Why we cannot explain cross-cultural differences in risk assessment

• Published in: Aggression and violent behavior Vol. 50; pp. 101346 - 12
• Main Authors: Schmidt, Stefanie, Heffernan, Roxanne, Ward, Tony
• Format: Journal Article
• Language: English
• Published: Tarrytown Elsevier Ltd 01.01.2020; Elsevier Science Ltd
• Subjects: Bias; Confusion; Crime; Cultural differences; Culture; Dynamic Risk Factors; Ethnicity; Explanation of Crime; Level of Service; Measurement; Predictive validity; Risk assessment; Risk factors; Risk Prediction; Level of Service; Ethnicity; Dynamic Risk Factors; Risk Prediction; Bias; Explanation of Crime
• ISSN: 1359-1789; 1873-6335
• DOI: 10.1016/j.avb.2019.101346

The prediction and explanation of crime currently relies predominantly upon the concept of dynamic risk factors (DRF). Evidence suggests that DRF vary across cultures with respect to their prevalence and their predictive validity (e.g., Olver, Stockdale, & Wormith, 2014). What remains unclear is whether the observed differences can be explained by real cultural differences in the causes of crime; and if this is the case, how does culture influence this? We suggest that confusion arises due to conceptual problems with DRF and their measurement. Because DRF are vague, composite, and value-laden constructs researchers are unable to minimize or control for the occurrence of construct or item bias when scores on risk measures are compared across cultures. This makes any further interpretation and adjustment to assessment or intervention unwarranted. If we do not know whether or how DRF cause violence within the cultural context of their development, we cannot possibly hope to explain this relationship in another culture. We suggest that there is a pressing need to provide coherent theories for research, risk assessment and treatment in the future. We will discuss possible ways forward in this paper.