The secret weapon to winning elections may be in where the votes are cast. The polling locations themselves could be a source of bias.

A paper published Thursday in the International Journal for the Psychology of Religion reports that survey respondents were more likely to express conservative viewpoints, distrust of a variety of ethnic and socioeconomic groups, and profess a belief in God when standing in front of a church.

“Several studies have demonstrated that behavior is affected by external stimuli,” said Jordan LaBouff, the University of Maine professor of psychology who headed the study. “This is the first test of religious stimuli on politics.”

Standing in front of a church, 54 percent of people expressed a certain belief in God. In front of a civic building that number was only 42 percent. The remainder in both instances indicated they were atheist or agnostic. But the shift in answers was not limited to God. Outside a church, survey takers identified themselves as more than 10 percentage points more conservative.

Perhaps most surprisingly of all, survey takers near a church expressed more negative views towards ethnic groups (such as “foreigners,” “Asian” and “African”), people of socio-economic extremes (“rich,” “poor”), “gay men” and “lesbian women” and non-Christian religious groups (“Jewish,” “Muslim”).

“A church environment is one of the most common polling places in the United States,” said LeBouff. “This might tip the scales for a voter in a church polling place on issues of, say, same sex partnership.”

The research was conducted in the Netherlands, at a location chosen for its diverse mix of religious backgrounds. LeBouff noted that surveys taken in America often disproportionately reflect this country’s Protestant majority.

Standing in front of a church seems like a subtle influence to have a major effect, but subtlety is in keeping with other tests on the effects of “priming.” LeBouff noted countless research that showed little things having big impacts.

A 2004 Yale study showed that a room with the mere presence of a briefcase increased feelings of competitiveness, while a backpack promoted cooperation. Another study, conducted by the United Kingdom’s University of Leicester showed that French wine outsold German wine at a five to one ratio when French music was played in the background, while German wine outsold French two to one with German music was played.

LaBouff’s results are particularly troubling when paired with a 2007 study conducted by researchers at The University of Pennsylvania, MIT and Stanford linking voting in a school to voting in favor of policies that benefitted schools. Schools, like churches are common polling places susceptible to bias.

If voting in churches and schools promotes an agenda, how is it constitutional? It might not be, says Jeremy Blumenthal, associate professor at Syracuse University College of Law. The Supreme Court has never tested the issue. Voting in churches has, however, come up in state courts as a violation of the first amendment.

Those Judges have rejected complaints about the constitutionality of church voting centers, emphasizing the opportunity for voters to vote in other ways such as absentee ballots. But in a 2011 paper in the Boston University Law Review, Blumenthal and co-author Terry Turnipspeed argue that the judges may not be giving proper weight to the subconscious nature of the influence.

“On the face of it [the judges' argument] makes sense,” said Blumenthal, but the flipside is that the bias is subconscious. If people don’t know the potential bias, they see have no reason to vote elsewhere.”

The Chicago Board of Elections is yet to assign polling stations for the 2012 elections, but last year they included the St. Demetrios, St. George, St Chrysostom and Cristo Rey churches, Hayt Elementary School, Newberry Math and Science Academy and Mather High School.