Assessing The Risk Of Domestic Violence Offenders

Assessing The Risk Of Domestic Violence Offenders PDF Version (720 KB)

Research summary
Vol. 16 No. 3
May 2011


How can we assess risk for reoffending among domestic violence offenders?


Domestic violence is an offence that causes considerable public concern. Additionally, domestic violence offenders constitute a meaningful proportion in prison and on community supervision (e.g., probation). Assessing their risk for reoffending has important implications for public safety and for the offender.

There has been considerable research over the past 20 years on which factors predict reoffending for domestic violence offenders (e.g., substance abuse). There has also been several risk instruments developed specifically to predict domestic violence reoffending. Risk assessment tools can be helpful at many stages in the criminal justice system, such as decisions about whether to grant release into the community (e.g., bail or parole), levels and conditions of supervision (e.g., how closely the offender is monitored, or whether to prohibit contact with the victim), and even length or type of sentence.


To examine domestic violence risk assessment, we reviewed the most researched risk instrument for domestic violence: The Spousal Assault Risk Assessment guide (SARA). The SARA was developed in Canada and is the most common risk tool for domestic violence offenders used in at least 15 countries and translated into 10 languages. The SARA has 20 items that are related to domestic violence offending. Based on the offender's score on these items, the risk evaluator makes an overall judgement about whether the offender poses a low, moderate, or high risk to reoffend violently towards their partner, as well as towards others.


Although it could benefit from further research and development, the SARA is a useful risk scale for domestic violence offenders. The 20 items considered in the SARA are generally supported by research, although some items have stronger support than others. It has a helpful coding manual to guide evaluators on the information they need, how to score the items, and how to interpret the SARA appropriately.

There have been 10 studies testing the SARA, mostly from Canada, but also from the United States, Spain, and Sweden. These studies suggest that the SARA is moderately accurate at predicting which offenders are more likely to reoffend with a domestic violence offence. The SARA appears to predict reoffending in diverse samples, such as offenders released from prison or on community supervision, as well as mentally disordered offenders. Many of these studies included some non-Caucasian offenders, but did not specifically examine how well the SARA works with minority offenders. It has not been tested with female offenders or juvenile offenders.

The SARA predicts reoffending similarly to most of the other tools specifically designed for domestic violence offenders. One advantage of the SARA, however, is that it has more research support than other scales, so evaluators can use it knowing that it has been validated with diverse groups of offenders. Although domestic violence risk scales seem to do a similar job at predicting recidivism, there are other risk scales designed for violent offenders (e.g., the Violence Risk Appraisal Guide) or general offenders (e.g., the Level of Service instruments) that may predict recidivism slightly better than the domestic violence scales. This suggests that optimal risk assessment should consider what makes domestic violence offenders similar to other offenders as well as what makes them a unique group.

Another important finding is that any risk assessment (using the SARA or any other instrument) that could negatively impact an offender should be completed by a properly trained evaluator who has access to relatively comprehensive information and is aware of the strengths and limitations of the risk tool being used.

Policy implications

1. The SARA is a defensible method for evaluating the risk of domestic violence offenders. It can help inform correctional decisions (e.g., bail, sentencing, treatment, parole, community supervision) in a way that can enhance the protection of the public while also applying the least restrictive sanction for the offender.

2. Although the SARA has more research support than other domestic violence risk tools, it has not been tested with female domestic violence offenders, or with Aboriginal offenders. This means that practitioners in the criminal justice system should exercise caution when evaluating the risk of these groups.

3. Although it is useful to have risk assessment tools like the SARA specifically designed for domestic violence offenders, these tools should complement risk assessment methods used for general offenders, rather than replace them.


For further information

Guy Bourgon, Ph.D.
Corrections Research
Public Safety Canada
340 Laurier Avenue West
Ottawa, Ontario
K1A 0P8
Tel (613) 991-2033
Fax (613) 990-8295

Date modified: