Actuarial Assessment of Sex Offender Recidivism Risk: A Validation of the German version of the Static-991

The Static-99 and the RRASOR are actuarial risk assessment tools for evaluating the risk of sexual and violent recidivism in sexual offenders. The Static-99 was developed in 1999 by Karl R. Hanson (Canada) and David Thornton (Great Britain) and is in the mean time regularly used for risk assessment in North America and some countries in Europe. The RRASOR can be described as a predecessor of the Static-99 and was published by Hanson in 1997. At first we translated the revised version of the Static-99 (Harris, Phenix, Hanson & Thornton, 2003) and adapted the instrument and the manual to the forensic context in Germany and Austria (Rettenberger & Eher, 2006). In this retrospective study, interrater reliability and concurrent validity of the RRASOR and of the German adaption of the Static-99 is presented. Furthermore we evaluated the predictive accuracy of the Static-99 and the RRASOR and compared their results. The instruments were validated from file information of Austrian sexuel offenders, who were convicted between 1968 and 2002. Both the Static-99 and the RRASOR had good interrater reliability and concurrent validity. The Static-99 showed good predictive validity for general (r = .41, AUC = .74), sexual (r = .35, AUC = .74) and violent (r = .41, AUC = .76) recidivism, whereas the predictive accuracy of the RRASOR was moderate for general (r = .29, AUC = .66), sexual (r = .30, AUC = .68) and violent (r = .28, AUC = .67) recidivism. The Static-99 exhibited a higher accuracy for the prediction of sexual offender recidivism. Although further validation studies on German-speaking populations of sex offenders are necessary, these results support the utility of the German version of the revised version of the Static-99 in improving risk assessment of sexual offenders.