New Study Finds Test Often Used in Sexually Violent Predator Evaluations Is Unreliable

A new study finds that the Psychopathy Checklist-Revised, which is often used to determine the civil commitment of offenders as sexually violent predators, is unreliable.  The study is “The Role and Reliability of the Psychopathy Checklist-Revised in U.S. Sexually Violent Predator Evaluations: A Case Law Survey” by DeMatteo, D., Edens, J. F., Galloway, M., Cox, J., Toney Smith, S. and Formon, D. in Law and Human Behavior (2013).

Here is the abstract from American Psychiatric Association PsychNET Direct:

The civil commitment of offenders as sexually violent predators (SVPs) is a highly contentious area of U.S. mental health law. The Psychopathy Checklist—Revised (PCL–R) is frequently used in mental health evaluations in these cases to aid legal decision making. Although generally perceived to be a useful assessment tool in applied settings, recent research has raised questions about the reliability of PCL–R scores in SVP cases. In this report, we review the use of the PCL–R in SVP trials identified as part of a larger project investigating its role in U.S. case law. After presenting data on how the PCL–R is used in SVP cases, we examine the reliability of scores reported in these cases. We located 214 cases involving the PCL–R, 88 of which included an actual score and 29 of which included multiple scores. In the 29 cases with multiple scores, the intraclass correlation coefficient for a single evaluator for the PCL–R scores was only .58, and only 41.4% of the difference scores were within 1 standard error of measurement unit. The average score reported by prosecution experts was significantly higher than the average score reported by defense-retained experts, and prosecution experts reported PCL–R scores of 30 or above in nearly 50% of the cases, compared with less than 10% of the cases for defense witnesses (κ = .29). In conjunction with other recently published findings demonstrating the unreliability of PCL–R scores in applied settings, our results raise questions as to whether this instrument should be admitted into SVP proceedings. (PsycINFO Database Record (c) 2013 APA, all rights reserved)

NPR covered the rise of the PCL-R in a 2011 story, describing how it began as a research tool and then was coopted by the criminal justice system to the dismay of its creator who “feared that the test, created purely for research purposes, might be used incorrectly in the real world and could hurt people.”