Assessing a Community-Based Restorative Justice Pilot: Design, Implementation, Evaluation

Thesis Abstract

This study addresses the persistent gap between formal punitive approaches and community-recognized justice by evaluating a community-based restorative justice (RJ) pilot designed to reduce reoffending, enhance victim satisfaction, and strengthen local social norms. The aim is to design, implement, and evaluate a scalable RJ pilot within urban and peri-urban neighborhoods, examining process integrity, stakeholder engagement, and substantive outcomes. Specific objectives are (1) to map the stakeholder landscape and co-design an RJ intervention tailored to local crime types and cultural contexts; (2) to implement a structured RJ process for 120 participants (60 offenders, 40 victims, 20 community mediators) across 12 months; (3) to assess process fidelity and participant experiences through a mixed-methods lens; (4) to evaluate impact on recidivism, victim satisfaction, restitution compliance, and community cohesion using rigorous empirical analysis; and (5) to develop a policy-ready model outlining scalable components and guardrails for wider adoption. The methodology integrates a mixed-methods design guided by restorative justice theory and procedural fairness concepts. The quantitative strand employs a quasi-experimental design with a matched comparison group (n ? 120 participants in intervention and n ? 120 in control) and uses regression discontinuity-inspired matching to mitigate selection bias. Key instruments include a standardized victim satisfaction survey, a recidivism tracking dataset from local police records over 18 months, and a fidelity checklist administered at three time points. The qualitative strand uses semi-structured interviews with 40 participants (20 victims, 12 offenders, 8 mediators) and 6 focus groups with community stakeholders, analyzed via thematic analysis. Validity and reliability are supported by triangulation, inter-rater reliability for coders (Cohen’s kappa > 0.70), and pilot testing of survey instruments. Ethical considerations address informed consent, confidentiality, potential re-traumatization, and safeguarding measures, with approval from an institutional review board. Analytical techniques include multivariate regression to estimate the program’s effect on recidivism and victim satisfaction, propensity score matching to balance baseline covariates, and time-to-event analysis for reoffending. Thematic analysis follows Braun and Clarke’s approach to elucidate perceived legitimacy, accountability, and social capital changes. A process evaluation framework (context, implementation, and mechanism) is applied to interpret deviations and fidelity. The study also integrates a cost—benefit perspective by comparing program costs against observed reductions in recidivism and increased restitution uptake. Expected findings indicate that the RJ pilot will produce statistically significant improvements in victim satisfaction (p < .05), a reduction in reconviction rates within 12–18 months post-intervention relative to the control group (hazard ratio < 0.75), and enhanced perceived procedural fairness among participants. Qualitative insights are anticipated to reveal mechanisms through which community mediation fosters accountability and repair, including strengthened neighborhood social ties and legitimacy of the justice process. The research is expected to identify critical implementation determinants such as mediator training quality, timeliness of case processing, and proactive community engagement as core drivers of success. The study contributes to knowledge by providing robust empirical evidence on the design, implementation, and evaluation of community-based RJ in a real-world urban setting, addressing gaps related to scalability, fidelity, and long-term outcomes. It advances theoretical understanding by integrating procedural justice and social capital theories with outcomes data, offering a nuanced account of how ecological context shapes RJ efficacy. The resulting model delineates a scalable blueprint for policymakers and practitioners, detailing entry criteria, stakeholder roles, governance structures, resource requirements, risk management, and an integrated monitoring and evaluation framework. The main conclusion posits that a well-designed, faithfully implemented community-based RJ pilot can meaningfully reduce recidivism and increase victim satisfaction while strengthening community cohesion, provided that fidelity, timely delivery, and genuine stakeholder engagement are prioritized. Policy recommendations include adopting formalized training programs for mediators, allocating dedicated funding for process evaluation, embedding RJ within local policing and community services, establishing transparent accountability mechanisms, and creating a scalable replication template with context-sensitive adaptations. Suggestions for future research emphasize longitudinal studies on long-term crime trends, comparative analyses across diverse communities, and exploration of digital mediation platforms as supplements to in-person processes.