The subject mortality problems of an experimental design which covers the period of student recruitment through posttesting over a 13-month period, is detailed in this study. Out of an effective applicant pool of 72 students randomly assigned to experimental and control groups, only 38 students remained at the end of the study with complete test data. Over one-fourth of the sample dropped out of school, with another 20 percent not tested due to mobility or unavailability. The ways attrition occurred, attrition time points, categories of attrition identification and the effects of attrition are discussed in light of a detailed analysis of student attrition. Data are analyzed for mortality effects according to a design presented by Jurs (1970). Methods of controlling or reducing attrition are discussed and suggestions for future study are presented. (Author/DEP)