Restorative justice and the evidence hierarchy: implications for policy and practice

Abstract

Restorative justice (RJ) has expanded rapidly across criminal justice systems and related institutional settings, accompanied by a large and growing empirical literature evaluating its effectiveness. Despite this growth, claims about RJ’s impacts often rest on evidence drawn from heterogeneous programs and research designs of varying rigor. This review evaluates the scientific evidence on restorative justice outcomes by assessing the existing literature against established standards of causal inference. Drawing on recent meta-analyses of adult and juvenile RJ programs, we assess inferential capacity rather than produce new estimates. We show that much of the RJ evidence base relies on research designs that limit causal interpretation, non-random referral mechanisms, and substantial program heterogeneity. While some studies report favorable associations between RJ participation and outcomes, these findings are often drawn from less-than-rigorous research designs. The review concludes by identifying gaps in the current evidence base and outlining methodological priorities necessary for generating more credible and generalizable conclusions about restorative justice programs.