Fredrick Butcher and Jeff M. Kretschmar - February 4, 2020
As the juvenile justice system has evolved, the use of statistics and data to inform decision-making have become commonplace. Risk assessment in the juvenile justice system is widely used across the United States to predict whether a youth is likely to recidivate.
Typically, risk assessments are structured interviews conducted by court personnel (e.g. probation officers) that provide a score to indicate a youth’s risk of reoffending. Risk assessments are generally grounded in the Risk Needs Responsivity model of offender rehabilitation, which argues for justice systems to match treatment programming to address criminogenic needs according to an offender’s risk to reoffend.
Depending on the jurisdiction, information from risk assessments influence decisions on the sentence a youth may receive, whether to formally process or divert and eligibility for programming, among others. Much of the scientific literature concludes that risk assessments reliably predict recidivism.
However, does reliably predicting recidivism mean that decisions are made equitably and that all youth, regardless of their race, have the same opportunities for diversion and other alternatives to detention? There are some reasons to doubt that risk assessments provide for equitable decision-making.
Risk assessments themselves are nothing new. Juvenile justice personnel have used interviews and file reviews to make judgements on a youth’s likelihood to recidivate for as long as the modern juvenile justice system has existed. These first generation risk assessments depended solely on the professional judgment of decision-makers.
As risk assessments evolved, an increasing emphasis was placed on empirical evidence. Further, newer risk assessments have been designed to take into account both static (e.g. criminal history) and dynamic (e.g. antisocial relationships) factors that predict justice-related outcomes. Most recent risk assessments are designed to be administered multiple times to measure change in risk and to guide placement in treatment programs.
The guiding principle behind the use of risk assessment is that youth who are at higher levels of risk should be provided with more intensive supervision and programming. Furthermore, intensive programming for low-risk youth may have the unintended consequence of increasing their risk for recidivism, in part due to their association with their higher-risk counterparts.
The ultimate goal of risk assessments is to provide objective measures of the likelihood of a youth to recidivate, which can then help judges and prosecutors make decisions based on data and evidence which, as it stands to reason, will reduce bias. On its face, this seems like a logical step to take.
However, consider that risk assessments are mostly developed using criminal justice data to predict recidivism. Recidivism, as indicated by rearrest or some other type of criminal justice system involvement, is dependent upon the biases inherent in the system. Research has shown that there is disproportionate minority contact at every point in the justice system, which places minority youth at greater risk for arrest, formal court processing, intensive supervision and incarceration.
Given that research has also shown that involvement in the juvenile justice system is, in and of itself, criminogenic, it is no surprise that minority youth are at greater risk of recidivism. As risk assessments are designed to use data to best predict recidivism, they are likely to reflect systematic inequalities. In other words, minority youth are more likely to be classified as high risk. In some cases, this may be because the youth is indeed at a high risk to reoffend. In other cases, minority youth are at a higher risk of misclassification simply due to structural realities that have very little to do with the individual.
In our research, there is some preliminary evidence to illustrate these concerns around racial equity in risk assessments. For example, in a sample of justice-involved youth who have behavioral health issues, we found that a significantly larger proportion of black youth compared to white youth were identified as high risk to recidivate. While the proportion of youth who were identified as moderate risk was nearly identical for both white and black youth, significant differences existed in the low and high risk categories.
These findings are not necessarily conclusive, but they do raise red flags about whether there is a racial disparity in how youth are classified by risk assessments. The risk assessment that we examined contains questions about a youth’s juvenile justice history, their associations with delinquent peers, whether they were suspended or expelled from school, whether their friends or families are associated with gang activity and their own associations with gangs, among others. There is potential racial bias in each of these questions, and further research into identifying the problematic questions is necessary.
The consequences of incorrect classification have tremendous implications at every stage of the system. In many states and jurisdictions, risk assessments have been used to determine an individual’s eligibility for diversion. Diversion occurs at various points of the justice system including at the front end, before formal processing and at the deeper end, where youth may be diverted from out-of-home placements and incarceration.
Risk assessments also significantly impact the likelihood of pretrial detention, which in turn, has a significant impact on the outcome of cases for youth identified as high risk. Jurisdictions may also use risk assessments to determine the intensity of probation supervision. This may include closer levels of monitoring, including how often a youth must meet with a probation officer or submit to a drug test.
Generally, failure rates of probation increase with the level and intensity of monitoring. Similarly, the dosage, length and intensity of treatment programming increases for youth at higher risk levels, which also has implications for failure rates. Furthermore, misclassification can have a significant impact, as there is evidence to suggest that placing youth who are at lower levels of risk with high-risk youth can increase their likelihood of recidivism.
To read more, click here.