Author Description

Internationally, governments are making renewed efforts to reduce rates of re-offending. Measuring progress against this objective is difficult because officially recorded reconviction rates are determined not only by the effectiveness of the criminal justice system in dealing with offenders, but also by the characteristics of offenders moving through the justice system. The Group Risk Assessment Model (GRAM) is a statistical technique designed to obtain more accurate estimates of trends in re-offending by adjusting for the characteristics of offenders coming through the justice system (see Smith & Jones 2008). This earlier work focussed only on offenders given non-custodial sanctions. The purpose of the current study was to extend GRAM to adjust for the characteristics of prisoners being released from the NSW adult prison system. A number of groups of prisoners were found to be at greater risk of reconviction within two years of release: younger offenders, Indigenous offenders, offenders released to parole (compared with those released without supervision), offenders who had served medium-length sentences (61-364 days, compared with those serving less than 61 days or greater than 364 days), offenders who had more prior convictions and offenders who had a prior conviction for a breach of a justice order, a non-aggravated violent offence or a theft offence. An application of the model revealed that, after adjusting for offender characteristics, there was no significant change in rates of re-offending among prisoners released in 2003 or 2004 when compared to prisoners released in 2002.