Static-99 Recidivism Tables (Generated from Logistic Regression) 5 Year Sexual Recidivism (%) Static-99 Score 0 1 2 3 4 5 6 7 8 9 10+ Total

N* Routine CSC Samples 2.3 3.2 4.3 5.7 7.7 10.2 13.4 17.4 22.3 28.2 34.9 752 Preselected High Risk Samples 8.3 10.3 12.8 15.7 19.1 23.1 27.7 32.7 38.2 44.0 50.0 1,163 10 Year Sexual Recidivism (%) Routine CSC Samples 1.8 2.6 3.9 5.7 8.2 11.8 16.7 23.0 30.8 39.8 49.7 342 Preselected High Risk Samples 13.0 15.8 19.1 23.0 27.3 32.1 37.3 42.8 48.5 54.3 59.9 735

*N is the total sample size used in the logistic regression analysis to generate predicted recidivism values. It is not the sample size with a particular Static-99 score. This is because logistic regression uses information on the relationship between Static-99 and recidivism in the complete dataset to generate predicted values. Note: Some of the 10 year CSC rates are lower than the 5 year rates due to sampling error (not all of the offenders in the 5 year sample were followed for the full 10 years).

Static-99 Recidivism Tables (Generated from Logistic Regression) 5 Year Violent Recidivism (%) Static-99 Score 0 1 2 3 4 5 6 7 8 9 10+ Total N* Routine CSC Samples 8.5 10.8 13.6 17.0 21.1 25.8 31.2 37.1 43.4 50.0 56.6 752 Preselected High Risk Samples 16.5 20.0 24.1 28.6 33.7 39.1 44.9 50.8 56.6 62.3 67.6 1,110 10 Year Violent Recidivism (%) Routine CSC Samples 8.5 11.4 15.1 19.7 25.3 31.8 39.2 47.0 55.1 62.8 70.0 342 Preselected High Risk Samples 25.5 29.5 33.8 38.4 43.2 48.2 53.2 58.1 62.9 67.4 71.7 790

*N is the total sample size used in the logistic regression analysis to generate predicted recidivism values. It is not the sample size with a particular Static-99 score. This is because logistic regression uses information on the relationship between Static-99 and recidivism in the complete dataset to generate predicted values.