Progress in violence risk assessment

Research summary
Vol. 10 No. 3
May 2005


What have we learned about evaluating the risk of future violence?


Assessing the risk of violence has always been one of the central tasks for those involved in the management of offenders in the criminal justice system. It is not an easy task. The scientific literature in the 1980s suggested that evaluators had limited ability to predict future violence, and that forensic experts were no better than lay people at this task. At the time, policies that treated "high risk" offenders different from "low risk" offenders were questioned because the decision-makers had difficulty telling these groups apart. The need for accurate risk assessments, however, has intensified in recent decades as social policy has increasingly emphasized community safety as the fundamental goal of intervention with offenders. Consequently, it is important to know the extent to which scientific evidence has advanced our ability to predict who will and who will not commit new violent offences.


The scientific literature on violence risk assessment was examined, focusing on major reviews conducted between 1980 and 2004. Also considered were the risk assessment methods actually used by correctional professionals and forensic mental health evaluators during this same time period.


During the past 20 years, there have been considerable advances in the ability to predict violent recidivism. The professional journals are now filled with articles in which violent re-offending is predicted with moderate to high accuracy. Major reviews have identified the individual characteristics most closely associated with violent offending (e.g., antisocial lifestyle, substance abuse, history of rule violation). Rather than debating whether violence can be predicted, researchers are now debating the best methods for risk assessment.

One of the major developments during the past 20 years has been the introduction of structured risk assessment tools. These tools come in two main types: a) structured professional guidelines, which only specify the relevant risk factors to consider, and b) actuarial measures, which specify both the items to consider as well as how to combine these items into an overall evaluation. These new approaches to risk assessment are far from perfect, but everyone agrees they are more accurate than the unguided professional judgement used to assess the risk for violence in the 1980s. This advance in research knowledge has also resulted in substantial changes in the methods used in applied risk assessments. Structured risk assessment tools are now commonly used in the forensic and correctional systems for decisions concerning sentencing, security placement, programming, and conditional release.

Policy implications

  1. Not all offenders are equally likely to commit a new violent offence. The goals of community protection can be effectively promoted when more resources and services are directed toward higher risk offenders than toward lower risk offenders.
  2. Decision-makers concerned with the risk of violence should play close attention to recent scientific developments. In particular, evaluators should include validated, structured assessment tools as part of their overall assessment of risk for violence.
  3. Ongoing training efforts are needed to ensure that evaluators and decision-makers have access to recent research findings.


For further information

R. Karl Hanson, Ph.D.
Corrections Research
Public Safety Canada
340 Laurier Avenue West
Ottawa, Ontario K1A 0P8
Tel: 613-991-2840
Fax: 613-990-8295

Date modified: