Guidelines for offender risk assessment

Guidelines for offender risk assessment PDF Version (105KB)

Research summary
Vol. 7 No. 6
November 2003


How does the research literature offer to guide the selection and use of offender risk assessment instruments?


A major goal for correctional systems is to differentiate offenders with respect to their risk to re-offend. This goal not only has obvious importance for public safety but it is also important for the operation of efficient and humane correctional systems. Being able to differentiate higher risk offenders from lower risk offenders helps correctional agencies make decisions about applying controls and targeting supervision and treatment resources.

Until recently, most decisions regarding offender risk were based on subjective judgements and simple measures of risk. Research in the area of offender risk assessment has exploded over the past couple of decades and there have been significant developments in offender risk assessment. The present state of the research is now in a much better position to inform agencies and practitioners about "what works" in risk assessment.


A review of the literature on offender risk assessment was conducted. The review considered studies from around the world although most of the research was from Canada, a recognized leader in the field of offender assessment. The review yielded clear patterns in the results that were sufficiently robust to formulate guidelines for the selection and use of offender risk instruments.


The accumulation of research evidence suggests a number of broad guidelines for offender risk assessment. One of the most consistent findings is that evidence-based, actuarial measures are more accurate in the prediction of offender re-offending or recidivism than professional, clinical judgement. Unfortunately, the research also shows that many practitioners still rely on subjective, professional opinions and poorly validated assessment instruments.

Beyond the general guideline to use actuarial risk instruments that are based on sound research evidence, a number of other recommendations can be derived from the research. First, practitioners should ensure that the assessment instruments that they use actually have evidence supportive of their use and that they have been designed for use with criminal offenders. Many offender assessment instruments in use today were developed for other populations (e.g., psychiatric patients) and they generalize poorly to offender groups.

Second, many offender risk scales consist of historical items that provide little direction in terms of treatment programming. Some of the newer offender instruments assess needs that, when successfully addressed, result in a reduction of risk. These "risk-need" instruments can be extremely valuable for managing risk through offender treatment. Finally, improvements in predictive accuracy can be attained when many different measures are used rather than relying on a single instrument.

Policy implications

  1. The assessment of offender risk should include objective, actuarial measures. Continued reliance on subjective, professional or clinical judgement to gauge risk is no longer empirically defensible.
  2. Practitioners and correctional agencies need to recognize that not all assessment instruments are suitable for offender populations. Test instruments developed on non-offender populations should be reviewed as to their appropriateness.
  3. Offender assessment instruments that also assess criminogenic needs are especially useful because of their ability to direct effective treatment programming.


For further information

James Bonta, Ph.D.
Public Safety Canada
340 Laurier Avenue West
Ottawa, Ontario
K1A 0P8
Tel 613-991-2831
Fax 613-990-8295

Date modified: