METHODOLOGICAL ISSUES IN THE ANALYSIS OF HUMAN SPERM CONCENTRATION DATA

We examined two methodological issues in the analysis of sperm concent ration data using a large database of sperm concentrations in healthy men that were collected at the University of Washington. We showed tha t the raw data were skewed and that log transformation should be used to assure that the data meet the assumptions underlying most statistic al estimation and testing procedures. We also addressed the issue of t he great variability in sperm concentrations within a single individua l and the necessity and utility of multiple sampling to reduce varianc e. We conclude that log-transformed data should be used for statistica l analysis of sperm concentration and recommend that such analyses be based on the geometric mean of several samples from each subject to re duce variability, increase accuracy of estimation, and improve statist ical power. This is particularly important when the objective is to de tect small but important differences or subtle effects.