Abstract
Purpose
Poythress, Edens, et al. (2010) recently used cluster analysis to identify subtypes of antisocial and psychopathic offenders using a diverse collection of theoretically important clustering variables. Two predicted subtypes, primary and secondary psychopathy, were identified, in addition to non-psychopathic and (unexpectedly) “fearful” psychopathic offenders. The purpose of the present research was to determine whether these clusters could be replicated using a single self-report measure, the Psychopathic Personality Inventory (PPI; Lilienfeld & Andrews, 1996).
Method: Study 1
We used discriminant function analysis (DFA) to predict cluster membership for the Poythress et al. subtypes based solely on the eight subscales of the PPI.
Introduction
Mental health professionals have been interested in the prevention of delinquent and criminal behavior and the rehabilitation of offenders for centuries (e.g., Raphael, Jacoby, Harryman, & Raphael, 1924). The dominant classification system for mental disorders in use in the United States today, the Diagnostic and Statistical Manual of Mental Disorders—4th Edition-Text Revision (DSM-IV-TR; American Psychiatric Association, 2000), categorizes most of those who engage in repetitive criminal conduct to be suffering from Antisocial Personality Disorder (ASPD). Multiple controversies bedevil the validity and utility of the ASPD diagnosis, however, with one frequent criticism being that it comprises a very heterogeneous group of individuals who share little in common other than a propensity to engage in irresponsible behavior, including crime. A considerable amount of recent research suggests that people who meet criteria for ASPD can be meaningfully subtyped into more homogeneous groups (described below) based on psychological characteristics. However, the clinical utility of these subtypes to inform treatment and management decision-making requires further investigation.
Another controversy surrounding the categorization of persons who engage in chronic antisocial behavior is the utility of self-report personality and attitudinal measures. Self-report measures have been criticized as problematic on a number of grounds, even though extant research suggests that many have considerable utility in identifying offenders who are prone toward violence or recidivism (Walters, 2006). Self-report measures offer numerous advantages over more labor-intensive assessment methods such as standardized interviews (Lilienfeld & Fowler, 2006) and a method for identifying subtypes of antisocial offenders based largely on self-report would be a major contribution to the assessment literature.
The present study builds on recent subtyping research by Poythress, Edens, et al. (2010), who cluster analyzed a large sample of individuals diagnosed with ASPD into more conceptually meaningful subgroups based on a complex combination of personality measures, including interview data and self-report questionnaires. In this report, we investigate the utility of a single, widely used self-report measure, the Psychopathic Personality Inventory (PPI; Lilienfeld & Andrews, 1996) to classify the original sample into these subtypes using discriminant function analysis (Study 1). In Study 2, we extend this work by examining the utility of this PPI-based system in a new sample of offenders who had completed the PPI as part of an earlier study (Edens, Poythress, Lilienfeld, Patrick, & Test, 2008). More specifically, we address whether prison inmates classified into subtypes are at differential risk for institutional misconduct.
Before presenting these findings, we provide a more extensive review of controversies concerning ASPD. We begin by highlighting conceptual debates regarding the relationship between ASPD and the related concept of psychopathic personality (psychopathy).
