Rational Discretionary Risk: The Judicial Risk x Sentence Model for Sex Offenders

Abstract

Given that sex crimes are considered to be dangerous offenses and are perceived to have a high risk of reoffending, sex offenders are subject to disproportionality harsh treatment in criminal institutions. Sentence disparity among sex offenders has been argued to increase as a result of judicial clinical judgment on risk assessment. Even though judicial discretion theory indicates that judicial focal concerns about individual risk would affect sentencing outcomes, there is little known about how judges “weight” risk factors within judicial discretion. The current study integrates a sentencing guideline framework and the utility of actuarial risk assessment instruments (e.g., RRASOR, Static-99R, MnSOST-R, SONAR) and constructs a judicial risk x sentence model (JRS) for sex offenders to address a research gap. This study attempts to answer (a) whether judicially imposed sentence lengths vary by risk level, (b) whether risk classification explains the differences between imposed sentence lengths and advisory sentence lengths, and (c) whether the third generation of risk assessment tools would better predict judicial discretion than the second generation of risk assessment instruments. The JRS model, a statistically-based approach for modeling “risk” into judicial discretion, demonstrated that judicially imposed sentence lengths for sex offenders vary by level of risk when the model controlled for legal- and extra-legal factors. Judges were more likely to depart from sentencing guidelines when sex offenders were assessed as high risk. Moreover, findings indicate that judicial discretion as a first generation risk assessment tool is the least reliable method of assessing individual risk and the least likely way to reduce sentence disparity compared to second and third generation risk assessment instrument criteria. It is worthy to note that the current study, in fact, revealed that judges would be thinking in ways similar to third generation risk assessment instruments when they “weight” risk in sentencing decisions. Even though the JRS model is a novel and exploratory approach, it could serve as an evidence-based proxy for judicial discretion in risk decision-making and better fulfill the doctrine of sentencing proportionality.