The Efficacy of Pretrial Risk Assessment Instruments: Exploring Questions of Predictive Validity

• In Event: Disparities in Pretrial Justice

Abstract

Over the last decade there has been considerable attention to the U.S. money bail system, including its connection to rising levels of pretrial detention marked by sizeable racial and ethnic disparities. In response, jurisdictions across the country have implemented a range of strategies to reduce their jail populations safely and efficiently. Among these, pretrial risk assessment tools have emerged as a leading approach to help jurisdictions reduce their reliance on financial forms of release. Evidence regarding the impact of these tools on pretrial processes or their efficacy in predicting pretrial failure, however, are somewhat mixed. In the current investigation, we draw on multiple administrative data sources to explore the impact of a pretrial risk assessment tool on a range of pretrial outcomes. In addition to considering the overall predictive validity of the tool, we focus on the contribution of its different components and, in a supplemental set of analyses, present a redesigned tool that is comparable in accuracy relying on fewer items. We discuss the implications of our findings for policy and practice, focusing considerable attention on the issue of predictive validity and how we determine the acceptable threshold of accuracy for high-stakes decisions in the criminal justice domain.

Author

• Jennifer Copp, Florida State University