A Simple Model to Understand Gender Discrepancies in Sexual Behavior Reports

Taryn Dinkelman, University of Michigan
David Lam, University of Michigan

This paper formalizes the intuition that in closed, heterosexual populations without misreporting or sampling bias, average number of sex partners reported by men and women should balance. With some simple examples and key parameter values for the fraction of sex workers and number of sex worker clients, we use a simple model to demonstrate that sampling bias could account for the range of gender gaps in reported number of partners. To understand a second stylized fact – that men report condom use more often than women do – we present a related model which depends on additional parameters. We show that a gender gap in condom use persists when (a) some individuals have multiple partners, (b) some individuals have most recent sex with a non-regular partner and (c) condom use differs across partner types. Our simple model highlights a new explanation for observed gender differences in reports of condom use.