Investigating racial disparities in violence risk assessment using the Spousal Assault Risk Assessment Guide-Version 3 (SARA-V3): Structured professional judgment ratings and recidivism among Indigenous and non-Indigenous individuals.

Racial disparities in criminal justice outcomes are widely observed. In Canada, such disparities are particularly evident between Indigenous and non-Indigenous persons. The role of formal risk assessment in contributing to such disparities remains a topic of interest to many, but critical analysis has almost exclusively focused on actuarial or statistical risk measures. Recent research suggests that ratings from other common tools, based on the structured professional judgment model, can also demonstrate racial disparities. This study examined risk assessments produced using a widely used structured professional judgment tool, the Spousal Assault Risk Assessment Guide-Version 3, among a sample of 190 individuals with histories of intimate partner violence. We examined the relationships among race, risk factors, summary risk ratings, and recidivism while also investigating whether participants' racial identity influenced the likelihood of incurring formal sanctions for reported violence. Spousal Assault Risk Assessment Guide-Version 3 risk factor totals and summary risk ratings were associated with new violent charges. Indigenous individuals were assessed as demonstrating more risk factors and were more likely to be rated as high risk, even after controlling for summed risk factor totals and prior convictions. They were also more likely to recidivate and to have a history of at least one reported act of violence that did not result in formal sanctions. The results suggest that structured professional judgment guidelines can produce disparate results across racial groups. The disparities observed may reflect genuine differences in the likelihood of recidivism, driven by psychologically meaningful risk factors which have origins in deep-rooted systemic and contextual factors. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

University Student Attitudes Toward Persons With a History of Adult Sexual Violence and Their Correlates With Policy, Treatment, and Management.

Empirical evidence suggests that postsecondary students are disproportionately impacted by sexual violence. Further evidence suggests that most persons convicted of crimes involving sexual violence return to the community, and that social factors, including public policy and community members' attitudes and perceptions, are key determinants of successful reintegration among these persons. Taken together, these observations suggest that students' perspectives should be considered in discourse related to reintegration. The current study aimed to assess the attitudes of a university undergraduate sample toward men convicted of adult sexual violence as well as their level of support for various community-based policies to manage this population upon release from custody. Participants (N = 333) completed a survey battery comprising measures of three dimensions of attitudes toward persons who have sexually offended, perceptions of recidivism and treatment response, and support for various incapacitation/control (IC) and rehabilitative/reintegration (RR)-based policies. Results varied depending on the dimension of attitudes measured, with the affective component appearing to be the most markedly negative. Participants endorsed a mixture of IC and RR policies, with attitudinal measures predicting policy endorsement controlling for relevant demographic variables. The results provide a framework for future research surveying a more representative sample of the Canadian public, while also providing useful information for policymakers relying on community support to reduce sexual offending.

On the Ethics and Practicalities of Artificial Intelligence, Risk Assessment, and Race.

Artificial intelligence (AI) has been put forth as a potential means of improving and expediting violence risk assessment in forensic psychiatry. Furthermore, it has been proffered as a means of mitigating bias by replacing subjective human judgements with unadulterated data-driven predictions. A recent ethics analysis of AI-informed violence risk assessment enumerated some potential benefits, ethics concerns, and recommendations for further discussion. The current review builds on this previous work by highlighting additional important practical and ethics considerations. These include extant technology for violence risk assessment, paradigmatic concerns with the application of AI to risk assessment and management, and empirical evidence of racial bias in the criminal justice system. Emphasis is given to problems of informed consent, maleficence (e.g., the known iatrogenic effects of overly punitive sanctions), and justice (particularly racial justice). AI appears well suited to certain medical applications, such as the interpretation of diagnostic images, and may well surpass human judgement in accuracy or efficiency with respect to some important tasks. Caution is necessary, however, when applying AI to processes like violence risk assessment that do not conform clearly to simple classification paradigms.

An Evaluation of the Reliability, Construct Validity, and Factor Structure of the Static-2002R.

The fundamental psychometric properties of the subscales found in the Static-2002R, an actuarial measure of sexual recidivism risk, were evaluated in the current study. Namely, the reliability, concurrent and construct validity, and factor structure of the Static-2002R subscales were examined with a sample of 372 adult male sex offenders. In addition to using validated measures of sexual violence risk to examine concurrent validity, construct-related measures taken from extant risk measures and psychometric tests were correlated with three of the subscales to assess overall construct validity. Moderate support was found for the reliability of the Static-2002R. The concurrent and construct validity of the General Criminality, Persistence of Sexual Offending, and Deviant Sexual Interest subscales were supported. Generally, these findings further support the Static-2002R as a valid sex offender risk appraisal instrument that encompasses multiple distinct, clinically relevant, risk domains.