The applicability of risk assessment tools to diverse offender populations is a relevant consideration for all criminal justice organizations. The primary benefit of structured risk assessment is that it enables the use of accumulated knowledge on risk factors to be reliably applied to individuals with accuracy, in order to predict future offending and, in turn, inform sentencing, custody security ratings, community release, supervision conditions, and treatment needs. However, one of the assumptions of standard risk assessments is that they apply equally to all offenders, even though they have primarily been developed on largely White male offenders. Therefore, their use with offenders of varied ethnic/racial, cultural backgrounds, and genders has been highly criticized. Some of the main considerations that fuel the conversation around the generalizability of risk assessment to varied genders and cultures include (a) the potential influence of cultural or gender bias, (b) inappropriate or inadequate measurement of current risk factors, and (c) lack of culturally relevant or gender-relevant risk factors. This article discusses considerations related to the use of risk assessment tools with individuals from diverse cultures, including a case example, followed by a focus on how risk assessment tools are used with female offenders.
When assessments are applied to groups that differ significantly from the construction samples, they are criticized for making the assumption that the assessments, and the risk items comprising the assessment, are equally valid to the new group. This criticism has been raised largely by those questioning the validity of these tools with ethnic minority groups, predominantly those overrepresented in criminal justice systems, such as African American and Hispanic peoples in the United States and Indigenous peoples in Canada, Australia, and New Zealand. One of the main concerns with using risk assessments developed on one group with different and diverse populations is the potential for cultural bias. This is particularly salient to issues pertaining to cross-cultural validity as well as minority groups who are overrepresented in many criminal justice systems around the world.
More specifically, the argument around cultural bias contends that generic risk tools ignore the unique historical and present context of many ethnic/racial groups that may be relevant to their offending trajectories. This is particularly relevant for ethnic/racial groups who have experienced the oppression of colonization, the vestiges of which in many cases persist today, through the social, political, and economic marginalization of these groups (see Case Example later in this article). It is argued that the use of these assessments, founded on the cultural values and experiences largely representative of Western society, not only improperly capture risk, they also serve to further disadvantage the cultural minority (through consistently higher scores on risk tools) when used for legal and criminal justice decision-making (e.g., security classification, program allocation, release back into the community).
Similarly, given the potential differences in the underlying values across cultural groups, the adequacy of the risk factors being measured within these tools has been called into question. Additionally, the lack of culturally relevant and culturally specific risk factors has been identified by some scholars and practitioners as a shortcoming of risk assessment, as it is applied to such varied racial populations. For example, having a history of emotional problems and victimization has been shown in some research to be predictive of offending for Indigenous populations but not for non-Indigenous offenders. It could be that this factor is more relevant to the offending patterns of Indigenous peoples given their differential history of victimization; in other words, it is more relevant to one cultural group than the other. Similarly, culturally specific factors are those that may be related to offending (e.g., loss of Indigenous language) exclusively for one group. There is a dearth, however, of empirical research in the area of cross-cultural validity and exploring cultural differences in these risk factors.
Case Example: Risk Assessment With Indigenous Offenders
Within the larger debate concerning risk assessment with minority offenders, many have specifically questioned the applicability of these tools with Indigenous offenders, given their overrepresentation in the criminal justice systems of Canada, Australia, and New Zealand. For example, in Canada, in 2016, Indigenous peoples represented one quarter of federal inmates, while only accounting for approximately 4% of the general population. One of the main arguments explaining their overrepresentation is the differential history and experience of Indigenous peoples compared to non-Indigenous people. Namely, it is argued that the history of colonization, isolation, and discrimination has resulted in Indigenous peoples disproportionately coming into contact with the justice system. Therefore, in order to better understand criminal behavior among this population, this cultural history must be considered in any assessment of risk.
In Canada, risk assessment instruments developed on non-Indigenous male offenders are, for the most part, also administered to Indigenous offenders. Meta-analytic reviews on the predictive validity of common risk factors (i.e., the Central Eight risk/need factors, as defined by Andrews and Bonta in 2010) and assessment tools (e.g., the Level of Service Inventory Scales) have shown, with consistency, that risk factors and assessments generally predict recidivism for Indigenous offenders (as is similarly found with other racial minority groups); however, they do so with less predictive accuracy than with White offenders. Given the unique history of Indigenous peoples, it has been suggested that culturally relevant and culturally specific factors may be important to better understand criminal behavior among this population.
Gender
The steady increase in female involvement in the criminal justice system, and a lack of similar growth in male offending, has prompted considerable interest in the assessment and management of female offenders. Given the emphasis on evidence-based practice, this had led to increasing questions regarding the applicability of standard risk assessment tools, largely developed on male offenders, to female offenders. The criticisms associated with applying existing assessment tools to female offenders are premised on evidence indicating that female offenders differ significantly from male offenders in both offending patterns and personal/ social histories. Compared with males, female offenders are most often arrested for alcohol- or drug-related offenses, minor property crimes, or public order offenses. They typically experience increased rates of social marginalization and poverty, physical and sexual abuse as children and adults, substance use problems, mental illness, and are more likely to be primary caregivers for children compared to male offenders. It has also been suggested that low self-esteem and self-efficacy, problematic relationships, and the experience of trauma are salient factors in understanding female criminality. Although male offenders also experience these issues, research has shown that these factors affect female offenders with greater frequency. Furthermore, it has been suggested that these factors uniquely impact pathways to crime for females and, in turn, have been referred to as gender-responsive factors. In contrast, risk factors typically referred to as gender-neutral are risk factors that are argued to apply equally to all offenders, such as the Central Eight risk/need factors.
The most commonly used risk assessment tools consist of gender-neutral factors and therefore, consistent with general theories of criminal behavior, they include the same items and weighting for all offenders; this implies that all risk/need items have the same relevance for both genders. When examining the extant literature, the predictive accuracy of existing, gender-neutral risk assessment tools, particularly those tools focused on dynamic factors, has been mixed when tested with female offenders. Beyond explanations associated with the tools themselves, these findings may be partially attributable to a lack of sufficient research to provide stable results. Overall, studies focusing solely on risk assessment with female offenders are few, though growing, and those studies that include female offenders within a larger sample do not consistently separate effect sizes by gender. Furthermore, even less is known about female offenders of diverse ethnic/racial backgrounds. One exception appears to be the Level of Service Inventory instruments, as they have been the most tested risk assessment tools with female offenders and results generally suggest acceptable predictive ability, although questions about overclassification remain.
Feminist scholars have argued that even if gender-neutral factors (e.g., employment, relationship problems) are predictive for female offenders, they are experienced differently by female offenders and, therefore, differentially affect the offending patterns of this group. Burgeoning research has also suggested that using standard gender-neutral risk assessment tools that do not include or emphasize factors more relevant for female offenders produces inaccurate risk estimates associated with the overclassification of female offenders. In turn, as the role of risk assessment has extended beyond simply risk prediction to also informing treatment, it is suggested that these standard tools do not identify other, potentially more relevant treatment targets for female offenders.
The practice of simply validating the existing tools on female offenders does not explore the arguably differential mechanisms underlying the relationship between gender-neutral risk factors and female offending. It also does not assist in determining the existence of more predictive and appropriate risk/need factors, and therefore, treatment targets, for this offender group. These concerns with gender-neutral tools coupled with qualitative and mixed-methods research suggesting the importance of gender-responsive factors have led feminist scholars to call for gender-informed risk assessment tools. Evidence to support this view is growing, as some studies have found that gender-responsive risk factors (e.g., parental stress, adult victimization, childhood sexual abuse) have added incremental predictive validity to existing gender-neutral risk tools. In turn, researchers have begun to develop risk assessment tools specifically for female offenders, typically including both gender-neutral and gender-responsive risk factors.
References:
- Andrews, D. A., & Bonta, J. (2010). The psychology of criminal conduct (5th ed.). United States: Anderson. Cincinnati, OH, US: Anderson Publishing Co.
- Andrews, D. A., Guzzo, L., Raynor, P., Rowe, R. C., Rettinger, L. J., Brews, A., & Wormith, J. S. (2012). Are the major risk/need factors predictive of both female and male reoffending? A test with the eight domains of the Level of Service/Case Management Inventory. International Journal of Offender Therapy and Comparative Criminology, 56, 113–133.
- Blanchette, K., & Brown, S. L. (2006). The assessment and treatment of women offenders: An integrative perspective. New York, NY: Wiley.
- Gutierrez, L., Wilson, H. A., Rugge, T., & Bonta, J. (2013). The prediction of recidivism with Aboriginal offenders: A theoretically-informed meta-analysis. Canadian Journal of Criminology and Criminal Justice, 55, 55–99. doi:10.3138/cjccj.2011.E.51
- Hannah-Moffat, K. (2009). Gridlock or mutability: Reconsidering “gender” and risk assessment. Criminology & Public Policy, 8, 209–219.
- Martel, J., Brassard, R., & Jaccoud, M. (2011). When two worlds collide: Aboriginal risk management in Canadian corrections. British Journal of Criminology, 51, 235–255. doi:10.1093/bjc/azr003
- Public Safety Canada. (2017). Corrections and conditional release statistical overview: Annual report 2016. Ottawa, Canada: Author.
- Shepherd, S. M. (2015). Finding color in conformity: A commentary on culturally specific risk factors for violence in Australia. International Journal of Offender Therapy and Comparative Criminology, 59, 1297–1307. doi: 10.1177/0306624X14540492