Unpacking the Impact of Restorative Justice in the Rise Experiments: Facilitators, Offenders, and Conference Non-Delivery

Restorative Justice (RJ) programs are often evaluated in terms of their outcomes, with little attention to the process. Typically we analyze average effects across individuals who experience RJ differently. The present dissertation unpacks these different effects in three separate inquiries utilizing data from the Reintegrative Shaming Experiments (RISE) conducted in Canberra, Australia from 1995 2000. First, we descriptively assess the extent RJ conference facilitators engender perceptions of procedural justice and legitimacy in offenders. We examine the number of conferences delivered (experience), sequential conferences (practice-makes-perfect) and the timing between conferences (skill maintenance). Certain conference facilitators are better than others from the outset. We recommend the identification of RJ facilitators who are good at promoting perceptions of procedural justice and legitimacy. Second, we utilize trajectory analysis and find the impact of RJ varies by offending group, with negative effects observed for Aboriginal offenders. Finally, utilizing multinomial logistic regression, we examine the characteristics associated with non-delivery of RJ. Randomized controlled trials, such as RISE, rely on treatment integrity to best assess the impact of the assigned treatment. From a policy standpoint, the most efficient use of resources would rely on successful conference delivery. We find that the time between random assignment and the first conference attempt is significantly related to successful delivery. This dissertation takes important steps in understanding the importance of unpacking the impact of RJ and helps inform who should conduct RJ conferences, what groups of individuals to include in future studies, and what impacts non-delivery of RJ conferences. Degree Type Dissertation Degree Name Doctor of Philosophy (PhD) Graduate Group Criminology First Advisor Lawrence Sherman Second Advisor Richard Berk Third Advisor John MacDonald This dissertation is available at ScholarlyCommons: http://repository.upenn.edu/edissertations/73