In 2008, a trio of scholars found the location of a polling place can affect how people vote. Specifically, they discovered that in the 2000 Arizona election, voters who cast their ballots in schools were more likely to support a school-funding initiative.
Those researchers raised the troubling question of whether casting ballots in churches — the most common polling location in America, according to the American Humanist Association — might alter voters’ decisions on social-values issues.
Newly published research provides evidence that it does.
“Polling locations can exert a powerful and precise influence on political attitudes and decision-making,” psychologist Abraham Rutchick of California State University, Northridge, writes in the journal Political Psychology. He describes five studies that, together, “suggest that the use of churches as polling places could be advantageous to politically conservative candidates and to supporters of conservative positions on abortion, same-sex marriage and other relevant issues.”
In one study, Rutchick examined the 2004 election in South Carolina’s Sixth Congressional District between Democratic incumbent James Clyburn and Republican challenger Gary McLeod. After controlling for party identification, he found that McLeod received 41 percent of the votes cast at churches, compared to 32 percent at secular locations. (Thirty-five churches were used as polling places, compared to 371 secular locations.)
A second South Carolina study examined the 2006 general election, which featured several proposed amendments to the state constitution. Rutchick compared voting results in churches and non-church polling places on two of the amendments. One sought to define marriage as between a man and a woman; another, more secular proposal involved restricting the state’s power of eminent domain.
After controlling for race, gender, age, party affiliation and the per capita number of churches in the county, he found that the definition-of-marriage amendment was supported by 83 percent of people voting in houses of worship compared to 81.5 percent of those voting elsewhere. Support for the eminent domain amendment was virtually identical in the two locations: 91.8 percent vs. 91.5 percent.
In a related field experiment, Rutchick and his colleagues asked 77 college undergraduates to play the role of insurance adjuster and evaluate two compensation claims: one for the use of an “abortion pill,” the other for a work-related injury. Half performed the task in the atrium of a nondenominational university chapel; the others did so in the lobby of an academic building on the same campus.
Those in the chapel gave significantly less money to the abortion-pill defendant than those in the academic building. In contrast, they gave more money to the workman’s comp defendant than those in the secular location, suggesting the chapel may have evoked general feelings of charity along with negative emotions regarding abortion. A subsequent study found that the effect was only present for Christians; non-Christian participants were unaffected by where the task was performed.
“These findings suggest that voting and other political decisions are subject to transient and contextual influences to a greater degree than is accounted for by many conceptualizations of voting behavior,” Rutchick concludes. “The expression of political attitudes, like the expression of other attitudes, depends in significant part on the environment in which the attitude is expressed.”
“Police stations, for instance, might activate respect for authority,” he writes. “Fire departments might activate helping norms.”
Perhaps, with the increased use of voting by mail, this issue will eventually become moot. Nevertheless, given the growing evidence that subtle stimuli can affect our attitudes, it’s odd to consider how seriously we take the separation of church and state, and how little thought we give to the partition of polls and pews.