Juvenile risk assessments are standardized rating instruments that measure criminogenic risk in court-involved youth. Juvenile court practitioners use scores from risk assessments to inform judicial decisions throughout case processing. It is critically important that risk scores accurately reflect court-involved youths' latent level of criminogenic risk; both artificially high and low scores incur significant detriments to youths, courts, and communities. In light of the consequences of risk misevaluation, there is urgent need to develop and evaluate alternate juvenile risk assessment measurement models. The current study aspired to improve measurement of criminogenic risk through the development of a Novel Scoring Algorithm which innovated upon current juvenile risk assessment scoring twofold: (1) it adjusted the weights of assessment items and domain sub-scores to reflect their correlation with latent constructs of criminogenic risk; and (2) it integrated the mitigating impact of prosocial protective factors into cumulative risk scores. Drawing upon a sample of 559 youth who entered the supervision of a county-level juvenile circuit court for the first time, the Novel Scoring Algorithm outperformed the current method of scoring (i.e., summing all unweighted risk factors) in both absolute and relative model fit. However, the Novel Scoring Algorithm yielded no incremental improvement in diagnostic accuracy, affirming the Scoring-as-Usual method as an acceptable procedure for assessing likelihood of recidivism in court-involved youth. Implications for effectively and equitably managing risk are discussed.
Read
