Summary
Aim
To compare reoffending rates between adults given a community service order (CSO) and those given a bond (both supervised and unsupervised) or suspended sentence (both supervised and unsupervised).
Method
Using propensity score matching, offenders who were given CSOs were matched to those given bonds and separately to those given suspended sentences. Bonds and suspended sentences were separately considered with supervision, without supervision and as a whole group (with and without supervision). The matched samples were then compared on reoffending rates with and without adjustment for potential covariates.
Results
The CSO group and the whole bond group were able to be matched successfully. The unsupervised bond group also was successfully matched to the CSO group. In both cases, reoffending rates were lower in the CSO group after controlling for other relevant factors. Neither the supervised bond group, nor the suspended sentence groups, could be matched successfully with the CSO group, and therefore conclusions from this analysis are not considered representative or reliable.
Conclusion
Adults given a CSO are less likely to reoffend than offenders given a bond, holding other relevant and available characteristics equal.