Abstract
Wollert's (2002) paper on actuarial tests that are used for the prediction of sexual recidivism (ATSRs) showed that the recidivism rates for the score groups in the MnSOST-R developmental sample were inflated, and that they “shrank” by as much as 44 percentage points on cross-validation. Doren and Dow (2002) present an analysis of Wollert's (2002) research and an alternative study that they claim demonstrated a lack of shrinkage. This commentary presents a critique of Doren and Dow's response. Various considerations indicate that the original cross-validation was appropriate. Doren and Dow's alternative was also found to be unacceptably flawed. Additional analyses cast doubt on the MnSOST-R's validity coefficients and its relevance for identifying likely recidivists, and a consensus was apparent that the MnSOST-R experience table for the developmental sample should not be used in civil commitment hearings. Finally, a number of guidelines for the development of ATSRs were reviewed. In particular, it was noted that test developers could make a valuable contribution to the field by sharing data about their tests with other professionals when asked to do so.