ABSTRACT
Social scientists have long recognized women as being both more religious and more liberal than men. Indeed, men have been disaffiliating from churches and the Democratic Party at higher rates than women for several decades. Moreover, we show evidence to suggest that churchgoing women, while less liberal than women in the general population, are nevertheless more liberal than churchgoing men. Given these trends, one might expect churches to have become more liberal in response to these changing demographics. However, using nationally representative data of American congregations and their adherents, we show that churches with a higher percentage of female congregants are actually more conservative. We suggest that the institutional structure of churches limits the ideological influence and contributions of women within the church community. Specifically, women are often excluded formally or informally from decision-making leadership positions in the church. When women are allowed greater access to leadership positions, however, we find that congregations do tend to be less politically conservative. These findings suggest that churches and other civic organizations, which are major direct and indirect political actors, may be insulated from changes in the partisan or ideological makeup of their membership through their allocation of important leadership roles.
Acknowledgements
The authors would like to thank Jim Guth, Mike Wagner, Gary Hollibaugh, participants at the Rooney Center for the Study of American Democracy, the editors, and anonymous reviewers for their helpful feedback on this research.
Disclosure statement
No potential conflict of interest was reported by the authors.
ORCID
Andre P. Audette http://orcid.org/0000-0001-7071-2693
Notes
1 Democrats include all respondents on a 7-point party identification scale who identified either as strong Democrats, weak Democrats, or as leaning toward the Democratic Party.
2 Although there are breakaway sects of the Catholic Church that ordain women (i.e., the Roman Catholic woman priests), the Vatican does not recognize these ordinations as legitimate.
3 A number of denominations invited their congregants to participate in oversamples for their religious tradition. These data are weighted to account for the oversampling of adherents from those traditions.
4 Significant zero coefficients are a result of rounding.
5 This model is not included in the article, as it applies only to Catholic attenders.