The recidivism rate for sex offenders is much lower than commonly assumed. Denying individual liberty on the basis of community protection requires a defensible mechanism for identifying those sex offenders who are likely to reoffend. Careful attention to empirical literature has the potential to improve risk assessments for sex offenders. The two major types of risk factors include static risk factors such as a history of childhood maladjustment or prior offenses and stable and acute dynamic risk factors that have the potential of changing. A number of historical and highly stable predictor variables have been documented (e.g., offense history and deviant sexual preferences), but the research on dynamic risk factors has been limited. Three plausible approaches to conducting risk assessments are the guided clinical approach, the pure actuarial approach, and the adjusted actuarial approach. Currently, each approach has demonstrated moderate and roughly equivalent predictive accuracy, although further research should enable actuarial approaches substantially to outperform and guide clinical approaches. Each approach can be expected reliably to identify a small subgroup of offenders with an enduring propensity to reoffend sexually. Methods are not available to identify individuals certain to reoffend, but special handling of high-risk offenders can be justified. Figures, table, and 73 references (Author abstract modified)