Ubuntu-Restorative Conferencing and Juvenile Recidivism in Rwandan Community Courts: A Difference-in-Differences Estimation of Indigenous Dispute-Resolution Exposure

Rwanda's community courts have incorporated Ubuntu-restorative conferencing into juvenile proceedings, yet the causal effect of exposure to this indigenous dispute-resolution practice on recidivism remains unidentified because participation is non-random and recidivism is observed only for adjudicated youth. No observational data are used or claimed. The analysis specifies a before-after-control-impact estimator with between-cluster standard deviation of 6.0, residual standard deviation of 4.5, and intraclass correlation of 0.64, and evaluates power across 2000 replications per design cell for cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points. The design requires 54 clusters to detect a 5-percentage-point reduction at 80% power. The credibility of indigenous-practice evaluations depends less on the choice of estimator than on the variance structure that community-court clustering imposes, and it derives the cluster-count thresholds below which a null result cannot be distinguished from an underpowered design.