Does the LSI-R Have Utility for Sex Offenders?

Abstract

Despite societal perception that sex offenders will repeat their crimes, research indicates these offenders are more likely to be generalists than sex offense-specific offenders. Sex offender-specific legislation has reinforced this erroneous perception while contributing to the excessive labeling of sex offenders as sexual recidivists. Additionally troubling is the lack of research on the efficacy of generalized risk/needs assessments for sex offenders. The present study fills this void by evaluating the adequacy of the Level of Service Inventory-Revised (LSI-R) for use with a sexual offending population. The predictive accuracy of the LSI-R for sexual and nonsexual recidivism outcomes was explored using a sample of 21,298 individuals released from New Jersey correctional facilities from 2004 to 2006. Results indicate that while the LSI-R does not have predictive utility for sexual offenses, it has utility for sex offenders overall. Policy implications of the usefulness of the LSI-R for this offending population are discussed.

In the eye of the beholder? An examination of the inter-rater reliability of the LSI-R and YLS/CMI in a correctional agency

“…For specific domains, 6 of the 10 were under .60 (an AUC of .50 suggest no relationship between the predictor and criterion). Other studies (see Caudy, Durso, & Taxman, 2013;Ostermann & Herrschaft, 2013;Ragusa-Salerno, Ostermann, & Thomas, 2013) also found similarly moderate relationships between recidivism and the LSI-R. Results with the YLS/CMI have also shown mixed results, with some studies showing substantial AUCs (Catchpole & Gretton, 2003;Viljoen, Elkovitch, Scalora, & Ullman, 2009) and others indicating more moderate scores (see Chu, Ng, Fong, & Teoh, 2012;Onifade et al, 2008).…”

Effects of Automating Recidivism Risk Assessment on Reliability, Predictive Validity, and Return on Investment (ROI)

“…For example, in a meta-analysis on the LS instruments, Olver et al (2014) found their predictive performance was significantly worse for U.S. offenders compared with those from other countries, including Canada. Moreover, in nondeveloper LSI-R validation studies in which performance with the area under the curve (AUC) was assessed, a predictive validity metric that is robust across different recidivism base rates (Smith, 1996), the area under the curve (AUC) ranged from a high of 0.51 to 0.74 (Barnoski, 2006;Barnoski and Aos, 2003;Caudy, Durso, and Taxman, 2013;Dahle, 2006;Davidson, 2012;Duwe, 2014;Duwe and Rocque, 2016;Fass, Heilbrun, DeMatteo, and Fretz, 2008;Flores, Lowenkamp, Smith, and Latessa, 2006;Folsom and Atkinson, 2007;Ostermann and Herrschaft, 2013;Ragusa-Salerno, Ostermann, and Thomas, 2013;Vose, Smith, and Cullen, 2013;Watkins, 2011;Zhang, 2014). In these 15 studies, which included more than 50 assessments of predictive validity in four countries outside the United States (Australia, Canada, China, and Germany) and four states within the United States (Iowa, Minnesota, New Jersey, and Washington), the average AUC for the LSI-R was 0.64.…”

Predicting Sex Offender Recidivism: Using the Federal Post Conviction Risk Assessment Instrument to Assess the Likelihood of Recidivism Among Federal Sex Offenders

“…Overall, the literature on using general risk/needs instruments on sex offenders shows that they perform well in predicting nonsexual reoffending behavior; however, their predictive effectiveness tends to degrade when assessing the likelihood that an offender will sexually recidivate (Hanson ; Hanson & Morton‐Bourgon ; Ragusa‐Salerno et al ). Most studies examining this issue have explored using the Level of Service Inventory (LS) scales to predict recidivism for various sex offender populations.…”

“…Most studies examining this issue have explored using the Level of Service Inventory (LS) scales to predict recidivism for various sex offender populations. In an examination of 1,602 sex offenders released from New Jersey correctional facilities, Ragusa‐Salerno et al () found that the Level of Service Inventory‐Revised (LSI‐R) had adequate predictive accuracy for arrests involving violent or other crimes, but had poor predictive utility for sexual rearrests. In an analysis of four different studies, Hanson () demonstrated that tools designed to predict sexual reoffending behavior performed better at sexual recidivism prediction compared to generalized risk assessment tools.…”