Discovering Dimensionality: Factor Analysis of Personality Disorder Spectra Scales for the MMPI-2

A longstanding debate within the field of psychological assessment centers on whether personality disorders (PDs) are better classified by categorical (i.e., syndrome) or dimensional models. In fact, the most recent version of the Diagnostic and Statistical Manual for Mental Disorders (DSM-5) includes an optional, dimensional model, the Alternative Model for Personality Disorders (AMPD). The current studies examined the underlying factor structure of the MMPI-2 PD Spectra Scales developed by Mulay et al. (2018) to determine whether the Scales were more reflective of a dimensional or categorical model for PDs. Study 1 consisted of exploratory factor and bifactor analyses of the entire Spectra Scales item pool (220 items), tested on a sample of 1,030 outpatients seen at a university community clinic. Results supported an exploratory factor model made up of six factors and an exploratory bifactor model made up of a general factor and five specific factors, both of which were fit to a 53-item subset of the original 220 PD Spectra Scales items following item deletion procedures. Study 2 sought to confirm the factor structures resulting from the Study 1 analyses using confirmatory factor and confirmatory bifactor analyses. Criterion validity (based on MCMI-II Clinical Scales) and reliability analyses of the final bifactor model were also completed in Study 2. Participants were 539 inpatients and outpatients seen at a military clinic. Results supported the bifactor model over the factor model. The final bifactor model demonstrated good criterion validity and excellent internal consistency for the model as a whole. The bifactor model is interpreted in the context of preexisting dimensional models for PDs, and potential clinical utility and recommendations for future research are discussed.