The PCL–R and capital sentencing: A commentary on “Death is different” DeMatteo et al. (2020a).

Abstract

DeMatteo et al. (2020a) published a Statement in this journal declaring that the Psychopathy Checklist-Revised (PCL-R) “cannot and should not” be used in U.S. capital-sentencing cases to assess risk for serious institutional violence. Their stated concerns were the PCL-R’s “imperfect interrater reliability,” its “variability in predictive validity,” and its prejudicial effects on the defendant. In a Cautionary Note, we (Olver et al., 2020) raised questions about the Statement’s evaluation of the PCL-R’s psychometric properties, presented new data, including a meta-meta-analysis, and argued that the evidence did not support the Statement’s declaration that the PCL-R “cannot” be used in high stakes contexts. In their reply, titled “Death is Different,” DeMatteo et al. (2020b) concurred with several points in our Cautionary Note, disputed others, asserted that we had misunderstood or mischaracterized their Statement, and dismissed our new data and comments as irrelevant to the Statement’s purpose. This perspective on our commentary is inimical to balanced academic discourse. In this article, we contend that DeMatteo et al. (2020b) underestimated the reliability and predictive validity of PCL-R ratings, overestimated the centrality of the PCL-R in sentencing decisions, and underplayed the importance of other factors. Most of their arguments depended on sources other than capital cases, including mock trials, Sexually Violent Predator (SVP) hearings, and studies that included the prediction of general violence. We conclude that the rationale for the bold “cannot and should not” decree is open to debate and in need of research in real-life venues.