Measuring Same-Sex Couples: The What, Who, and How of Misreporting on Relationship and Sex

Jamie M. Lewis, U.S. Census Bureau
Nancy Bates, U.S. Census Bureau
Matthew Streeter, U.S. Census Bureau

Prior research on same-sex couples has identified relative group size as a key issue in obtaining accurate estimates. If a small percentage of opposite-sex couples is misreported as same-sex, the impact on estimates of same-sex couples is substantial. Despite the Census Bureau’s development of a revised relationship item, inconsistency in reports of couples’ relationship versus sex persists. Here, we investigate relationship-sex inconsistency in the 2013 American Housing Survey. Results suggest inconsistencies are due more to errors on the relationship item than sex. Both inconsistent reports and mismarks on relationship are most common for couples whose relationship was reported as same-sex married, followed by same-sex unmarried, opposite-sex unmarried, and opposite-sex married couples. Older respondents and larger households are associated with mismarks on relationship, and returning households are less likely to have sex misreported. We suggest some strategies, such as educating interviewers about relative group size, to improve estimates of same-sex couples.