|dc.description.abstract||Studies using latent profile analysis of the Psychopathy Checklist – Revised (PCL-R) report that groups differ on external correlates relevant to forensic evaluations. Some of these external correlates are being used to support the use of the PCL-R in high stakes forensic settings, such as capital sentencing (Olver et al., 2020). For subtype findings to be useful in practice, clinicians need to be able to reliably assign offenders to correct subgroups.
The current study contained two parts, both of which aimed to address whether individuals can accurately classify PCL-R profiles into their correct subtype as found by subtyping research performed by McCallum and colleagues (2020). In the first part of the study, psychology doctoral students (N = 12) were able to classify profiles with moderate to high accuracy, with some differences based on subtype and whether the data were presented as mean item scores or summed facet scores. The overall difference in accuracy between the mean item and summed facet scores was not statistically significant.
The second part of the study asked clinicians to classify PCL-R profiles into their empirically based subtype. The profiles were presented as a score sheet similar to the Scoring Grid presented in the manual (Hare, 2003) to increase generalizability. Clinicians (N = 37) were better able to classify the prototypic subtype, with more difficultly on the sociopathic and callous-conning subtypes. As prototypicality of the profiles lowered, clinicians had more difficulty with accurate classification.
Overall, this study shows that individuals are better at classifying some subtypes compared to others; however, more research is needed to investigate what differences are causing discrepant classification accuracy between the subtypes. Further, although the overall difference between the two presentation methods in the first study (mean item scores and summed facet scores) was not statistically significant, participants stated that the profiles presented as mean item scores were easier to classify. This may be because they better map on to the way the subtypes are presented in the research literature. Researchers may want to consider presenting their findings in a manner more consistent to what practitioners have in the field to facilitate more effective use of their findings.||