Show Summary Details

Page of

 PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, CRIMINOLOGY AND CRIMINAL JUSTICE (criminology.oxfordre.com). (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 20 September 2018

A Review of the Validity of Juvenile Risk Assessment Across Race/Ethnicity

Summary and Keywords

Juvenile risk assessment instruments have provided juvenile courts with the opportunity to make standardized decisions concerning sentences and intervention needs. Risk assessments have replaced the reliance on professional decision-making practices in which court officials relied on their hunches or previous experience to determine what to do with youth once they became involved in corrections. A primary goal of juvenile risk assessment is to improve case management and help courts focus resources on juveniles who exhibit the greatest intervention needs. Further, juvenile risk assessments play a critical role in estimating which juveniles will likely reoffend by identifying factors that increase the propensity of future offending. Although some researchers believe that the implementation of standardized juvenile risk assessments is a good strategy for reducing biased decision-making for racial/ethnic minorities, other researchers have called into question the extent to which risk assessments overestimate risk for certain juveniles, especially those in minority groups who have a history of being marginalized due to their race, culture, or ethnicity. This article provides an overview of how well juvenile risk assessment instruments predict future delinquency across race and ethnicity. The review suggests that in general, risk assessments do a good job in predicting recidivism across racial/ethnic groups for diverse populations inside and outside the United States. However, there is still some room for improvement concerning the assessment of risk and needs for ethnic minorities. In addition, while there are some studies that do not report the predictive validity of risk assessment scores across race/ethnicity, risk assessments overall seem to be a promising effort to correctly classify and/or identify juveniles who are at greatest risk for future recidivism.

Keywords: juvenile risk assessment, predictive validity, race, risk, needs, systematic review, ethnicity, classification

Since the establishment of the first juvenile court in 1899, class and racial disparities in the juvenile justice system have been a continual concern (Leiber, Bishop, & Chamlin, 2011; Platt, 1969). For instance, the Child Savers Movement, which was a movement in the United States that largely influenced the development of the juvenile court, has been criticized as being based on social control rather than rehabilitative efforts (Fox, 1996; Platt, 1969). More specifically, some scholars have raised concerns regarding the extent to which the development of the juvenile justice system was based on an act of benevolence for children labeled as needing services as opposed to an attempt to discipline, impose personal/moral values, and regulate the behavior of individuals (Fox, 1996; Platt, 1969; Ward, 2009). In fact, Ward (2009) suggests that the juvenile court is an institution of injustice and inequity that has engaged in mistreatment of racial minorities for generations. To date, these debates continue and call into question the fairness of court practices for all racial groups.

In addition to race, there are concerns about the extent to which the juvenile court was designed to systematically target and control historically marginalized groups perceived as being a threat to society. Some researchers, practitioners, and social activists believe that parentless children, socioeconomically disadvantaged youth, and juveniles from urban ethnic minority communities are among the most targeted subgroups within society. As a result, courts have administered laws and created policies that have widened the net of care, which has consequently involved more juveniles in the court system and labeled more juveniles as being in need of services (Fox, 1996; Platt, 1969). This concern of net-widening dates back to the start of the juvenile court. For instance, there has been a long-standing history of disproportionate contact of specific groups involved in the juvenile court system (e.g., ethnic minorities). Some believe that it is due to oversurveillance (e.g., increased policing) and policies (e.g., zero-tolerance school policies) that increase the likelihood that ethnic minority youth would come in contact with the court. This early and disproportionate response to certain juvenile subgroups has raised many questions regarding the fairness of policies and practices at all stages of the juvenile justice process. For this reason, court officials are interested in determining whether practices like the use of juvenile risk assessments exacerbate or perpetuate racial and class disparities within the juvenile justice system.

Attribution and the Assessment of Risk

One hypothesis used to support the adoption of juvenile risk assessment tools stems from attribution theory, which suggests that individuals rely on preconceived assumptions to make decisions (Albonetti, 1991) and characterize others based on personal feelings and belief systems, which are influenced by a host of sociological mechanisms (Leiber, Bishop, & Chamlin, 2011; Rodriguez, 2013). These sociological influences (e.g., stereotypes) can have both a positive and a negative impact on how individuals are treated. As a strategy to minimize the influence of factors like negative stereotypes, juvenile and adult courts use risk assessment tools to help inform their practice concerning what to do with individuals who are involved in the justice system.

Attribution bias can occur in two directions. First, an offender who is high risk can be seen as low risk and thus not receive intervention services. Second, an offender who is low risk can be seen as high risk and thus receive unnecessary intervention services. Given the potential of attribution error, effecting marginalized groups (e.g., ethnic minorities in the United States), courts have moved toward standardized risk assessments to guide decisions concerning appropriate treatment of juvenile offenders. Even with the use of risk assessments, there remains some controversy around fairness, although research suggests that these concerns are unwarranted (Skeem & Lowenkamp, 2016). More research is necessary to validate risk assessment across racial groups and better understand if risk assessments promote equity and fairness. In the case of a juvenile offender who was arrested or received a petition due to some allegations of a delinquent act, a juvenile court must determine if the juvenile is at risk for future recidivism and how dangerous he or she is to society. Andrews, Bonta, and Hoge (1990) recommend that the administration of a risk assessment occur before a juvenile has a formal hearing with a judge or a court official. For many juvenile courts, trained court officers, probation officers, intake officers, social workers, or research professionals are responsible for administering a risk assessment interview. Once an interview is completed by a trained professional, measures are completed to determine if a youth has a low, moderate, or high risk of future recidivism. Assessing a juvenile’s risk provides the court with information concerning intervention needs and also informs a judge if a dismissal or diversion program is most suitable for a juvenile.

In contrast, without the use of juvenile risk assessment instruments, practitioners must use professional (i.e., clinical) judgment to make decisions and predict future behavior (Onifade, Petersen, Bynum, & Davidson, 2011). This is very complicated in the case of juvenile offenders because most juveniles will desist from committing future criminal acts or age out of crime. Further, research suggests that the use of professional judgments or hunches in place of standardized risk assessments can lead to a biased or overestimated measure of risk for some juveniles, especially those from disadvantaged backgrounds (Onifade et al., 2011). Given that racial and ethnic minorities are more likely to be perceived as high risk due to negative stereotypes, the adoption of formal tools like juvenile risk assessments is suggested to be one strategy to reduce attribution error (Leiber, Bishop, & Chamlin, 2011; Rodriguez, 2013).

The Use of Risk Assessments

Juvenile risk assessment instruments provide court officials (i.e., court referees, judges, and probation/juvenile court officers) with a standardized approach to making informed decisions concerning a juvenile offender’s risk for future offending and level of intervention needs. As mentioned previously, prior to the implementation of juvenile risk assessment instruments, decision-makers used subjective approaches that included relying on hunches, professional experience, or clinical judgments when determining a juvenile’s disposition or outcome. Given the level of subjectivity and unconscious biases (e.g., gender, race, and socioeconomic status) that may influence judgment, court practitioners replaced traditional subjective approaches to decision-making with modern risk assessment tools that focused on specific criminogenic risk factors that predict recidivism. While risk assessments have evolved over the years, there remains a major philosophical question. Given that most juvenile risk assessment tools were developed and normed for predominately White and male populations, it is essential to understand the extent to which these tools are valid across diverse populations. Since the generalizability of risk assessments across race is still unclear. This review provides some insight into how risk assessment models perform for individuals inside and outside the United States.

The RNR Model

The Risk Need Responsivity (RNR) model is the dominant theoretical framework used in the development of offender risk assessment instruments to predict recidivism (Lipsey, Howell, Kelly, Chapman, & Carver, 2010; Onifade, Nyandoro, Davidson, & Campbell, 2010; Schwalbe, 2007; Polaschek, 2012). The RNR model involves three major components: the identification of criminogenic risk, specific intervention needs, and aspects of responsivity (i.e., barriers to treatment). Together, these three components have been used to shape offender assessment and treatment models in an attempt to reduce the likelihood of recidivism. Overall, the RNR has guided the development of many risk assessment tools in the field of juvenile justice.

Risk Principle

Criminogenic risk is assessed in order to determine the risk for recidivism, as well as the level of services needed by a particular youth (Barnes et al., 2016; Gatti, Tremblay, & Vitaro, 2009; Polaschek, 2012). There are two types of criminogenic risk factors found in risk assessment tools: static factors and dynamic factors. Static factors are aspects of the individual’s experience that are not malleable in a treatment program, such as history of antisocial behavior or criminal involvement (Abracen et al., 2004; Ryan, Williams, & Courtney, 2013). Dynamic factors are criminogenic risk factors that have the potential to change. Dynamic factors commonly found in risk assessment tools include involvement in antisocial leisure, lack of recreational activities, involvement with delinquent peers and acquaintances, substance use, poor parental relationships/supervision, and problems in school. Together, static and dynamic criminogenic risk factors estimate if a juvenile has a low, moderate, or high risk of recidivism (McCafferty, 2016). Offenders who are at low risk generally have about a 10% chance of recidivism, moderate-risk offenders have about a 30% chance, and high-risk offenders have a 50%–60% chance (Latessa & Lovins, 2010; Lovins & Latessa, 2013). The primary goal of identifying criminogenic risk in offenders is to place those with higher risk into more intensive programs.

It has been well documented in the literature that placing low-risk offenders in intensive treatment actually has the potential to increase the likelihood of recidivism (Gatti et al., 2009; Latessa & Lovins, 2010; Lowenkamp & Latessa, 2005). One issue is that placing low-risk offenders in programs with moderate- and high-risk offenders puts low-risk offenders in contact with delinquent peers and exposes them to antisocial behaviors (Gatti et al., 2009; Latessa & Lovins, 2010; Lowenkamp & Latessa, 2005; Shapiro, Smith, Malone, & Collaro, 2010). Another issue is that overprogramming low-risk juveniles versus the use of diversion or dismissal can lead to negative labels. These stigmatizing labels ultimately lead to self-fulfilling prophecies, resulting in more delinquency. It is the risk principle that forecasts criminal behavior and allows courts to focus on juveniles who are most at risk for future offending.

Need Principle

The need principle involves assessing and targeting a juvenile offender’s needs in order to reduce the likelihood of future involvement with the juvenile justice system. Once a juvenile’s risk is determined, the need principle is critical to determining treatment delivery. According to the RNR model and the current best practice literature on offender rehabilitation, offenders should be given treatment that targets their specific dynamic needs, as identified by risk assessment tools (Andrews, Bonta, & Wormith, 2006; Polaschek, 2012). For example, an offender identified as high risk in the area of peer relationships might benefit from a program targeting prosocial peer acquisition. The primary goal of the need principle is to match services with risk. The need principle provides an opportunity to assist practitioners with providing individualized treatment (Andrews et al., 2006; Henggeler, Belton, & Smith, 1992; Polaschek, 2012). Some researchers recommend that treatment should be multisystemic in nature, as well as targeting multiple risk factors and tailoring services based on unique needs of the juvenile offender (Huey, Henggeler, Brondino, & Pickrel, 2000; Latessa & Lovins, 2010). On the other hand, programs that target only one or two needs, to the exclusion of others that are present in the offender’s situation, are sometimes less effective (Latessa & Lovins, 2010).

Responsivity Principle

The final principle in the RNR model is the responsivity principle, which informs how rehabilitative efforts, responses, and treatment should be provided. Assumption of the principle suggests that individuals respond differently to services, that there is a need to provide an appropriate level of intervention/treatment dosage, and that there is a need to identify and ameliorate potential barriers that will interfere with the intervention or cause potential harm to the juvenile. Ultimately, the responsivity principle in the RNR model is an attempt to determine the most efficacious treatment modality for the offender by identifying barriers to the treatment process.

There are two types of responsivity recognized in the RNR framework: general responsivity and specific responsivity. General responsivity refers to the use of cognitive-behavioral and social learning practices (Andrews & Bonta, 2010; Taylor, 2016). Specific responsivity consists of strategizing with knowledge of the offender’s characteristics (Andrews & Bonta, 2010; Lovins & Latessa, 2013; Taylor, 2016). One example of specific responsivity is the language barrier (Latessa & Lovins, 2010). Offenders must be able to speak the same language and comprehend the instructions given to them by the treatment provider. Another example is intellectual capability (Latessa & Lovins, 2010). Some treatments require introspection and articulate discussion of thoughts, which might not be appropriate for someone with a diminished capability for expressing himself or herself verbally. For these offenders, the use of purely behavioral interventions would likely be more effective. As the RNR model demonstrates, risk assessments are necessary in order to inform practitioners of the level of treatment needed, the type of treatment that will have positive effects, and the aspects of the treatment that should be focused on to best support offenders and rehabilitation efforts.

History and Evolution of Offender Risk Assessment

Current risk assessments, which are based on the RNR framework, have played a critical role in helping court practitioners and judges make decisions about appropriate responses to juvenile offenders. As juvenile risk assessments have evolved over the years, their efficacy as a tool for treatment has also increased (Schwalbe, 2007). To date, risk assessments have gone through four generations of changes.

First-generation risk assessments relied on professional judgment—known as clinical assessment—to estimate an offender’s risk for recidivism (Schwalbe, 2007; Vaswani & Merone, 2014). Generally, clinical judgment has been demonstrated to be less effective than approaches that take advantage of statistical data (Bengston & Langstrom, 2007; Lovins & Latessa, 2013; McCafferty, 2016). Second-generation risk assessments were grounded in statistical evidence—known as actuarial approaches—and focused on static factors such as criminal history (i.e., a history of having contact or repeat contact with the court; Schwalbe, 2007; Vaswani & Merone, 2014). Focusing on static factors did not benefit the process of rehabilitation because static factors cannot be changed and do not tell practitioners where to target interventions. Third-generation risk assessments incorporated dynamic risk factors (Schwalbe, 2007; Vaswani & Merone, 2014). Investigating dynamic risk factors provides specific information to practitioners, such as current drug use, family environment, the leisure activities of an offender, and family circumstances, on which they can base treatment. Fourth-generation assessments include case management and provide information about gender and other potential barriers to treatment in order to better inform practitioners (Lovins & Latessa, 2013; Schwalbe, 2007; Vaswani & Merone, 2014).

Although studies have demonstrated that juvenile risk assessments are promising and critical to the decision-making process (Schwalbe, Fraser, Day, & Cooley, 2006; Stams, 2015), more research is needed to determine the extent to which juvenile risk assessments are valid across specific subgroups—especially race and ethnicity. This issue of predictive validity is important because it determines how well a measure is performing and identifying which juveniles are at greatest risk for recidivism. Many court decisions are made based on being classified as low, moderate, or high risk according to the classification tool. For this reason, it is imperative that these instruments are fair and do not disproportionately put certain groups in high- or low-risk categories.

Criteria for Good Risk Assessments

During the 1990s, there was an academic debate concerning the characteristics of a good risk assessment. As a result of this academic discussion, the National Council on Crime and Delinquency (NCCD, 1997) suggested that instruments should involve all of the following: validity, reliability, fairness across all juvenile offenders, and utility (a system useful for court personnel). It is important that a measure be deemed valid because it allows judges to identify juveniles who are likely to return to the juvenile justice system. Reliability provides judges with information concerning whether juveniles who have the same characteristics are being placed in the same or similar treatment programs. This is important because risk assessment measures should be able to identify a given set of needs for juveniles who look similar in terms of needs and provide recommendations for the treatment; these recommendations should be consistent across juvenile offenders.

Further, risk assessments must be equitable. Risk assessment instruments should be fair and not contain any implicit or explicit biases due to demographic factors such as race, gender, or socioeconomic status. Finally, utility must be determined to see if the instrument is efficient and easy to implement. This is important because risk assessments should improve case management by providing opportunities for courts to save time and resources. Improving utility also allows various actors of the court to speak the same language, which provides a consistent way of evaluating processes and describing juveniles. More contemporary suggestions of objective risk assessments emphasize the extent to which the scoring process is objective, the ease and reliability of measurement, and the extent to which it predicts criminal behavior (Ausin, Dedel, & Weitzer, 2005). As can be seen from a comparison of these guidelines for risk assessment, objectivity, reliability, and the predictive validity of the assessments remain a focal concern. Unfortunately, more research is needed to make a determination about these characteristics of risk assessment.

More specifically, one major gap in the literature is the extent to which risk assessments are valid, reliable, fair, and useful for juveniles from diverse racial and ethnic backgrounds. For instance, when looking at the predictive validity of assessments, the extent to which these instruments are valid for minority groups (i.e., nondominant/nonmajority) is still unclear. For example, the findings of validation studies in the United States regarding the accuracy of risk assessment in African-American populations are seldom reported. Further, studies that examine the validity of risk assessment across race have found that some risk assessment tools predict future offending for all groups, some show variations in predictive validity by race, and some studies do not report how well the standardized tools perform by race. Given this research is a fairly new area, current trends remain inconclusive. While it is assumed that risk assessment tools are predictive and equitable across all racial groups, it is essential to take a closer look at how well risk assessments perform, especially as it relates to their predictive efficacy. Given that courts rely heavily on these risk assessments and that the development of these tools were not based on diverse samples of youth, it is critical to examine how accurate these tools are at predicting future offending.

Criticisms of Risk Assessment

As mentioned previously, juvenile risk assessments are intended to be administered by juvenile court officers upon a juvenile offender’s first contact with the court, in the form of an interview and use of collateral information such as previous criminal record and school records. The information on the assessment is intended to be used to place the juvenile in appropriate treatment based on needs and risk to recidivate information provided by the assessment. Although risk assessments offer information about best practices and next steps in offender treatment, they are not always used as intended (Miller & Maloney, 2013). Miller and Maloney (2013) found that while practitioners almost invariably filled out the assessments, decisions were often not made on the basis of the results. Another common argument against the use of risk assessments is the inclusion of static factors. Some researchers say that the use of static factors such as arrest history may affect minority youth differentially due to surveillance effects (Harcourt, 2015; Kempf-Leonard, 2007; Ryan et al., 2013).

A Review of Risk Assessment by Race

Overview of Systematic Review

Juvenile risk assessments were examined to understand how well they predict future recidivism by racial group/ethnicity/nationality. The systematic review identified 54 juvenile risk assessment validation studies that were published in peer-review academic journals from 2000 to 2016. We used the university library system and Google Scholar search. Keywords used to locate articles included the following: juvenile risk assessment; validation study; predictive validity; predicting recidivism; risk assessment tool; race; ethnicity; and the names of specific risk assessment tools in combination with the keywords. To cross-check and make sure that all the juvenile risk assessment tool names had the opportunity to be part of the review, a published list of juvenile risk assessment tools available on the Office of Juvenile Justice and Delinquency Prevention (OJJDP) website was used to compile a list of common assessment tools used in the field.

Evaluating Predictive Validity of Risk Assessments

Evaluations of the predictive validity of risk assessment tools have gone through a series of methodological evolutions. Early work in the evaluation of risk assessment has taken advantage of the chi square statistic. In this procedure, the risk level acquired from the interpretation of a risk assessment is used, as well as the recidivism numbers. If the assessment is measuring the risk of recidivism accurately, then those who recidivate should have been more likely to be placed in the high-risk category. While this methodology does a good job of determining differences in risk level, it does not reflect differences in the raw score.

The area under the curve (AUC) statistic and logistic regression have been among the most commonly used methods of evaluating predictive validity. This method uses the raw score as a predictor variable. AUC is used to determine the likelihood that a person with a higher score will have the outcome variable, compared to one with a lower score. This measure estimates and compares true positives and false positives and determines how much greater than chance a risk score performs to recidivism (Barnes et al., 2016). This measure is among the most common and appropriate approaches to estimating recidivism and evaluating predictive validity because it adjusts for the base rate of occurrence. Logistic regression also uses raw score as a predictor variable, with recidivism being the outcome variable. Both AUC and logistic regression provide a p-value. When examining the validity of risk assessments, these two methods are most commonly reported.

Table 1 lists the risk assessments identified and discussed in this article, and also provides the year that each assessment was developed, the abbreviated name of the assessment, and the assessment generation. Table 2 lists the studies and, for each one, the sample size, the name of the risk assessment used, the race/ethnicity/nationality of the participants (if reported), and the predictive validity of the assessment. The goal of this review was to determine the extent to which the juvenile risk assessment tool was valid across races/ethnicities.

In the event that a study did not examine racial or ethnic differences directly, its sample is described. In the event that a study was an attempt to validate an assessment in a country or culture outside the U.S., then the country or culture is listed instead. In this review, the predictive validity of juvenile risk assessment tools was examined. For studies published in the United States, any reference to the racial majority referred to juveniles who were classified as White, White American, or White (Non-Hispanic). For studies that took place outside the United States, the racial majority was defined by the dominant/majority race in the country (e.g., Aboriginal groups were identified as the minority group). A significant p-value indicates that the juvenile risk assessment measure significantly predicted future delinquency or had a significant probability of correctly classifying juveniles who reoffended, as defined by future contact with the court system (i.e., official records). In Table 2, ns indicates which findings are nonsignificant.

Overview of Articles Retrieved for Systematic Review

Of the 54 juvenile risk assessment studies found, 44% were conducted in the United States and 56% were conducted in other countries (e.g., Japan or Singapore). The sample size of juveniles across studies ranges from 35 to 15,072. In addition, 64.81% of studies that examined the validity of risk assessment using the AUC, which is a statistic used to estimate the rate of true positives (i.e., how accurately individuals were classified as higher-risk recidivate). The next common statistical methods used to determine predictive validity included linear regression and survival modeling. Racial group classifications (e.g., African American, Non-White, Native, and Aboriginal) varied across the 54 studies.

As shown in Table 2, 50% of the studies identified reported the predictive validity of risk assessments for Black youth, while 87.5% reported the validity of risk assessments for White youth. About 20.83% of the studies reported the predictive validity for Hispanic youth, and 41.67% reported the validity of risk assessment for native populations (i.e., Aboriginal youth or Native Americans). Further, 8.33% of studies divided the youth into White/Non-White categories to evaluate predictive validity.

The Validity of Risk Assessment Across Race and Ethnicity

A risk assessment total score is used to examine the overall predictive validity of a measure. Risk assessments vary in the number of items. Each measurement item or question generally represents 1 point. Each point represents an increased likelihood of reoffending. The sum of each measurement item is called the total score; this cumulative number is used by court officials to determine if a juvenile presents a low, moderate, or high risk of reoffending. For example, the Youth Level of Service/Case Management Inventory (YLS/CMI) uses a total score of 0–8 for low risk, 9–22 for moderate risk, 23–34 for high risk, and 35–41 for very high risk. Higher risk assessment scores suggest higher propensity for recidivism.

Although assessments generally give cut-off scores for levels of risk, assessments that are intended to be used as structured professional judgments do not. The Structured Assessment for Violence Risk in Youth (SAVRY) is one such tool. Decision-makers are not given criteria from which to ascertain risk levels; rather, they are responsible for making informed decisions using the information presented in the assessment results. Most validation studies are most concerned with how well total assessment scores predict future recidivism for juvenile offenders overall. As a result, this review focuses on the performance of the total risk assessment score.

According to Table 2, risk assessment total scores are generally found to be fairly predictive of recidivism across all race groups. This was true for both the United States and other countries. The risk assessment tools that reported the overall validity of the risk assessment scores by race/ethnicity found that the juvenile risk assessment tool significantly predicted future offenses, with p-values ranged from .001 to .05. However, it is important to note that a significant number of the 54 studies (n = 30) do not report on how well the juvenile risk assessment performs across racial/ethnic groups. This was most commonly true for risk assessment studies conducted outside the United States, where the conceptualization of race may have a different meaning. Further, there was one study that only examined White juveniles.

For the studies conducted in the United States that examined the validity across race and ethnicity, the common minority groups studied included White, Black, and Latino/Hispanic juveniles. While there were some variations across racial groups concerning the level/strength of predictive validity (e.g., White juveniles had slightly better/higher AUC scores), the overall scores significantly predicted future recidivism in all but one of the studies that included racial minorities. Shepherd, Singh, and Fullam (2015) found that the YLS/CMI was a significant predictor or recidivism only for the White offenders in a sample of 207 juveniles. Only 20.83% of the studies identified found that the risk assessment was not predictive for certain groups. Finally, two studies found that the initial risk assessment total score assigned to each juvenile was not predictive for neither the majority nor the minority groups (see Barnes et al., 2016; Edens & Cahill, 2007). Of the validation studies evaluated, the YLS/CMI was the most commonly used instrument that examined the validity of risk assessment across diverse racial/ethnic groups.

Race and Risk Assessment Future Research and Considerations

In addition to the need to report on the validity of risk assessment across diverse groups, there is a need to further examine the fairness and utility of the cut-off scores that are used to determine if a juvenile is low, moderate, or high risk of recidivism. Some research suggests that there may be valid concerns about the inconsistency in the scoring threshold for determining if a juvenile is low, moderate, or high risk. This is important because the practitioners are instructed to use the risk level as a decision-making tool; practitioners are told to give high-risk offenders more treatment than low- or moderate-risk offenders (Lovins & Latessa, 2013). Offenders are given more or less supervision and treatment depending on the risk score. For this reason, more consideration should be given to examining risk levels and their validity in future research. Fewer studies today examine the efficacy of low, moderate, and high risk as decision-making criteria for individual assessments. There are two types of errors that an assessment can facilitate: an assessment can overpredict or underpredict recidivism risk. Overprediction is identifying an offender as high risk to recidivate when they are actually at moderate or low risk. Underprediction involves treating offenders as low or moderate risk to recidivate when they are actually at high or very high risk. The creation of cut-off scores requires a decision about the relative danger or inconvenience of making an error in either direction. The current review suggests somewhat promising findings; however, more research is necessary to examine the degree to which these measures perform well for all racial groups (especially in the area of risk level/risk category classification).

In general, this review suggests that risk assessment is a valid strategy for estimating the risk of reoffending among juvenile offender populations. While the overall performances of such measures may vary across subgroups (i.e., total score is a stronger predictor for White offenders than for Non-White offenders), most risk assessments significantly predict recidivism for juveniles in general, regardless of racial group.

Review of the Literature and Primary Sources

Modern risk assessments are largely based on the RNR model, which suggests that if you identify risk factors that increase involvement in crime, identify what a juvenile needs, and assess the level of youth responsivity, then individuals will have a decreased likelihood of involvement in delinquent activities. A crucial element to predicting future offenses, the tool heavily relies on researchers’ ability to identify the most important risk factors needed to estimate delinquency. As a result, many researchers have explored both proximal and distal factors that may influence pathways of delinquency.

Measuring individual, microlevel, and macrolevel criminogenic risk factors, by way of risk assessments, has been an ongoing strategy for addressing the disproportionate number of minority offenders who have come in contact with the justice system. Researchers have identified eight factors that are central to assessing criminogenic risk, and thus estimating future delinquency, known as the “Central Eight” (Andrews et al., 1990; Latessa, Listwan, & Koetzle, 2013). Andrews et al. (1990) ranked these factors and further classified them into two categories based on their level of impact on future delinquency. These two groups of variables are known as the “Big Four” and the “Moderate Four.” The “Big Four” variables include the following: history of antisocial behavior (e.g., criminal history); antisocial personality (e.g., impulsivity and aggressive behavior); antisocial cognition (e.g., antisocial attitudes and beliefs); and antisocial associates (e.g., having delinquent friends or peers). The “Moderate Four” include substance abuse, family/marital relationships, school/work, and program-social recreation.

Together, the Central Eight have been used to construct risk assessments, such that the more factors that are present in a juvenile’s life, the more likely they will be to recidivate. Consequently, the higher the risk score, the more intervention services that a juvenile will receive when coming in contact with corrections. Further, these factors are used to match risk with needs so that services that a juvenile receives map to dynamic risk factors.

Given that most risk assessments have been developed based on White male samples, the question of the accuracy of these instruments across racial groups remains. More specifically, there are questions concerning the extent to which risk assessments exacerbate or overestimate the risk of recidivism for racial minorities. There are also questions regarding the extent to which these measures are arbitrarily increasing the risk of recidivism. For instance, the use of static variables like criminal history is often factored into risk assessment measurements. Given that criminal history may disproportionately affect ethnic groups with a history of marginalization (e.g., groups that are oversurveillanced), careful considerations must be made concerning the true contribution of static variables.

Results from this systematic review suggest that among the most widely used risk assessment instruments, there is not much difference in predictive validity by race. However, while raw scores and risk-level grouping have not been shown to have differential predictive validity, the use of clinical overrides has yet to be examined at length. The term clinical overrides refers to the ability of an assessor to change the final risk level of an offender based on the assessor’s own judgment, regardless of the actual risk score. There is not a large base of knowledge about which offenders get their assessments overridden and how this differentially affects groups of juveniles. Further, there is little information about how certain groups of offenders might differ on certain subscales, although they have similar final scores. More research is needed to examine additional ways that risk assessments may differ by racial group.

The observations in this review are consistent with current research. Skeem and Lowenkamp (2016) found that risk assessments are more free of bias and provide standardized ways to improve racial disparities involved in court decision-making. In general, given the evolution of risk assessment, the current performance of such measurements seems to be developing in a promising direction. At the same time, it is important to note that the quality and implementation of risk assessment measures differ. Researchers must evaluate the quality of these measures across subgroups to ensure that they do not cause harmful and disparate outcomes for certain groups.

Current research on juvenile risk assessments that report on the predictive accuracy of measures seems promising, but there still are some assessments that have not been examined as it pertains to how well they perform across specific groups. Given the efforts of corrections and the public to promote a system of fairness and justice, a standard for using such instruments should require reporting validity across racial groups and further examining other processes and practices that may threaten the equity and fairness of such tools.

CSG Justice Center. (2017). Risk Assessment Instruments Validated and Implemented in Correctional Settings in the United States. Retreived from https://csgjusticecenter.org/wp-content/uploads/2014/07/Risk-Assessment-Instruments-Validated-and-Implemented-in-Correctional-Settings-in-the-United-States.pdf.

CSG Justice Center (2017). Risk Assessment Instrument Guide. Retreived from https://csgjusticecenter.org/wp-content/uploads/2014/07/Risk-Instruments-Guide.pdf.

Development Services Group, Inc. (2015). Risk and needs assessment for youths. Washington, DC: Office of Juvenile Justice and Delinquency Prevention. Retrieved from http://www.ojjdp.gov/mpg/litreviews/RiskandNeeds.pdf.Find this resource:

    Further Reading

    Andrews, D. A., & Bonta, J. (2010). The psychology of criminal conduct. New York: Routledge.Find this resource:

      Andrews, D. A., Bonta, J., & Wormith, S. J. (2004). The Level of Service/Case Management Inventory (LS/CMI): User’s manual. Toronto: Multi Health Systems.Find this resource:

        Bonta, J. (2002). Offender risk assessment guidelines for selection and use. Criminal Justice and Behavior, 29(4), 355–379.Find this resource:

          Bonta, J. (2007). Offender risk assessment and sentencing. Canadian Journal of Criminology and Criminal Justice, 49(4), 519–529.Find this resource:

            Bonta, J., & Andrews, D. A. (2007). Risk-need-responsivity model for offender assessment and rehabilitation. Rehabilitation, 6, 1–22.Find this resource:

              Bonta, J., LaPrairie, C., & Wallace-Capretta, S. (1997). Risk prediction and re-offending: Aboriginal and non-aboriginal offenders. Canadian Journal of Criminology, 39, 127.Find this resource:

                Bonta, J., Law, M., & Hanson, K. (1998). The prediction of criminal and violent recidivism among mentally disordered offenders: A meta-analysis. Psychological Bulletin, 123(2), 123–142.Find this resource:

                  Burgess, E. M. (1925). The working of the indeterminate sentence law and the parole system in Illinois. Springfield: Illinois Parole Board.Find this resource:

                    Craig, L. A., Gannon, T. A., & Dixon, L. (Eds.). (2013). What works in offender rehabilitation: An evidence-based approach to assessment and treatment. Hoboken, NJ: John Wiley & Sons.Find this resource:

                      Grigorenko, E. L. (Ed.). (2012). Handbook of juvenile forensic psychology and psychiatry. New York: Springer.Find this resource:

                        Latessa, E. J., Listwan, S. J., & Koetzle, D. (2013). What works (and doesn’t) in reducing recidivism. New York: Routledge.Find this resource:

                          Otto, R. K., & Douglas, K. S. (Eds.). (2011). Handbook of violence risk assessment. New York: Routledge.Find this resource:

                            Ward, G. (2012). The black child savers: Racial democracy and American juvenile justice. Chicago: University of Chicago Press.Find this resource:

                              References

                              Abracen, J., Mailloux, D. L., Serin, R. C., Cousineau, C., Bruce, P., Looman, M., & Looman, J. (2004). A model for the assessment of static and dynamic factors in sexual offenders. Journal of Sex Research, 41(4), 321–328.Find this resource:

                                Albonetti, C. A. (1991). An integration of theories to explain judicial discretion. Social Problems, 38(2), 247–266.Find this resource:

                                  Andrews, D. A., & Bonta, J. (2010). Rehabilitating criminal justice policy and practice. Psychology, Public Policy, and Law, 16(1), 39–55.Find this resource:

                                    Andrews, D. A., Bonta, J., & Hoge, R. D. (1990). Classification for effective rehabilitation: Rediscovering psychology. Criminal Justice and Behavior, 17(1), 19–52.Find this resource:

                                      Andrews, D. A., Bonta, J., & Wormith, J. S. (2006). The recent past and near future of risk and/or need assessment. Crime & Delinquency, 52(1), 7–27.Find this resource:

                                        Ausin, J., Dedel, K., & Weitzer, R. J. (2005). Alternatives to the secure detention and confinement of juvenile offenders. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice Delinquency Prevention.Find this resource:

                                          Baglivio, M. T., & Jackowski, K. (2013). Examining the validity of a juvenile offending risk assessment instrument across gender and race/ethnicity Youth Violence and Juvenile Justice, 11(1), 26–43.Find this resource:

                                            Baglivio, M. T. (2009). The assessment of risk to recidivate among a juvenile offending population Journal of Criminal Justice, 37, 596–607.Find this resource:

                                              Barnes, A. R., Campbell, N. A., Anderson, V. A., Campbell, C. A., Onifade, E., & Davidson, W. S. (2016). Validity of initial, exit, and dynamic juvenile risk assessment: An examination across gender and race/ethnicity Journal of Offender Rehabilitation, 55(1), 21–38.Find this resource:

                                                Bechtel, K., Lowenkamp, C. T., & Latessa, E. (2007). Assessing the risk of re-offending for juvenile offenders using the Youth Level of Service/Case Management Inventory Journal of Offender Rehabilitation, 45(3–4), 85–108.Find this resource:

                                                  Bengston, S., & Langstrom, N. (2007). Unguided clinical and actuarial assessment of re-offending risk: A direct comparison with sex offenders in Denmark. Sex Abuse, 19, 135–153.Find this resource:

                                                    Book, P., Thomas, J., & Steinke, D. (2004). Predictors of recidivism in a community-based residential program for delinquent youth. Journal for Juvenile Justice Services, 19(1&2), 101–120.Find this resource:

                                                      Catchpole, R. H., & Gretton, H. M. (2003). The predictive validity of risk assessment with violent young offenders. A 1-year examination of criminal outcome Criminal Justice and Behavior, 30(6), 688–708.Find this resource:

                                                        Chu, C. M., Goh, M. L., & Chong, D. (2015). The predictive validity of SAVRY ratings for assessing youth offenders in Singapore: A comparison with YLS/CMI ratings Criminal Justice and Behavior, 43(6), 793–810.Find this resource:

                                                          Chu, C. M., Lee, Y., Zeng, G., Yim, G., Tan, C. Y., Ang, Y., . . . Ruby, K. (2015). Assessing youth offenders in a non-Western context: The predictive validity of the YLS/CMI ratings Psychological Assessment, 27(3), 1013–1021. .Find this resource:

                                                            Chu, C. M., Yu, H., Lee, Y., & Zeng, G. (2014). The utility of the YLS/CMI-SV for assessing youth offenders in Singapore Criminal Justice and Behavior, 41(2), 1437–1457.Find this resource:

                                                              Corrado, R. R., Vincent, G. M., Hart, S. D., & Cohen, I. M. (2004). Predictive validity of the psychopathy checklist: Youth version for general and violent recidivism Behavioral Science and the Law, 22, 5–22.Find this resource:

                                                                Dolan, M. C., & Rennie, C. E. (2008). The structured assessment of violence risk in youth as a predictor of recidivism in a United Kingdom cohort of adolescent offenders with conduct disorder Psychological Assessment, 20(1), 35–46.Find this resource:

                                                                  Douglas, K. S., Epstein, M. E., & Poythress, N. G. (2008). Criminal recidivism among juvenile offenders: Testing the incremental and predictive validity of three measures of psychopathic features Law and Human Behavior, 32(5), 423–438.Find this resource:

                                                                    Edens, J. F., & Cahill, M. A. (2007). Psychopathy in adolescence and criminal recidivism in young adulthood: Longitudinal results From a multiethnic sample of youthful offenders Assessment, 4(1), 58–64.Find this resource:

                                                                      Flores, A. W., Travis, L. F., & Latessa, E. J. (2003). Case classification for juvenile corrections: An assessment of the Youth Level of Service/Case Management Inventory (YLS/CMI). Cincinnati: Division of Criminal Justice.Find this resource:

                                                                        Fox, J. A. (1996). Trends in juvenile violence: A report to the United States Attorney General on current and future rates of juvenile offending. Washington, DC: US Department of Justice, Bureau of Justice Statistics, pp. 1–15.Find this resource:

                                                                          Gatti, U., Tremblay, R. E., & Vitaro, F. (2009). Iatrogenic effect of juvenile justice. Journal of Child Psychology and Psychiatry, 50(8), 991–998.Find this resource:

                                                                            Gavazzi, S. M., Yarcheck, C. M., Sullivan, J. M., Jones, S. C., & Khurana, A. (2008). Global risk factors and the recidivism rates in a sample of first-time misdemeanant offenders International Journal of Offender Therapy and Comparative Criminology, 52(3), 330–345.Find this resource:

                                                                              Gossner, D., & Wormith, J. S. (2007). The prediction of recidivism among young offenders in Saskatchewan. Canadian Journal of Police & Security Services, 5(1/2), 70–82.Find this resource:

                                                                                Gretton, H. M., McBride, M., Hare, R., O’Shaunessy, R., & Kumka, G. (2001). Psychopathy and recidivism in adolescent sex offenders. Criminal Justice and Behavior, 28(4), 427–449.Find this resource:

                                                                                  Hay, C., Widdowson, A. O., Bates, M., Baglivio, M. T., Jackowski, K., & Greenwald, M. A. (2016). Predicting recidivism among released juvenile offenders in Florida: An evaluation of the Residential Positive Achievement Change Tool Youth Violence and Juvenile Justice, 16(1), 97–116.Find this resource:

                                                                                    Harcourt, B. (2015). Risk as a proxy for race: The dangers of risk assessment. Federal Probation, 27, 237–243.Find this resource:

                                                                                      Henggeler, S. W., Belton, G. B., & Smith, L. A. (1992). Family preservation using mutisystemic therapy: An effective alternative to incarcerating serious juvenile offenders. Journal of Consulting and Clinical Psychology, 60(6), 953–961.Find this resource:

                                                                                        Hilterman, E. L. B., Nicholls, T. L., & Nieuwenhuizen, C. V. (2014). Predictive validity of risk assessments in juvenile offenders: Comparing the SAVRY, PCL:YV, and YLS/CMI with unstructured clinical assessments Assessment, 21(3), 324–339.Find this resource:

                                                                                          Hodges, K., & Kim, C. (2000). Psychometric study of the Child and Adolescent Functional Assessment Scale: Prediction of contact with the law and poor school attendance. Journal of Abnormal Child Psychology, 38(3), 287–297.Find this resource:

                                                                                            Huey, S. J., Henggeler, S. W., Brondino, M. J., & Pickrel, S. G. (2000). Mechanisms of change in multisystemic therapy: Reducing delinquent behavior through therapist adherence and improved family and peer functioning. Journal of Consulting and Clinical Psychology, 68(3), 451–467.Find this resource:

                                                                                              Jung, S., & Rawana, E. P. (1999). Risk and need assessment of juvenile offenders. Criminal Justice and Behavior, 26(1), 69–89.Find this resource:

                                                                                                Kempf-Leonard, K. (2007). Minority youths and juvenile justice Youth Violence and Juvenile Justice, 5(1), 71–81.Find this resource:

                                                                                                  Krysik, J., & LeCroy, C. W. (2002). The empirical validation of an instrument to predict risk of recidivism among juvenile offenders. Research on Social Work Practice, 12(1), 71–81.Find this resource:

                                                                                                    Latessa, E. J., Listwan, S. J., & Koetzle, D. (2013). What works (and doesn’t) in reducing recidivism. New York: Routledge.Find this resource:

                                                                                                      Latessa, E., & Lovins, B. (2010). The role of offender risk assessment: A policy maker guide. Victims and Offenders, 5, 203–219.Find this resource:

                                                                                                        Leiber, M., Bishop, D., & Chamlin, M. B. (2011). Juvenile justice decision-making before and after the implementation of the disproportionate minority contact (DMC) mandate. Justice Quarterly, 28(3), 460–492.Find this resource:

                                                                                                          Leistico, A. R., & Salekin, R. T. (2003). Testing the reliability and validity of the Risk, Sophistication-Maturity, and Treatment Amenability Instrument (RST-i): An assessment tool for juvenile offenders. International Journal of Forensic Mental Health, 2(2), 101–117.Find this resource:

                                                                                                            Lipsey, M. W., Howell, J. C., Kelly, M., Chapman, G., & Carver, D. (2010). Improving the effectiveness of juvenile justice programs. Washington, DC: Center for Juvenile Justice Reform at Georgetown University.Find this resource:

                                                                                                              Lodewijks, H. P. B., Doreleijers, T. A. H., & Ruiter, C. (2008). SAVRY risk assessment in violent Dutch adolescents: Relation to sentencing and recidivism Criminal Justice and Behavior, 35(6), 696–709.Find this resource:

                                                                                                                Lovins, B., & Latessa, E. (2013). Creation and validation of the Ohio Youth Assessment System (OYAS) and strategies for successful implementation. Justice Research and Policy, 15(1), 67–93.Find this resource:

                                                                                                                  Lowenkamp, C. T., & Latessa, E. J. (2005). Increasing the effectiveness of correctional programming through the risk principle: Identifying offenders for residential placement. Correctional Programming & Risk, 4(2), 263–290.Find this resource:

                                                                                                                    Marczyk, G. R., Heilburn, K., Lander, T., & DeMateo, D. (2003). Predicting juvenile recidivism with the PCL:YV, MAYSI, and YLS/CMI International Journal of Forensic Mental Health, 2(1), 7–18.Find this resource:

                                                                                                                      Marshall, J., Egan, V., English, M., & Jones, R. M. (2005). The relative validity of psychopathy versus risk/needs-based assessments in the prediction of adolescent offending behavior Legal and Criminological Psychology, 11(2), 197–210.Find this resource:

                                                                                                                        McCafferty, J. T. (2016). Unjust disparities? The impact of race on juvenile risk assessment outcomes. Criminal Justice Policy Review. Advance online publication.Find this resource:

                                                                                                                          McGowan, M. R., Horn, R. A., & Mellott, R. N. (2011). The predictive validity of the Structured Assessment of Violence Risk in Youth in secondary educational settings Psychological Assessment, 23(2), 478–486.Find this resource:

                                                                                                                            Meyers, J. R., & Schmidt, F. (2008). Predictive validity of the Structured Assessment for Violence Risk in Youth (SAVRY) with juvenile offenders Criminal Justice and Behavior, 35(3), 344–355.Find this resource:

                                                                                                                              Miller, J., & Maloney, C. (2013). Practitioner compliance with risk/needs assessment tools: A theoretical and empirical assessment Criminal Justice and Behavior, 40, 716–736.Find this resource:

                                                                                                                                Odgers, C. L., Reppucci, N. D., & Moretti, M. M. (2005). Nipping psychopathy in the bud: An examination of the convergent, predictive, and theoretical utility of the PCL-YV among adolescent girls. Behavioral Sciences & the Law, 23(6), 743–763.Find this resource:

                                                                                                                                  Olver, M. E., Stockdale, K. C., & Wong, S. C. P. (2012). Short- and long-term prediction of recidivism using the Youth Level of Service/Case Management Inventory in a sample of serious young offenders Law and Human Behavior, 36(4), 331–344.Find this resource:

                                                                                                                                    Onifade, E., Nyandoro, A. S., Davidson, W., & Campbell, C. (2010). Truancy and patterns of criminogenic risk in a young offender population Youth Violence and Juvenile Justice, 8(1), 3–18.Find this resource:

                                                                                                                                      Onifade, E., Davison, W., & Campbell, C. (2009). Risk assessment: The predictive validity of the Youth Level of Service Case Management Inventory with African Americans and girls Journal of Ethnicity in Criminal Justice, 7, 205–221.Find this resource:

                                                                                                                                        Onifade, E., Davidson, W., Campbell, C., Turke, G., Malinowski, J., & Turner, K. (2008). Predicting recidivism in probationers with the Youth Level of Service Case Management Inventory (YLS/CMI) Criminal Justice and Behavior, 35(4), 474–483.Find this resource:

                                                                                                                                          Onifade, E., Petersen, J., Bynum, T., & Davidson, W. (2011). Multilevel recidivism prediction incorporating neighborhood socioeconomic ecology in juvenile justice risk assessment. Criminal Justice and Behavior, 38, 840–853.Find this resource:

                                                                                                                                            Platt, A. (1969). The rise of the child-saving movement: A study in social policy and correctional reform. The ANNALS of the American Academy Policy and Social Science, 38(1), 21–38.Find this resource:

                                                                                                                                              Polaschek, D. L. L. (2012). An appraisal of the Risk-Need-Responsivity (RNR) model of offender rehabilitation and its application in correctional treatment. Legal and Criminological Psychology, 17, 1–17.Find this resource:

                                                                                                                                                Quist, R. M., & Matshazi, G. M. (2000). The Child and Adolescent Functional Assessment Scale (CAFAS): A dynamic predictor of juvenile recidivism. Adolescence, 35(137), 181–192.Find this resource:

                                                                                                                                                  Rembert, D. A., Henderson, H., & Pirtle, D. (2014). Differential racial/ethnic predictive validity Youth Violence and Juvenile Justice, 12(2), 152–166.Find this resource:

                                                                                                                                                    Rennie, C., & Dolan, M. (2010). Predictive validity of the Youth Level of Service/Case Management Inventory in custody sample in England Journal of Forensic Psychiatry & Psychology, 21(3), 407–425.Find this resource:

                                                                                                                                                      Risler, E. A., Sutphen, R., & Shields, J. (2000). Preliminary validation of the Juvenile First Offender Risk Assessment Index. Research on Social Work Practice, 10(1), 111–126.Find this resource:

                                                                                                                                                        Rodriguez, N. (2013). Concentrated disadvantage and the incarceration of youth: Examining how context affects juvenile justice. Journal of Research in Crime and Delinquency, 50(2), 189–215.Find this resource:

                                                                                                                                                          Ryan, J. P., Williams, A. B., & Courtney, M. E. (2013). Adolescent neglect, juvenile delinquency and the risk of recidivism. Journal of Youth Adolescence, 42, 454–465.Find this resource:

                                                                                                                                                            Schmidt, F., Campbell, M. A., & Houlding, C. (2011). Comparing analyses of the YLS/CMI, SAVRY, and PCL: YV in adolescent offenders: A 10-year follow-up into adulthood Youth Violence and Juvenile Justice, 9(1), 23–42.Find this resource:

                                                                                                                                                              Schmidt, F., Hoge, R. D., & Gomes, L. (2005). Reliability and validity analyses of the Youth Level of Service/Case Management Inventory Criminal Justice and Behavior, 32(3), 329–344.Find this resource:

                                                                                                                                                                Schmidt, F., McKinnon, L., Chattha, H. K., & Brownlee, K. (2006). Concurrent and predictive validity of the Psychopathy Checklist: Youth Version across gender and ethnicity Psychological Assessment, 18(4), 393–401.Find this resource:

                                                                                                                                                                  Schwalbe, C. (2007). Risk assessment for juvenile justice: A meta-analysis. Law and Human Behavior, 31(5), 449–462.Find this resource:

                                                                                                                                                                    Schwalbe, C. S. (2009). Risk assessment stability: A revalidation study of the Arizona Risk/Needs Assessment Instrument Research on Social Work Practice, 19(2), 205–213.Find this resource:

                                                                                                                                                                      Schwalbe, C. S., Fraser, M. W., & Day, S. H. (2007). Predictive validity of the Joint Risk Matrix with juvenile offenders. A focus on gender and race/ethnicity Criminal Justice and Behavior, 34(3), 348–361.Find this resource:

                                                                                                                                                                        Schwalbe, C. S., Fraser, M. W., Day, S. H., & Cooley, V. (2006). Classifying juvenile offenders according to risk of recidivism: Predictive validity, race/ethnicity, and gender Criminal Justice and Behavior, 33(3), 305–324.Find this resource:

                                                                                                                                                                          Schwalbe, C. S., Fraser, M. W., Day, S. H., & Arnold, E. M. (2004). North Carolina Assessment of Risk (NCAR) Journal of Offender Rehabilitation, 40(1–2), 1–22.Find this resource:

                                                                                                                                                                            Shapiro, C. J., Smith, B. H., Malone, P. S., & Collaro, A. L. (2010). Natural experiment in deviant peer exposure and youth recidivism. Journal of Clinical Child & Adolescent Psychology, 39(2), 242–251.Find this resource:

                                                                                                                                                                              Sharkey, J. D., Furlong, M. J., Jimerson, S. R., & O’Brien, K. M. (2003).Evaluating the utility of a risk assessment to predict among male and female adolescents. Education and Treatment of Children, 26(4), 467–494.Find this resource:

                                                                                                                                                                                Shepherd, S. M., Leubbers, S., & Dolan, M. (2013). Gender and ethnicity in juvenile risk assessment Criminal Justice and Behavior, 40(4), 388–408.Find this resource:

                                                                                                                                                                                  Shepherd, S. M., Leubbers, S., Ferguson, M., Ogloff, J. R., & Dolan, M. (2013). The utility of the SAVRY across ethnicity in Australian young offenders Psychology, Public Policy, and Law, 20(1), 31–45.Find this resource:

                                                                                                                                                                                    Shepherd, S. M., Singh, J. P., & Fullam, R. (2015). Does the Youth Level of Service/Case Management Inventory generalize across ethnicity? Journal of Forensic Mental Health, 14(3), 193–204.Find this resource:

                                                                                                                                                                                      Skeem, J., & LowenKamp, C. (2016). Risk, race, and recidivism: Predictive bias and disparate impact Criminology, 54(4), 680–712.Find this resource:

                                                                                                                                                                                        Snowden, R. J., Gray, N. S., & Taylor, J. (2010). Risk assessment for violence in individuals from an ethnic minority group National Journal of Forensic Mental Health, 9, 118–123.Find this resource:

                                                                                                                                                                                          Stams, G. J. J. (2015). From Criminogenic Risk to rehabilitation: Is there a need for a culturally sensitive approach? International Journal of Offender Therapy and Comparative Criminology, 59(12), 1263–1266.Find this resource:

                                                                                                                                                                                            Stockdale, K. C., Olver, M. E., & Wong, S. C. (2010). The Psychopathy Checklist: Youth Version and adolescent and adult recidivism: Considerations with respect to gender, ethnicity, and age Psychological Assessment, 22(4), 768–781.Find this resource:

                                                                                                                                                                                              Takahashi, M., Mori, T., & Kroner, D. G. (2013). A cross-validation of the Youth Level of Service/Case Management Inventory (YLS/CMI) among Japanese juvenile offenders Law and Human Behavior, 37(6), 389–400.Find this resource:

                                                                                                                                                                                                Taylor, L. R. (2016). General responsivity adherence in juvenile drug treatment court: Examining the impact on substance-use outcome. Journal of Drug Issues, 46(1), 24–40.Find this resource:

                                                                                                                                                                                                  Thompson, A. P., & McGrath, A. (2012). Subgroup differences and implications for contemporary risk-need assessment with juvenile offenders Law and Human Behavior, 36(4), 345–355.Find this resource:

                                                                                                                                                                                                    U.S. Department of Health and Human Services. National Council on Crime and Delinquency. (1997). Child Abuse and Neglect: Improving Consistency in Decision Making.Find this resource:

                                                                                                                                                                                                      Van der Put, C., Stams, G. J. J. M., Dekovic, M., & van der Laan, P. H. (2012). Predictive validity of the Washington State juvenile court pre-screen assessment in the Netherlands: The development of a new scoring system. Assessment, 21(1), 92–107.Find this resource:

                                                                                                                                                                                                        Vaswani, N., & Merone, L. (2014). Are there risks with risk assessment? A study of the predictive validity of the Youth Level of Service–Case Management Inventory with young offenders in Scotland. British Journal of Social Work, 44, 2163–2181.Find this resource:

                                                                                                                                                                                                          Vincent, G. M., Chapman, J., & Cook, N. E. (2011). Risk-needs assessment in juvenile justice. Predictive validity of the SAVRY, racial differences, and the contribution of needs factors Criminal Justice and Behavior, 38(1), 42–62.Find this resource:

                                                                                                                                                                                                            Ward, G. (2009). The “other” child-savers: Racial politics of the parental state. In A. M. Platt (Ed.), The child savers: The invention of delinquency (40th anniversary ed., pp. 1–68). Piscataway, NJ: Rutgers University Press.Find this resource:

                                                                                                                                                                                                              Appendix

                                                                                                                                                                                                              Table 1. List of Risk Assessment Names, Year of Development, Abbreviations, and Generation

                                                                                                                                                                                                              Risk Assessment Name

                                                                                                                                                                                                              Year Developed

                                                                                                                                                                                                              Abbreviation

                                                                                                                                                                                                              Assessment Generation

                                                                                                                                                                                                              Antisocial Process Screening Device

                                                                                                                                                                                                              2001

                                                                                                                                                                                                              APSD

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              Arizona Risk/Needs Assessment

                                                                                                                                                                                                              1998

                                                                                                                                                                                                              ARNA

                                                                                                                                                                                                              3rd*

                                                                                                                                                                                                              Child and Adolescent Functional Assessment Scale

                                                                                                                                                                                                              1989

                                                                                                                                                                                                              CAFAS

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              Carlson Psychological Survey

                                                                                                                                                                                                              1981

                                                                                                                                                                                                              CPS

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              First Offender Assessment Index

                                                                                                                                                                                                              2000

                                                                                                                                                                                                              FORAI

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              Global Risk Assessment Device

                                                                                                                                                                                                              2003

                                                                                                                                                                                                              GRAD

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              HCR-20 Risk Management Guide

                                                                                                                                                                                                              1997

                                                                                                                                                                                                              HCR-20

                                                                                                                                                                                                              1st

                                                                                                                                                                                                              Los Angeles County Needs Assessment Instrument

                                                                                                                                                                                                              2003

                                                                                                                                                                                                              LAC

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              Massachusetts Youth Screening Instrument

                                                                                                                                                                                                              2000

                                                                                                                                                                                                              MAYSI

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              Ministry Risk/Need Assessment

                                                                                                                                                                                                              1994

                                                                                                                                                                                                              MRNAF

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              North Carolina Assessment of Risk

                                                                                                                                                                                                              2004

                                                                                                                                                                                                              NCAR

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              National Council on Crime and Delinquency Risk Assessment

                                                                                                                                                                                                              2002

                                                                                                                                                                                                              NCCD

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              Orange County Risk Scale

                                                                                                                                                                                                              1998

                                                                                                                                                                                                              OCRS

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              Ohio Youth Assessment System–Disposition Tool

                                                                                                                                                                                                              2013

                                                                                                                                                                                                              OYAS-Dis

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              Positive Achievement Change Tool

                                                                                                                                                                                                              2005

                                                                                                                                                                                                              PACT

                                                                                                                                                                                                              4th

                                                                                                                                                                                                              Psychopathy Checklist Youth Version

                                                                                                                                                                                                              1994

                                                                                                                                                                                                              PCL: YV

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              Risk, Sophistication-Maturity, and Treatment Amenability Instrument

                                                                                                                                                                                                              1998

                                                                                                                                                                                                              RST-I

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              Structured Assessment of Violent Risk in Youth

                                                                                                                                                                                                              2001

                                                                                                                                                                                                              SAVRY

                                                                                                                                                                                                              1st

                                                                                                                                                                                                              Violence Risk Appraisal Guide

                                                                                                                                                                                                              2006

                                                                                                                                                                                                              VRAG

                                                                                                                                                                                                              2nd

                                                                                                                                                                                                              Washington State Juvenile Court Assessment

                                                                                                                                                                                                              1998

                                                                                                                                                                                                              WJSCA

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              Youth Level of Service/Case Management Inventory

                                                                                                                                                                                                              2002

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              3rd

                                                                                                                                                                                                              Note: (*) Highlights instruments that were difficult to determine the generation. Risk assessment generations are not commonly reported in research studies. Authors reviewed measures and determined generation based on a reviewing risk assessment instruments or the manuals.

                                                                                                                                                                                                              Table 2. Predictive Validity of Juvenile Risk Assessment Across Race/Ethnicity/Nationality

                                                                                                                                                                                                              Author(s)

                                                                                                                                                                                                              Year Article Was Published

                                                                                                                                                                                                              Sample Size

                                                                                                                                                                                                              Measure(s)

                                                                                                                                                                                                              Race/Ethnicity/Nationality

                                                                                                                                                                                                              Method (e.g., AUC)

                                                                                                                                                                                                              Overall Predictive Validity

                                                                                                                                                                                                              Predictive Validity (Majority group)

                                                                                                                                                                                                              Predictive Validity (Minority group)

                                                                                                                                                                                                              Baglivio & Jackowski

                                                                                                                                                                                                              2013

                                                                                                                                                                                                              15,072

                                                                                                                                                                                                              PACT

                                                                                                                                                                                                              White, Black, Hispanic

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Baglivio

                                                                                                                                                                                                              2009

                                                                                                                                                                                                              8,132

                                                                                                                                                                                                              PACT

                                                                                                                                                                                                              White, Non-White

                                                                                                                                                                                                              Regression

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Barnes et al.

                                                                                                                                                                                                              2016

                                                                                                                                                                                                              360

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              White, Non-White

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              Bechtel et al.

                                                                                                                                                                                                              2007

                                                                                                                                                                                                              4,482

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              White, Non-White

                                                                                                                                                                                                              Regression

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              Book et al.

                                                                                                                                                                                                              2004

                                                                                                                                                                                                              153

                                                                                                                                                                                                              CPS

                                                                                                                                                                                                              Regression

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              Catchpole et al.

                                                                                                                                                                                                              2003

                                                                                                                                                                                                              76

                                                                                                                                                                                                              SAVRY,

                                                                                                                                                                                                              YLS/CMI, PCL: YV

                                                                                                                                                                                                              White, Aboriginal, Asian

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Chu et al.

                                                                                                                                                                                                              2015

                                                                                                                                                                                                              163

                                                                                                                                                                                                              SAVRY

                                                                                                                                                                                                              Singaporean

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              Chu et al.

                                                                                                                                                                                                              2015

                                                                                                                                                                                                              3,264

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              Singaporean

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              Chu et al.

                                                                                                                                                                                                              2014

                                                                                                                                                                                                              3,264

                                                                                                                                                                                                              YLS/CMI (short version)

                                                                                                                                                                                                              Singaporean

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              Corrado et al.

                                                                                                                                                                                                              2004

                                                                                                                                                                                                              128

                                                                                                                                                                                                              PCL: YV

                                                                                                                                                                                                              White, Canadian Aboriginal,

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              Dolan & Rennie

                                                                                                                                                                                                              2008

                                                                                                                                                                                                              99

                                                                                                                                                                                                              SAVRY,

                                                                                                                                                                                                              PCL: YVns

                                                                                                                                                                                                              White, Asian, Afro-Caribbean, Oriental

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              Douglas & Epstein

                                                                                                                                                                                                              2008

                                                                                                                                                                                                              83

                                                                                                                                                                                                              PCL: YVns, APSDns,

                                                                                                                                                                                                              CPS

                                                                                                                                                                                                              White, Black

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              Edens & Cahill

                                                                                                                                                                                                              2007

                                                                                                                                                                                                              75

                                                                                                                                                                                                              PCL: YV

                                                                                                                                                                                                              White, Black, Hispanic

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              Flores et al.

                                                                                                                                                                                                              2003

                                                                                                                                                                                                              1,679

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              Not specified

                                                                                                                                                                                                              Regression

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              Gavazzi et al.

                                                                                                                                                                                                              2008

                                                                                                                                                                                                              711

                                                                                                                                                                                                              GRAD

                                                                                                                                                                                                              White, Black

                                                                                                                                                                                                              Chi square

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              Gossner & Wormith

                                                                                                                                                                                                              2007

                                                                                                                                                                                                              94

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              Canadian Non-Aboriginal, Canadian Aboriginal

                                                                                                                                                                                                              Regression

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Gretton et al.

                                                                                                                                                                                                              2001

                                                                                                                                                                                                              220

                                                                                                                                                                                                              PCL: YV

                                                                                                                                                                                                              White, Native Indian

                                                                                                                                                                                                              Regression

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              N/A

                                                                                                                                                                                                              N/A

                                                                                                                                                                                                              Hay et al.

                                                                                                                                                                                                              2016

                                                                                                                                                                                                              4,700

                                                                                                                                                                                                              PACT

                                                                                                                                                                                                              White, Black, Hispanic

                                                                                                                                                                                                              Regression

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              Hilterman et al.

                                                                                                                                                                                                              2014

                                                                                                                                                                                                              105

                                                                                                                                                                                                              SAVRY, YLS/CMI,

                                                                                                                                                                                                              PCL: YV

                                                                                                                                                                                                              Spanish, European, South American, North African/Asian

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              Hodges & Kim

                                                                                                                                                                                                              2000

                                                                                                                                                                                                              1,460

                                                                                                                                                                                                              CAFAS

                                                                                                                                                                                                              Regression

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Jung & Rawana

                                                                                                                                                                                                              1999

                                                                                                                                                                                                              250

                                                                                                                                                                                                              MRNAF

                                                                                                                                                                                                              White, Native American

                                                                                                                                                                                                              ANOVA

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              Krysik & LeCroy

                                                                                                                                                                                                              2002

                                                                                                                                                                                                              4,754

                                                                                                                                                                                                              NCCD

                                                                                                                                                                                                              White, Black, Hispanic, Native American

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              Leistico & Salekin

                                                                                                                                                                                                              2003

                                                                                                                                                                                                              126

                                                                                                                                                                                                              RST-I

                                                                                                                                                                                                              White, Black

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              Lodewijks et al.

                                                                                                                                                                                                              2008

                                                                                                                                                                                                              117

                                                                                                                                                                                                              SAVRY

                                                                                                                                                                                                              Dutch

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Marczyk et al.

                                                                                                                                                                                                              2003

                                                                                                                                                                                                              95

                                                                                                                                                                                                              PCL:YV, MAYSI, YLS/CMI

                                                                                                                                                                                                              White, Black, Hispanic

                                                                                                                                                                                                              Regression

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              Marshall et al.

                                                                                                                                                                                                              2005

                                                                                                                                                                                                              94

                                                                                                                                                                                                              PCL: YV,

                                                                                                                                                                                                              YLS/CMIns

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              McCafferty

                                                                                                                                                                                                              2016

                                                                                                                                                                                                              2,608

                                                                                                                                                                                                              OYAS–Disposition Tool

                                                                                                                                                                                                              White, Black

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              McGowan et al.

                                                                                                                                                                                                              2011

                                                                                                                                                                                                              87

                                                                                                                                                                                                              SAVRY

                                                                                                                                                                                                              White, Black, Hispanic, Native American

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              Meyers & Schmidt

                                                                                                                                                                                                              2008

                                                                                                                                                                                                              121

                                                                                                                                                                                                              SAVRY

                                                                                                                                                                                                              White, Native Canadian

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Odgers et al.

                                                                                                                                                                                                              2005

                                                                                                                                                                                                              125

                                                                                                                                                                                                              PCL: YV

                                                                                                                                                                                                              White, Native American, Hispanic

                                                                                                                                                                                                              Regression

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              Olver et al.

                                                                                                                                                                                                              2012

                                                                                                                                                                                                              167

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              White, Canadian Aboriginal

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              Onifade et al.

                                                                                                                                                                                                              2010

                                                                                                                                                                                                              308

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              White, Black, Hispanic, Multiracial

                                                                                                                                                                                                              Regression

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Onifade et al.

                                                                                                                                                                                                              2009

                                                                                                                                                                                                              968

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              White, Black

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Onifade et al.

                                                                                                                                                                                                              2008

                                                                                                                                                                                                              328

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              White, Black, Multiracial, Hispanic

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              Quist & Matshazi

                                                                                                                                                                                                              2000

                                                                                                                                                                                                              35

                                                                                                                                                                                                              CAFAS

                                                                                                                                                                                                              White, Latino, Black

                                                                                                                                                                                                              Chi square

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              Rembert et al.

                                                                                                                                                                                                              2014

                                                                                                                                                                                                              480

                                                                                                                                                                                                              LAC

                                                                                                                                                                                                              Blacka, Hispanic

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              Rennie & Dolan

                                                                                                                                                                                                              2010

                                                                                                                                                                                                              140

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              White

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Risler et al.

                                                                                                                                                                                                              2000

                                                                                                                                                                                                              223

                                                                                                                                                                                                              FORAI

                                                                                                                                                                                                              Black, Native American, Asian

                                                                                                                                                                                                              Regression

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              Schmidt et al.

                                                                                                                                                                                                              2005

                                                                                                                                                                                                              107

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              White, Canadian Native

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Schmidt et al.

                                                                                                                                                                                                              2011

                                                                                                                                                                                                              128

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              PCL: YV SAVRY

                                                                                                                                                                                                              White, Native Canadian

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Schmidt et al.

                                                                                                                                                                                                              2006

                                                                                                                                                                                                              130

                                                                                                                                                                                                              PCL: YV

                                                                                                                                                                                                              White, Native Canadian

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              Schwalbe et al.

                                                                                                                                                                                                              2007

                                                                                                                                                                                                              536

                                                                                                                                                                                                              NCAR

                                                                                                                                                                                                              White, Black

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Schwalbe et al.

                                                                                                                                                                                                              2006

                                                                                                                                                                                                              9,534

                                                                                                                                                                                                              NCAR

                                                                                                                                                                                                              White, Black

                                                                                                                                                                                                              Regression

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              Schwalbe et al.

                                                                                                                                                                                                              2004

                                                                                                                                                                                                              464

                                                                                                                                                                                                              NCAR

                                                                                                                                                                                                              White, Black

                                                                                                                                                                                                              Regression

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Schwalbe

                                                                                                                                                                                                              2009

                                                                                                                                                                                                              9,398

                                                                                                                                                                                                              ARNA

                                                                                                                                                                                                              White, Black, Latino, Native American

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Sharkley et al.

                                                                                                                                                                                                              2003

                                                                                                                                                                                                              159

                                                                                                                                                                                                              OCRS

                                                                                                                                                                                                              White, Mexican American, Filipino, African American

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Shepherd et al.

                                                                                                                                                                                                              2013

                                                                                                                                                                                                              175

                                                                                                                                                                                                              SAVRY

                                                                                                                                                                                                              English-Speaking Background, Culturally and Linguistically Diverse, Indigenous

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Shepherd et al.

                                                                                                                                                                                                              2015

                                                                                                                                                                                                              207

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              English Speaking Background, Culturally and Linguistically Diverse, Indigenous

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              Snowden et al.

                                                                                                                                                                                                              2010

                                                                                                                                                                                                              1,016

                                                                                                                                                                                                              VRAG; HRC-20

                                                                                                                                                                                                              White, Black

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              Stockdale et al.

                                                                                                                                                                                                              2010

                                                                                                                                                                                                              161

                                                                                                                                                                                                              PCL: YV

                                                                                                                                                                                                              White, Canadian Aboriginal

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              ns

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Takahishi et al.

                                                                                                                                                                                                              2013

                                                                                                                                                                                                              405

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              Japanese

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .01

                                                                                                                                                                                                              Thompson & McGrath

                                                                                                                                                                                                              2012

                                                                                                                                                                                                              3,568

                                                                                                                                                                                                              YLS/CMI

                                                                                                                                                                                                              Indigenous Australian, Australian Ethnic

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              Van Der Put et al.

                                                                                                                                                                                                              2012

                                                                                                                                                                                                              520

                                                                                                                                                                                                              WJSCA

                                                                                                                                                                                                              Dutch, Non-Dutch

                                                                                                                                                                                                              AUC

                                                                                                                                                                                                              p < .05

                                                                                                                                                                                                              Vincent et al.

                                                                                                                                                                                                              2011

                                                                                                                                                                                                              480

                                                                                                                                                                                                              SAVRY

                                                                                                                                                                                                              White, Black, Hispanic

                                                                                                                                                                                                              Regression

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              p < .001

                                                                                                                                                                                                              Note: ns = Nonsignificant findings.

                                                                                                                                                                                                              (——) = Data or information concerning predictive validity was not reported in research studies for overall, majority, or minority group.

                                                                                                                                                                                                              N/A = Not applicable.

                                                                                                                                                                                                              Note: a The risk assessment was not significant for the minority group(s) examined in the study.