Construct validation of a self-report psychopathy scale: does Levensonʹs self-report psychopathy scale measure the same constructs as Hareʹs psychopathy checklist-revised?

In an effort to validate Levenson, Kiehl and Fitzpatrickʹs [Levenson, M. R., Kiehl, K. A., & Fitzpatrick, C. M. (1995). Assessing psychopathic attributes in a noninstitutionalized population. Journal of Personality and Social Psychology, 68, 151–158]. Self-report Psychopathy Scale (SRPS) we compared it to Hareʹs [Hare, R. D. (1991). The Hare Psychopathy Checklist-Revised. Toronto: Multi-Health Systems] (PCL-R) and examined its relation to criminal activity and a passive avoidance task. Participants were 270 Caucasian and 279 African-American participants in a minimum security state prison. Confirmatory factor analysis provided modest support for the original SRPS factor structure. Although diagnostic concordance of the two instruments ranged from poor to fair, the SRPS and the PCL-R were significantly correlated and both showed similar patterns of correlations to measures of substance abuse and criminal versatility. Both measures were also predictive of performance on a passive avoidance task. While this constellation of findings provides some evidence for the construct validity of the SRPS, it also suggests that the SRPS may not measure the same construct as the PCL-R and further refinement of the instrument appears to be warranted.