RESUMO
In the current study, we used a sample of predominantly African-American women with high rates of trauma exposure (N = 434) to examine psychometric properties of the Personality Inventory for DSM-5-Brief Form (PID-5-BF). We compared model fit between a model with five correlated latent factors and a higher-order model in which the five latent factors were used to estimate a single "general pathology" factor. Additionally, we computed estimates of internal consistency and domain interrelations and examined indices of convergent/discriminant validity of the PID-5-BF domains by examining their relations to relevant criterion variables. The expected five-factor structure demonstrated good fit indices in a confirmatory factor analysis, and the more parsimonious, higher-order model was retained. Within this higher-order model, the first-order factors accounted for more variance in the criterion variables than the general pathology factor in most instances. The PID-5-BF domains were highly interrelated (rs = .38 to .66), and convergent/discriminant validity of the domains varied: Negative Affectivity and Detachment generally showed the hypothesized pattern of relations with external criteria, while Antagonism and Disinhibition displayed less consistent and discriminant relations. Results are discussed in terms of the costs and benefits of using brief pathological trait measures in samples characterized by high levels of psychopathology.