Abstract

We determined the cut-off score for the Korean version of BDI in order to establish empirical and clinical criteria for screening clinical depressives from the populations with two different prior probabilities of clinical depression. To this end, BDI was administered to 89 patients who were diagnosed by psychiatrists and clinical psychologists as depression and 272 normal adults aged from 20 to 45. The cut-off score assuming the prior probability of .50 was turned out to be sixteen, which was determined by randomely selecting 89 normal subjects and analysing their BDI scores along with the scores of the 89 patients by a probit analysis. The cut-off score assuming the prior probability of .30 was twenty-two, which was determined by subjecting the BDI scores of the entire subjects to the same analysis. The cut-off score of sixteen under the assumption of .50 prior probability classified 71.6% of the depressives and 77.3% of the normals correctly. That of twenty-two under the assumption of .30 prior probability classfied 60.2% of the depressives and 94.8% of the normals correctly. The situations where each of the two cut-off scores would be appropriate or not were discussed.