Source: (2013) NSW Bureau of Crime Statistics and Research. Crime and Justice Bulletin. Number 172.

Aim: To determine whether the NSW Forum Sentencing program is more effective than the conventional sentencing process in reducing recidivism. Method: Offenders referred to Forum Sentencing in 2011 were matched with offenders who were ‘eligible’ for Forum Sentencing but who were sentenced in a NSW Local Court where Forum Sentencing was not operating. These two groups were matched on a large number of covariates using propensity score techniques and were then compared on the time to first new proven offence using Cox regression. All offenders in both groups were followed up for a minimum of 6 months after finalisation of their index offence. The analysis was conducted using an intention-to-treat research design. Results: Of the 575 offenders referred to Forum Sentencing, 552 could be matched with an ‘equivalent’ offender in the control group based on the covariates measured at the index court appearance. Cox regression analyses showed that there was no significant difference between the matched groups in the time to first new offence. This lack of an effect remained even after controlling for other relevant covariates. Conclusion: This study finds no evidence that offenders who are referred to the NSW Forum Sentencing program are less likely to re-offend than similar offenders who are dealt with through the normal sentencing process. (summary from the report)