Published online 23 June 2008 | Nature | doi:10.1038/news.2008.908

News

Where you vote affects how you vote

The location of the polls could sway an election.

polling placeVoting in a school might make voters favour education-friendly politicians.KEVIN LAMARQUE/Reuters/Corbis

In November 2004, Christian Wheeler stood in line at a local church and waited to cast his ballot in the US presidential election, which pitted President George W. Bush against the democratic candidate, John Kerry. As Wheeler waited, his thoughts began to wander.

“I was thinking about the election and how Bush was highly affiliated with religion,” says Wheeler, “and it occurred to me that this church couldn’t possibly be a neutral location. This has to be affecting people’s thoughts.”

So Wheeler, a professor of marketing at Stanford University in California, decided to study whether the location of a polling station can influence how people vote. The results, published this week in the Proceedings of the National Academy of Sciences1, suggest that his early musings may have been correct: in an Arizona election, those who voted in schools were slightly more likely to support a proposition to increase funding for education.

The results add further detail to the already complex picture of the influences on voters' behaviour. “We used to think that voting was a simple process where people gathered information about a candidate and made a decision before they got to the voting booth,” says Jon Krosnick, a social scientist at Stanford University who was not affiliated with this study. “We now know that’s not true. Some arrive at the booth conflicted, or haven’t thought it through. They confront that ballot and need to make a decision with very little time and information.”

This leaves voters more vulnerable to being swayed by their surroundings. Election planners often try to minimize outside influences by shuffling the order in which choices are presented on the ballots, and requiring campaigners to remain a specified distance away from the polling place.

But Wheeler’s findings suggest that more subtle environmental cues can also influence voter behaviour.

Tough choices

In the United States, elections are not always as simple as selecting candidates for public office. Voters may also be asked to weigh propositions that can range from simple social issues to complex funding decisions. In some states, voters receive election guides that can be well over 100 pages long, detailing the text of the propositions as well as arguments for and against each one.

Wheeler and his colleagues decided to study the 2000 general election results from Arizona. In addition to selecting the president and other elected officials, voters also cast ballots for several propositions. One proposed to increase the rate of sales tax to raise money for education.

The researchers found that 56% of people who were assigned by their local election commission to vote in a school supported the education initiative. But among those who did not vote in schools, 54% voted in favour of the education proposal.

That difference shrank to 1% when other mitigating factors, such as whether voters lived near schools, were taken into account. It's a small difference, notes Krosnick, but could still be significant. “In our current elections, this type of effect is certainly big enough to reverse an election outcome,” he says.

Primed response

In a separate experiment, the researchers enrolled 327 participants and asked them to view a series of pictures and then vote on a proposal that was identical to the Arizona education initiative. Those who had seen pictures of schools beforehand were more likely to support the school funding initiative.

Further analysis showed that certain participants – such as those who were parents – were more likely to vote for the school initiative regardless of whether they had been ‘primed’ with pictures of schools. But the difference between parents and non-parents disappeared if non-parents were shown images of schools ahead of time.

ADVERTISEMENT

Despite these effects, no one in the study thought the images had affected their vote.

Although he feels the study is valuable, Krosnick struggles to come to terms with the simplicity of the results. Not everyone has a good association with schools, he notes. “Lots of people have bad experiences with schools,” he says. “and that might make them a little less favorable to the initiative.” It will be important to extend the work to additional elections in other states, he says, to determine just how universal the response is.

“If it is true, it would be an amazing insight into the political process,” says Krosnick. “It would be a mechanism by which nefarious people could influence election outcomes.” 

  • References

    1. Berger, J., Meredith, M. & Wheeler, S. C. Proc. Natl Acad. Sci. USA doi:10.1073/pnas.0711988105 (2008).
Commenting is now closed.