Age, risk assessment, and sanctioning: Overestimating the old, underestimating the young.

Abstract

While many extoll the potential contribution of risk assessment to reducing the human and fiscal costs of mass incarceration without increasing crime, others adamantly oppose the incorporation of risk assessment in sanctioning. The principal concern is that any benefits in terms of reduced rates of incarceration achieved through the use of risk assessment will be offset by costs to social justice-which are claimed to be inherent in any risk assessment process that relies on variables for which offenders bear no responsibility, such as race, gender, and age. Previous research has addressed the variables of race and gender. Here, based on a sample of 7,350 federal offenders, we empirically test the predictive fairness of an instrument-the Post Conviction Risk Assessment (PCRA)-that includes the variable of age. We found that the strength of association between PCRA scores and future arrests was similar across younger (i.e., 25 years and younger), middle (i.e., 26-40 years), and older (i.e., 41 years and older) age groups (AUC values .70 or higher). Nevertheless, rates of arrest within each PCRA risk category were consistently lower for older than for younger offenders. Despite its inclusion of age as a risk factor, PCRA scores overestimated rates of recidivism for older offenders and underestimated rates of recidivism for younger offenders. (PsycINFO Database Record

Understanding the Role of Neighborhood Typology and Sociodemographic Characteristics on Time to Recidivism Among Adjudicated Youth

“…In this study, the YLS/CMI juvenile risk assessment was successful in identifying criminogenic risk and capturing the likelihood of future recidivism. Without risk assessment tools, professionals are more likely to rely on arbitrary decision making, which leads to overestimating risk of recidivism (Monahan et al, 2017). Given risk assessments’ ability to identify the small percentage of adjudicated youth who have the greatest likelihood for recidivism, there is potential to divert juveniles out of the systems and/or hone in on intervention needs that appropriately match criminogenic risk areas for adjudicated youth who will likely return.…”

Public safety assessment

“…We use a moderator regression technique commonly cited in psychological studies and testing literature (e.g., Cleary, ; Sackett, Borneman, & Connelly, ). As a result, we can estimate four regression models to assess the extent to which “a given score will have the same meaning regardless of group membership (e.g., an average risk score of X will relate to an average recidivism rate of Y for all relevant [sub] groups)” (Monahan, Skeem, & Lowenkamp, , p. 193). Predictive bias is tested by assessing the extent to which subgroups have similar (i.e., not significantly different) intercepts and slopes (i.e., they possess similar regression lines).…”

“…For example, Monahan et al. () followed this approach with a data set of 7,350, which is much smaller than either of the data sets used here (Lin, Lucas, & Shmueli, ).…”

“…We follow recent practices and set more stringent statistical significance levels at p < .001 as a result of the large sample sizes used for both jurisdictions. For example, Monahan, Skeem, and Lowenkamp () followed this approach with a data set of 7,350, which is much smaller than either data set used here.…”

Assessing Risk in North Dakota Juvenile Probation: A Preliminary Examination of the Predictive Validity of the Youth Assessment and Screening Instrument

“…While the authors’ results indicate high predictive accuracy among this sample of African American and Native American youth, the power of the analyses were insufficient to produce statistically significant outcomes (Obuchowski et al, 2004). Research specific to special populations, including female youth, is still needed (for more on the complicated use of risk assessment with these special populations see Monahan et al, 2017; Shepherd et al, 2013; Skeem et al, 2016; Skeem & Lowenkamp, 2016; Snowden et al, 2010).…”