Controversies: Can we Predict Offender Risk?

About the controversies series: Controversies publications present the academic literature pertaining to an issue, and take one empirically-based side. These publications are intended for distribution and are extremely useful for academics, professionals or laypeople who are interested in issues in psychology and law. They provide a means to disseminate empirically based information in a way that is both quick and useful. Controversies publications undergo the EAPL-S peer review process and editing before publication.

About the author: This article was written by Danielle Andrewartha, EAPL-S representative for Australia and Ph.D. student at Monash University.

View this document in it's full glory by downloading the pdf here.



Remember to also read our fact sheet on forensic risk assessment!


Decisions concerning parole, sentencing and the like require careful predictions of risk. In relation to persons who have committed criminal offences, the relevant risks to the community relate to an offender’s likelihood of reoffending on release. The accurate prediction of such risk is, therefore, of significance to the applicable decision-maker as it will impact the length of sentence imposed, whether or not parole is granted and, if so, the provisions of parole, as well as the necessity for and conditions of any supervision orders. Such matters are particularly relevant in the case of repeat offenders and offenders who have committed serious sexual or violent offences as in these circumstances protection of the public from the threat of recidivism is paramount. [1] The risk of criminal recidivism is clinically assessed by a qualified psychologist or psychiatrist through informal interviews, clinical judgment, and/or consultations with other clinicians [2], and/or by the use of an actuarial risk prediction instrument(s) [3, 7]. Over the past decade or so, an abundance of objective risk assessment tools (i.e., actuarial risk prediction instruments) have been developed for the courts’ reliance in determining offender risk.

“[P]eople [may be deprived] of personal liberty, a most fundamental human right, on a prediction of dangerousness, based largely on the opinions of psychiatrists which can only be, at best, an educated or informed “guess”.”– The Hon Michael Derby, retired Australian high court judge

Evidence for

Concern over the accuracy of clinical opinion in gauging offender risk led to the development of these predictive instruments, which are regarded as providing greater validity and reliability than subjective clinical judgment [4, 5]. Proponents of actuarial instruments being used as means of assessing risk accurately argue that such tools “…locate an offender within a broad population of offenders and evaluate statistically the likelihood of their reoffending, based upon identified aspects of past behaviour, childhood and adolescent development, life events, current attitude, and personality traits… From the results of such studies, the predominant risk factors are isolated, coded, and weighted, then reproduced as an actuarial risk prediction instrument composed of these items.” [6] They argue this enables the clinician implementing the instrument to ascertain the relevant degree of risk posed by the offender, for example: low, medium-low, medium, medium-high, high.

Evidence against

Criticisms have been made of both clinical judgment and actuarial instruments for use in predicting offender risk. Unaided clinical opinions (those reached via interview, etc. without the use of an assessment tool) have been found to have high false-positive rates (a finding of risk where none exists) and false-negative rates (a finding of no risk where one exists). Further, the ability of mental health professionals to predict risk accurately has been questioned generally [7]. Actuarial risk prediction instruments also have their limitations. For instance, they can be rigid, and “dominated by assessment by static risk factors that are principally historical and give limited importance to dynamic risk factors such as accommodation, employment, education, and antisocial attitudes”[8].

What it all means

As demonstrated above, the courts rely on mental health professionals’ clinical judgment and/or actuarial instrument(s) to assist decision-makers in evaluating the risks posed by offenders, particularly in the case of repeat or serious offenders. The degree to which such predictions reflect accurately the true risk posed cannot be stated with absolute confidence, despite empirical studies attesting to the merit of the various actuarial risk prediction instruments [3, 8].

In conclusion, the short answer is, “Yes, we can predict offender risk”, but only with some degree of confidence given that the efficacy of such predictions is variable and uncertain. Thus, while clinical judgment and actuarial tools may be useful to the applicable decision-maker in ascertaining the likely risk posed by an offender, there is presently no definitive means of assessing an offender’s risk at any point in time. As a result, these aids are only one aspect of the evaluation of risk undertaken by the ultimate decision-maker as part of the broader assessment regarding sentencing, parole and the like.

The quote wall

“Every day many thousands of predictions are made by parole boards, deans' admission committees, psychiatric teams, and juries hearing civil and criminal cases. …these are high stakes indeed. To use the less efficient of two prediction procedures in dealing with such matters is not only unscientific and irrational, it is unethical. To say that the clinical-statistical issue is of little importance is preposterous.”

– William Grove and Paul Meehl (deceased), on the importance of developing sound statistical procedures for accurately predicting behaviour.

“Actuarial risk prediction scales may contribute important information to psychological and psychiatric risk assessments...”

- S. Smallbone, psychologist, & J. Ransley, lawyer and academic author.

“Actuarial risk assessment instruments cannot be used to estimate an individual's risk for future violence with any reasonable degree of certainty and should be used with great caution or not at all. The 95% confidence intervals were large for risk estimates at the group level; at the individual level, they were so high as to render risk estimates virtually meaningless."

– Dr. Stephen Hart and his colleagues discuss the problem of the large “margins of error” in risk assessment.

“[A]ctuarial tools are best used to support, rather than supplant, expert assessments of risk.”

– Eric Silver, an expert on assessing violent offending and violent victimization, and Lynette Chow-Martin, an actuarial scientist.

Where can I get more information?

Information from the EAPL-S is available online, in PDF, or as paper brochures sent through the mail. If you would like to have EAPL-S publications, you can order hardcopies for a nominal fee through This email address is being protected from spambots. You need JavaScript enabled to view it. .


  1. Kirby, M (ed.). (2012). The Laws of Australia. Australia, NSW: Thomson Law Book Company Limited.
  2. Hart, S. D. (1998). The role of psychopathy in assessing risk for violence: Conceptual and methodological issues. Legal And Criminological Psychology, 3(Part 1), 121-137. doi:10.1111/j.2044-8333.1998.tb00354.x
  3. Grove, W. M., & Meehl, P. E. (1996). Comparative efficiency of informal (subjective, impressionistic) and formal (mechanical, algorithmic) prediction procedures: The clinical–statistical controversy. Psychology, Public Policy, And Law, 2(2), 293-323. doi:10.1037/1076-8971.2.2.293
  4. Quinsey, V.L., Harris, G.T., Rice, M.E., & Cormier, C.A. (2006). Violent
    offenders: Appraising and managing risk (2nd ed.). Washington, DC: American
    Psychological Association.
  5. Davis, M., & Ogloff, J. (2008). Key considerations and problems in assessing risk for violence. In D. Canter, & R. Zukauskiene (eds). Psychology and law: Bridging the gap. Hampshire, England: Ashgate publishing ltd.
  6. Freckelton, I., & Selby, H. (2003). Expert evidence: Law, practice, procedure and advocacy. Sydney: LBC Information Service.
  7. Cocozza, J.J. , & Steadman, H. J. (1976): The failure of psychiatric predictions of dangerousness: Clear and convincing evidence. Rutgers Law Review, 29(5), 1084- 1101.
  8. Hilton, N., Harris, G. T., & Rice, M. E. (2006). Sixty-Six Years of Research on the Clinical Versus Actuarial Prediction of Violence. The Counseling Psychologist, 34(3), 400-409. doi:10.1177/0011000005285877

Controversies series information

EAPL-S publications are in the public domain and may be reproduced or copied without the permission from the European Association of Psychology and Law Student Society (EAPL-S). EAPL-S encourages you to reproduce them and use them in your efforts to improve awareness of issues in psychology, corrections and law. Citation of the European Association of Psychology and Law as a source is appreciated. However, using these materials inappropriately can raise legal or ethical concerns, so we ask you to use these guidelines:

  • EAPL-S does not endorse or recommend any commercial products, processes, or services, and publications may not be used for advertising or endorsement purposes.
  • EAPL-S does not provide specific medical advice or treatment recommendations, legal action or referrals; these materials may not be used in a manner that has the appearance of such information.
  • EAPL-S requests that organizations not alter publications in a way that will jeopardize the integrity and "brand" when using publications.
  • Addition of EAPL and EAPL-S logos and website links may not have the appearance of EAPL-S endorsement of any specific commercial products or services or medical treatments or legal services.
If you have questions regarding these guidelines and use of EAPL-S publications, please contact the EAPL-S at This email address is being protected from spambots. You need JavaScript enabled to view it.