Why is the juvenile re-offending rate higher than expected?

Release Date: 3 February 2011

Research by the NSW Bureau of Crime Statistics and Research has shown that the observed rate of re-offending among juvenile offenders convicted in 2006/7 was significantly higher than the predicted rate of re-offending.

The discrepancy, though small, (57.9% observed rate compared to 54.7% predicted rate) was statistically significant.

The re-offending predictions were made using an actuarial model known as the Group Risk Assessment Model (GRAM) developed by the NSW Bureau of Crime Statistics and Research.

GRAM calculates a predicted rate of re-offending, based on characteristics of the offender, such as age, gender, Indigenous status, offence and prior criminal record.

The model was developed by the Bureau to help evaluate the Government's commitment under the State Plan to reduce the proportion of offenders who re-offend within 24 months of a finalised court appearance or Youth Justice Conference.

Two factors appear to explain the higher than expected rate of re-offending:

The higher than expected rate of reoffending among juveniles dealt with via a Youth Justice Conference in 2006-07 may mean Youth Justice Conferences are becoming less effective, but it could also be a result of increased police surveillance or tougher policing.

The percentage of juveniles referred to a Youth Justice Conference for a justice procedure offence (e.g. breach of a court order) nearly doubled between 2004/5 and 2006/7. It is possible police are placing juveniles given non-custodial orders under closer scrutiny or taking a tougher line in relation any breach.

Commenting on the findings, the Director of the Bureau, Dr Don Weatherburn said that the Bureau would be looking at ways to improve the accuracy of GRAM while working with the Department of Juvenile Justice to more thoroughly investigate the effectiveness of Youth Justice Conferencing.

Further enquiries: Dr Don Weatherburn 9231-9190, 0419-494-408