The Importance of Reassessment: How Changes in the LSI-R Risk Score Can Improve the Prediction of Recidivism

Abstract

The Level of Service Inventory-Revised (LSI-R) is a correctional assessment tool that identifies the individual risk and needs of criminal offenders. The predictive validity of the LSI-R has the support of numerous studies, large samples, and multiple meta-analyses. The dynamic nature of many of the items in the LSI-R suggests that offenders are capable of altering their likelihood for future reoffending. However, despite the hundreds of studies conducted on the LSI-R, there have been only two to investigate whether or not changes in LSI-R score were empirically related to recidivism (Raynor, 2007; Vose, Lowenkamp, Smith, & Cullen, 2009). The current study attempts to add to this literature by assessing the predictive and dynamic validity of the LSI-R. Furthermore, this study is the first to investigate the relationship between changes in LSI-R domain scores and recidivism. The results support the LSI-R as a valid instrument in predicting recidivism and suggest future research on risk/needs assessments should include examinations of the percentage change in risk scores.