Cross-validation of Levenson's Psychopathy Scale in a sample of federal female inmates.

Levenson, Kiehl, and Fitzpatrick's Self-Report Psychopathy Scale (LSRPS) is evaluated to determine the factor structure and concurrent validity of the instrument among 430 federal female inmates. Confirmatory factor analysis fails to validate the expected 2-factor structure. Subsequent exploratory factor analysis reveals a 3-factor structure (egocentric, antisocial, and callous), where each factor accounts for significant variance in scores on several theoretically relevant measures. Higher scores on the antisocial factor of the LSRPS are associated with a history of varied psychopathological and negative legal outcomes, suggesting evidence of concurrent validity. However, the egocentric and callous factors do not seem to measure precisely the same construct as the primary psychopathy factor from the Levenson et al. study. The 3-factor structure proposed here has been proposed by other researchers and found in other samples of psychopathy in female inmates. Implications for both research and clinical practice using the LSRPS with female inmates are discussed.