RESUMEN
PURPOSE: Contrary to public opinion, empirical studies have consistently shown that persons convicted of a sexual offense (PCSO) are less likely to recidivate with a general offense. While researchers often point toward the surreptitiousness of sexual offending to explain low rates of recidivism, this paper tests a novel explanation: SOs recidivate at lower rates than persons convicted of a non-sexual offense (PCNSO) because they are more often revoked to prison before they are able to commit a new crime, perhaps owing to more restrictive post-release supervision guidelines. METHODS: Using a sample of 196,468 unique male releases, the difference in general and sexual recidivism between PCSO (n = 29,420) and PCNSO was assessed through survival analyses (Cox regression models). RESULTS: Results demonstrated that PCSO were significantly less likely to be reconvicted for a general crime, but more likely for a sex offense. They were also more likely to be reincarcerated due to a revocation without a new sentence. Accounting for revocations, the difference in reconviction risk lessens between the groups but does not disappear. CONCLUSIONS: This analysis provides evidence that differences in community supervision are contributing to the difference in recidivism rates between PCSO and PCNSO. Implications and future research are discussed.