Psychopathy, IQ, and Violence in European American and African American County Jail Inmates Zach Walsh, Marc T. Swogger, and David S. Kosson Rosalind Franklin University of Medicine and Science The accuracy of the prediction of criminal violence may be improved by combining psychopathy with other variables that have been found to predict violence. Research has suggested that assessing intelli- gence (i.e., IQ) as well as psychopathy improves the accuracy of violence prediction. In the present study, the authors tested this hypothesis by using a contemporary measure of psychopathy, the Psychopathy Checklist—Revised (R. D. Hare, 2003), in a sample of 326 European American and 348 African American male offenders. The postdictive power of psychopathy was evident for both ethnic groups and robust across most changes in the operationalization of violence and the analysis conducted, whereas the postdictive power of IQ was not. No Psychopathy  IQ interactions were identified. Implications of these results for violence prediction are discussed. The prediction of violence is an important clinical challenge, and even a small increase in predictive power may have significant practical benefit. The failure of unaided clinical judgment to ac- curately predict violence (Monahan, 1984) has necessitated the development of more complex approaches to prediction that in- corporate multiple variables (Dolan & Doyle, 2000). One variable of particular importance in predicting violence is psychopathy (Hare, 2003), which has proven robust in its ability to predict violence independent of other factors (Hare, Clark, Grann, & Thornton, 2000). IQ is another variable that has been demonstrated to predict violent behavior (Gendreau, Goggin, & Little, 1997; Raine, 1993). The ability of these variables to independently predict violence makes an assessment of their combined predictive power both theoretically and practically important, and the com- bination of variables such as these into actuarial methods for violence prediction characterizes an important trend within foren- sic psychology (Dolan & Doyle, 2000). Although such approaches have proven effective at maximizing prediction, they do not pro- vide information about important interactions between predictors. Indeed, it is not well established whether IQ and psychopathy predict violence independently or through an interactive mecha- nism. A series of studies by Heilbrun (1979, 1982, 1990) led him to propose that IQ moderates the effect of psychopathy on violence such that low IQ psychopathic participants present greater risk for violence than do high IQ psychopathic participants or both high and low IQ nonpsychopathic participants. In contrast, a study by Holland, Beckett, and Levi (1981) supports an additive linear model relating violence to both IQ and psychopathy. These studies measured psychopathy with various self-report measures, a method Heilbrun (1990) described as “far from ideal” (p. 148). Since then, the development of the Psychopathy Check- list—Revised (PCL–R; Hare, 2003) has contributed greatly to the power of psychopathy as an independent predictor of violence. However, to our knowledge there are no published tests of Heilbrun’s hypothesis with the PCL–R. The present study was designed to provide such a test. In particular, we examined whether psychopathy and intelligence interact to predict violence, with low IQ psychopathic offenders committing the greatest number of violent crimes, or whether psychopathy predicts violence equally well across the full range of intelligence scores in an offender sample. Psychopathy can be studied both continuously and via compar- isons of extreme groups (Hare, 2003). To examine prediction across the full range of psychopathy and IQ scores, we conducted hierarchical multiple and logistic regressions. One advantage of these techniques is that they allow examination of relationships between a predictor and a criterion after controlling for other predictors. However, because Heilbrun’s hypothesis addressed specific comparisons of high IQ and low IQ groups, we also tested the hypothesis via a set of a priori planned comparisons. Several of the relationships between psychopathy and maladap- tive behavior identified in European American offenders do not replicate among African American offenders (Doninger & Kosson, 2001; Kosson, Smith, & Newman, 1990; Lorenz & Newman, 2002). The validity of IQ estimates across ethnic groups has also been questioned (Fish, 2002), and numerous authors have noted important differences in the accrual of charges in European Amer- ican and African American criminals (Barak, Flavin, & Leighton, 2001). In light of these potentially important differences in the Zach Walsh, Marc T. Swogger, and David S. Kosson, Department of Psychology, Rosalind Franklin University of Medicine and Science. The research and preparation of this article were supported in part by National Institute of Mental Health Grants MH49111 and MH57714 to David S. Kosson. We thank Larry Lesza, Charles de Filippo, and the staff of the Lake County Jail in Waukegan, Illinois, for their consistent coop- eration and support during the conduct of this research. We also thank Carolyn Abramowitz, Katherine Aires-Byrnes, Maria Banderas, Nick Don- inger, Seoni Llanes-Macy, Andrew Mayer, Sarah Miller, and Elizabeth Sullivan for interviewing the inmates. Correspondence concerning this article should be addressed to Zach Walsh, Department of Psychology, Rosalind Franklin University of Med- icine and Science, 3333 Green Bay Road, North Chicago, IL 60064. E-mail: zcwalsh@hotmail.com Journal of Consulting and Clinical Psychology Copyright 2004 by the American Psychological Association 2004, Vol. 72, No. 6, 1165–1169 0022-006X/04/$12.00 DOI: 10.1037/0022-006X.72.6.1165 1165