Reliability and Validity of the Level of Service Inventory-Revised Among Federally Incarcerated Sex Offenders

Abstract

Nonsexual criminogenic needs are often overlooked in the assessment and treatment of sex offenders because of the focus and emphasis on issues surrounding sexual arousal. Augmenting sexual arousal information with objective evaluations of nonsexual criminogenic needs offers a comprehensive management and treatment strategy for this type of offender. The Level of Service Inventory-Revised (LSI-R) is an objective risk/need assessment instrument, which has demonstrated considerable promise among juvenile, adult, and female offender groups. Unfortunately, there is no documented research on the LSI-R with sample populations of sex offenders. In this present study, the LSI-R was completed on a representative sample of 216 Canadian, federally incarcerated sex offenders as part of an intake assessment. The study sample was composed of 74 sexual aggressors of adult victims, 54 extrafamilial child molesters, and 88 familial child molesters. Consistent with previous research among nonsexual offenders, the LSI-R displayed acceptable psychometric properties (alpha = .89). Convergent validity was also demonstrated with measures of relevant general criminogenic constructs as well as those specific to sex offenders (i.e., deviant sexual arousal and denial/minimization). Additionally, comparisons between the three offender groups indicated that familial child molesters scored significantly lower on the LSI-R total score as well as on the Criminal History, Education/Employment, Accommodation, Companions, and Attitude/Orientation subcomponents.