How accurately can researchers measure criminal history, sexual deviance, and risk of sexual recidivism from self-report information alone?

Abstract

ABSTRACT Sexual recidivism risk measures are primarily scored using official documentation (e.g. criminal records), but such reviews are time-consuming, and limited by the quality and availability of relevant information. In this study, we examined the agreement between self-reported and official file information. We conducted secondary analyses on two datasets in which 24 and 27 adult males convicted of sexual offences provided self-report information under confidential conditions, which we used to score the Static-99 and the Screening Scale for Pedophilic Interests. Criminal history information was reliable across both studies, whereas victim characteristics were not. We also used self-reports to create a self-report risk scale – the Sexual Offence Self-Report Risk Scale, which was positively correlated with the Static-99 across both studies (r = .73 and .56). Our results suggest that some self-report information gathered under confidential conditions can be reliable and provide acceptably valid estimates of relative risk for research purposes when official documentation is limited.