Abstract
Using data from the 1984-2014 General Social Survey, we employ cross classified random effects models to examine age-period-cohort changes in political tolerance by religion, incorporating religious belief, belonging, and behavior into the analyses. The results show large cohort and period based changes in political tolerance of the various religious traditions. The results also suggest that the primary driver of changes over time in political tolerance is best attributed to period effects, not cohort replacement, although there is some evidence of a cohort effect among Roman Catholics. Finally, our results demonstrate that aggregate data analyses can lead to noticeably different results and thus can mask substantial differences between groups.
Get full access to this article
View all access options for this article.
