Predicting Sex Offender Recidivism

• In Event: Advances in Risk Assessment and Prediction

Abstract

The main goal of the proposed study is to investigate whether nontraditional statistical tools better predict sex offenders’ recidivism. Doing so, traditional and nontraditional methods will be compared, and the performances will be evaluated based on related specificity and sensitivity. Using official sexual offending data, this study aims to seek ways to achieve a more accurate prediction through a comparative framework.

Authors

• Turgut Ozkan, Arkansas State University Jonesboro

• Stephen Clipper, University of Alabama

• Alex R. Piquero, University of Texas at Dallas

• Michael Baglivio, G4S Youth Services, LLC

• Kevin Wolff, The Graduate Center, John Jay College / CUNY