Illuminating the dark figure of sexual recidivism

Detected or reported ("observed") rates of sexual reoffending have long been recognized as underestimating the occurrence of actual sexual recidivism. Past attempts to bridge the gap between the two rates have been unsuccessful. Scurich and John try to reverse this course by presenting a simulation model to estimate the predicted actual sexual recidivism rates among individuals convicted of sexual offenses based on three parameters; they also apply these data to calibrate the sexual recidivism rates from four sexual recidivism studies. The accuracy of the predicted actual sexual recidivism rates is wholly dependent upon the reliability of the inputs to the model. This analysis relies upon scientific studies and literature to delve into the precision of the parameters of Scurich and John in relation to the accuracy of their predicted actual sexual recidivism rates and the validity of the calibration process. The results reveal that some of the assumptions by Scurich and John about the parameters are supported empirically, while others are not. Overall, the simulation model parameters suffer from significant deficiencies that likely produce inaccurate predicted actual sexual recidivism rates. Moreover, the methodologies of the comparison studies used in the calibration process do not actually meet the requirements of the analytic strategy of Scurich and John, which effectively invalidates their findings. Until computational strategies are employed that account for linear and nonlinear effects of model parameters, closing the gap between observed and actual sexual recidivism rates will remain elusive.