Developing Non-Arbitrary Categories for Sexual Offender Risk Communication: Construct Validity and the Quantification of “Riskiness”.
Risk assessment is an essential role of professionals working with offender populations. Thorough evaluations typically involve extensive effort, including interviews with the offenders, file reviews, collateral contacts and the scoring of structured risk assessment instruments. The results of such assessments are often summarized in one of three words: low, medium or high risk. Although both evaluators and decision-makers like these nominal risk categories, they are essentially arbitrary, in that they do not share a common meaning across different settings and jurisdictions. An offender who is low risk in the context of conditional release decisions may be unacceptably high risk to run a daycare.
In this workshop participants will have the opportunity to explore and discuss current practice, and develop options for improving risk communication using non-arbitrary language. There are two broad approaches to defining risk categories. The first approach, favoured by proponents of structured professional judgement, defines risk in terms of the presence or absences of clinically-significant problems worthy of treatment and supervision resources. The credibility of statements comes from the research on the construct validity of the assessment measures. For example, a statement that an offender has paraphilic sexual interests requires a theoretical definition of paraphilia, as well as evidence of the reliability and validity of the specific assessment procedures used. Given the problem with self-report among sexual offender populations, this workshop will provide suggestions on how to use offence history and other externally observable behaviors to identify psychologically-meaningful risk factors.
The second approach to defining “riskiness” quantifies the prediction information contained in structured risk scales. In general, prediction tools contain two types of information: relative and absolute risk. Relative risk concerns the extent to which recidivists are different from non-recidivists. Relative risk can be quantified by percentile ranks and risk ratios. For example, an offender can be described as being in the top 10% in terms of risk of violence, or 4 times more likely to re-offend with a violent offence than are offenders in the middle of the risk distribution. Absolute risk concerns the recidivism rate of the group to which the offender most closely belongs (40% over 5 years).
In this workshop, participants will learn the general concepts needed for non-arbitrary risk communication, and specifically how to apply these concepts to risk assessments for sexual offenders based on Static-99R, Static-2002R, STABLE-2007 and ACUTE-2007.
R. Karl Hanson, Ph.D. is a Senior Research Scientist with Public Safety Canada and Adjunct Research Professor in the Psychology Department of Carleton University, Ottawa, Canada. Dr. Hanson is one of the leading researchers in the field of sexual offender risk assessment and treatment. Dr. Hanson has published more than 100 articles, including several significant meta-analyses, and was co-developer of the most widely used risk assessment tool for sexual offenders, including the Static- 99R and STABLE-2007. He is a Fellow of the Canadian Psychological Association, and was the 2002 recipient of the Significant Achievement Award from the Association for the Treatment of Sexual Abusers (ATSA).