Women can keep the vote: no evidence that hormonal changes during the menstrual cycle impact political and religious beliefs.
Recently, Durante, Rae, and Griskevicius (2013) reported that women’s menstrual cycle phase affected their religiosity, voting preferences, social (but not fiscal) political attitudes, and preferences for U.S. presidential candidates. The direction of these effects seemingly depended on women’s relationship status. The authors argued that during the fertile phase of the menstrual cycle (relative to the infertile phase), women in committed relationships become more religious and socially conservative, which causes them to shift toward preferring the more conservative presidential candidate. In contrast, women who are not in committed relationships reportedly show the opposite effect; namely, during peak fertility, single women shift to being less religious and more socially liberal, and therefore prefer the more liberal presidential candidate.Durante et al. suggest that “because ovulation might lead married women to become more sexually interested in men who are not their partner, and because it is especially costly for such women to cheat on their partner, increased religiosity and conservatism might function to decrease the likelihood of behaviors that might harm the relationship” (p. 1009). This reasoning contrasts with most evolutionary psychologists’ theorizing, which generally contends that pair-bonded females are more likely to engage in extrapair sex during the peak fertility phase in order to acquire better genes for offspring (e.g., Penton-Voak et al., 1999).We attempted to directly replicate the findings of Durante et al. Assessing the robustness of their findings seems warranted for several reasons. First, the findings depart strikingly from common-sense ways of thinking about political and religious behavior, implying markedly greater fickleness in women’s attitudes relative to those of men—something that, to our knowledge, has not been noted by pollsters and political scientists. Second, there is great variability among menstrual cycle studies on preferences in how fertility is calculated (e.g., there is variability in the number of days categorized as fertile and as infertile, the specific days counted in each category, and which days are excluded altogether) and in what moderators are examined. Such inconsistency potentially introduces flexibility into analytic methods, which endangers replicability (Harris, Chabot, & Mickes, 2013; Pashler & Wagenmakers, 2012; Simmons, Nelson, & Simonsohn, 2011).
IN - Psychol Sci. 2014 May 1;25(5):1150-2. PMID: 24570262Psychol Sci. 2014 May 1;25(5):1147-9. doi: 10.1177/0956797613520236. Epub 2014 Feb 25.
1University of California, San Diego.