Exploring the Hierarchical Structure of the MMPI–2–RF Personality Psychopathology Five in Psychiatric Patient and University Student Samples

Abstract

In this study our goal was to examine the hierarchical structure of personality pathology as conceptualized by Harkness and McNulty's (1994) Personality Psychopathology Five (PSY–5) model, as recently operationalized by the MMPI–2–RF (Ben-Porath & Tellegen, 2011) PSY–5r scales. We used Goldberg's (2006) “bass-ackwards” method to obtain factor structure using PSY–5r item data, successively extracting from 1 to 5 factors in a sample of psychiatric patients (n = 1,000) and a sample of university undergraduate students (n = 1,331). Participants from these samples had completed either the MMPI–2 or the MMPI–2–RF. The results were mostly consistent across the 2 samples, with some differences at the 3-factor level. In the patient sample a factor structure representing 3 broad psychopathology domains (internalizing, externalizing, and psychoticism) emerged; in the student sample the 3-factor level represented what is more commonly observed in “normal-range” personality models (negative emotionality, introversion, and disconstraint). At the 5-factor level the basic structure was similar across the 2 samples and represented well the PSY–5r domains.

Acknowledgments

Martin Sellbom is now at The Australian National University, Canberra, ACT, Australia.

Notes

Digman (1997) proposed an alpha factor that accounts for the positive correlations among scales measuring Big Five Agreeableness, Conscientiousness, and Emotional Stability (low Neuroticism), and a beta factor that accounts for the positive correlations between scales measuring Big Five Openness to Experience and Extraversion. Digman interpreted alpha and beta, touted as the “Big Two” personality traits, to represent broad personality dimensions of socialization (alpha) and self-actualization (beta).

We also conducted the analyses using SPSS and principal components analysis (PCA) with varimax rotation. This procedure produced a nearly identical pattern of results as that obtained with our analyses. Component and factor loading matrices from both the principal components and ESEM analyses are available on request.