Author: Mason, Ron
Title: Analysis of the Tasmania Police Risk Assessment Screening Tool (RAST)
Summary: Safe at Home involves a range of initiatives, expanded and new services that represent a significant change in the way the Tasmanian Government responds to family violence: family violence is treated as a crime rather than a private matter. Included in the Safe at Home Strategy was the implementation of a Risk Assessment Screening Tool (RAST) developed by Tasmania Police and the Department of Justice. The RAST is utilised by operational police at the attendance of family violence incidents to assist in assessing the risk of a victim experiencing future violence. In May 2008, Tasmania Police contracted the Tasmanian Institute of Law Enforcement Studies [TILES], of the University of Tasmania, to undertake a statistical analysis of the police Risk Assessment Screening Tool [RAST] to determine the validity and reliability of the RAST with respect to the Tasmanian population. The study concluded that, whilst the current scoring system employed has modest predictive utility (AUC .602), with an increased risk of misclassification in the medium and high categories, those characteristics identified through regression modelling provided a significantly greater level of accuracy (AUC .726). This reflects good predictive utility in that it is correct in predicting repeat offending in nearly 75% of cases (i.e. in approximately 3 out of 4 cases). It should be noted that this compares favourably with other risk assessment tools. With the primary purpose of the RAST being to predict the likelihood of re-offending, the Analysis found that the most useful risk factors on the RAST are those that reflect static variables. Static variables include age, gender, criminal history and education, and tend to be objective rather than subjective. Problems identified with the present RAST included structural issues whereby risk factors may have unintended deleterious and/or cumulative effects on the likelihood of re-offending. Consequent of these conclusions, it has been recommended that a new weighting system be considered with emphasis placed on those risk factors that have a cumulative effect on re-offending. The Analysis identified potential for improvements to the RAST that could increase its predictive utility from modest to good.
Details: Hobart, Tasmania: Tasmanian Institute of Law Enforcement Studies, 2009. 68p.
Source: Internet Resource
Year: 2009
Country: Australia
URL:
Shelf Number: 119555
Keywords:
Domestic Violence (Tasmania)Family ViolencePolicingReoffendingRisk Assessment |
