Cross-cultural reliability and rater bias in forensic risk assessment: a review of the literature

Risk assessment instruments are frequently employed in criminal justice settings to predict recidivism and inform treatment plans for offenders. Modern approaches to risk assessment, (i.e. Structured Professional Judgment and actuarial instruments), have been found to accurately predict recidivism for both juvenile and adult offenders, although concerns have been raised regarding their applicability to Non-White cultural groups. Furthermore, research has suggested that rater bias may impact upon risk assessment scoring in various aspects of the risk assessment process. This paper explores whether rater bias may offer an explanation for the differential application and predictive validity of risk assessment for some cultural groups. Firstly, literature examining the differential scoring of risk assessment instruments across cultures was explored. Following this, existing literature on rater variability in risk assessment - encompassing areas such as interrater reliability, adversarial allegiance, professional override and rater training - was synthesised. Despite the lack of research specifically examining rater cultural bias in risk assessment, this paper finds that the risk assessment process may be vulnerable to rater biases. At the very least, the notion of rater cross-cultural bias in risk assessment warrants further investigation, and suggestions for future research are offered.