A multigroup item response theory analysis of the Psychopathy Checklist-Revised

Item response theory was used to investigate the functioning of the Psychopathy Checklist-Revised (PCL-R; R. D. Hare, 1991, 2003)in several offender populations. With male criminal offenders (N = 3,847) as a reference group, differential item functioning analyses were performed for 3 comparison groups: female criminal offenders (N = 1,219), male forensic psychiatric patients (N = 1,246), and male criminal offenders scored from file reviews (N = 2,626). Results are discussed in the context of the 2-factor, 4-facet model for the PCL-R (R. D. Hare, 2003; J. Parker, G. Sitarenios, & R. D. Hare, 2003). Application of a multigroup graded response model to all 4 groups suggests scalar equivalence may hold at least approximately for each population, although the PCL-R provided slightly greater information about the latent trait of psychopathy for male criminal offenders scored from the standard procedure.