Sexual Histories: Does It Matter Who Is Asking?

Deven Hamilton, University of Washington

Objective: To compare basic data on sexual behavior gathered in 9 different U.S. surveys, all of which used nationally representative samples. Methods: Descriptive analysis and permutation tests are used to assess whether studies have significantly different estimates for the number of partners reported, controlling for the race, age, sex and marital status composition of the sample. Topcoding, year of data collection and survey administration were investigated as potential contributors to differences in reporting. Results: Studies showed significant differences in reported numbers of partners within demographic subgroups. Effects were smaller in magnitude than those associated with the demographic attributes themselves. There were no significant effects of study year. Study design factors that significantly influence responses include top-coding and mode of administration. In contrast to previous research findings, face to face interviews generated higher estimates of sexual partners than either telephone surveys or SAQs. This trend held for both males and females.