Decomposing Trends in Attitudes Toward Gay Marriage, 1988–2006*

Objective. The objective of this article is to examine the trend in attitudes toward gay marriage through the analysis of data from the General Social Survey. Methods. Using linear decomposition techniques, I explain the change in attitudes toward gay marriage from 1988 to 2006. Results. Attitudes significantly liberalized over time; 71 percent opposed gay marriage in 1988, but by 2006, this figure dropped to 52 percent. Approximately two-thirds of this change was due to an intracohort change effect, or individuals' modifying their views over time, and one-third was due to a cohort succession effect, or later cohorts replacing earlier ones. This pattern was replicated across many subgroups of the U.S. public, including age, sex, residential, educational, and religious groups. Conclusion. The results suggest that the use of the “equality/tolerance” framing of gay marriage by its supporters and other societal events or “moments” may have convinced some people who used to disapprove of gay marriage in 1988 to approve of it by 2006.