Recidivism in the Federal Justice System: A Novel Use of FJSP Data

• In Event: Using Administrative Data for Evidence Building: Findings from the BJS’ Federal Justice Statistics Program

Abstract

In criminal justice research there is a rich literature on recidivism, usually centered on developing prediction instruments, evaluating programs/interventions and analyzing correlates of recidivism outcomes. Moreover, the nature of recidivism analyses is one in which studies are usually limited to examination at a specific point in the criminal justice process, i.e., beginning after an offender has been released from jail or prison. However there is much still to learn about the relationships between factors predicting and/or influencing recidivism, events occurring throughout criminal justice process and descriptions collected therein. In this paper we describe how the Federal Justice Statistics Program (FJSP), an on-going data collection program sponsored by the Bureau of Justice Statistics, offers new potential for conducting recidivism analyses using data combined across different stages of the criminal system among federal offenders. Specifically the FJSP compiles data on nearly all phases of the criminal process: prosecution, pretrial supervision, sentencing, prison stay, and post-release supervision. We argue that such data provide a platform for analysts to consider and study recidivism (during both pretrial release and post-conviction periods) as a collection of experiences throughout the criminal justice process, and that there are significant new insights to be learned by doing so.

Authors

• Jeremy Luallen, Abt Associates, Inc.

• Christina Dyous, Abt Associates, Inc.

• Ryan Kling, Abt Associates, Inc.