Psychopathy is a personality trait associated with persistent antisocial behavior. Research has documented the staggering costs associated with antisocial behavior result from the actions of a few individuals, many of whom exhibit psychopathic traits. Thus, the importance of identifying and treating these individuals is of paramount concern. The majority of psychopathy research utilizes adult Caucasian male participants; however, the validity of the construct in youth, females, and minorities remains unresolved (Sharp & Kine, 2008). Furthermore, the factor structure of psychopathy is the subject of considerable debate (e.g., Neumann, Kosson, & Salekin, 2007 and Cooke, Michie, & Skeem, 2007).
This dissertation examined psychopathic traits in a large sample of adjudicated adolescents in an effort to better understand the extent to which results from adult males generalize to other populations. The Global Risk Assessment Device (GRAD; Gavazzi, Slade, Buettner, Partridge, Yarcheck, & Andrews, 2003) is a risk and needs classification device for adolescent offenders used by court personnel for rehabilitation recommendations prior to sentencing. Exploratory and confirmatory factor analyses were performed on GRAD items in an effort to develop a measurement model of psychopathy and investigate race and sex differences. Next, regression analyses were employed for construct validation purposes.
Results indicated a three factor model of psychopathy provided the best fit for Caucasian males, consistent with the results of Cooke and Michie (2001). The model developed with Caucasian males also fit well for samples of Caucasian females, African-American males, and African-American females. The measurement model predicted a significant amount of variance in criminal behavior as well as a number of variables related to externalizing and internalizing symptoms. The Impulsivity/Conduct Problems factor was strongly and consistently related to all of these outcome variables, suggesting it represents a risk factor for both externalizing and internalizing psychopathology. The Callous-Unemotional Traits factor was also related to antisocial behavior, albeit less so than Impulsivity/Conduct Problems. Narcissism was positively related to violence. A few noteworthy race and sex differences emerged. First, the model predicted outcome variables as well or better for females as it did for males. Second, the model predicted serious crime less well for African-Americans than for Caucasians.
Overall, results suggest psychopathy is a valid and useful construct for understanding youth antisocial behavior across sex and race. Including additional items for each of the three psychopathy factors in order to ensure sufficient coverage of each dimension may enhance the GRAD as a risk and needs classification device for adolescent offenders. Future research with the GRAD would also benefit from the inclusion of well-validated measures of externalizing and internalizing psychopathology as well as collateral reporters.