After a brief overview of empirical data, the relative threat to the validity of prevalence estimates posed by fabrication versus nondisclosure is weighed. Then various methodological choices and their relationship to the magnitude of prevalence estimates are examined. Addressed are the definitions underlying the studies, the questions used to elicit reports of rape, the context in which rape questioning occurred, the confidentiality of the responses, the method of data collection, and the sample integrity. The author recommends a clear conceptualization of rape as the basis for any research on its scope. If men and boys are to be included in incidence studies, care must be taken to ensure that their data are accurate counterparts of rape prevalence among women. Questions about a range of sexually assaultive experiences are legitimate; however, the methodology should allow for separate calculation of legally defined, completed, and attempted rape to facilitate comparison across studies. Other recommendations are offered for the design of studies of rape prevalence. 1 table and 74 references