National Geographic News: NATIONALGEOGRAPHIC.COM/NEWS
 

 

Voter Decision Affected by Polling Place, Study Finds

John Roach
for National Geographic News
June 23, 2008
 
Politicos take note: Where people are assigned to vote can influence how they vote.

Researchers in a new study found voters were more likely to approve a school-funding initiative if they were assigned to cast their ballot at a school.

Numerically the effect was small: In the precincts analyzed, the initiative earned 2 percent more votes at schools than it did from voters assigned to churches, community centers, and other locations.

But "it's big in [the] sense that it would be big enough to tip the scales in a close election," said study leader Jonah Berger, an assistant professor of marketing at the University of Pennsylvania in Philadelphia.

For example, in 2000, George W. Bush defeated Al Gore by 0.009 percent of the votes officially counted in Florida to win the U.S. presidency.

The research appears online today in the journal Proceedings of the National Academy of Sciences.

Behavioral Priming

The study supports the psychological concept known as "behavioral priming"—that simple stimuli in the environment can influence judgment and behavior.

In a classic example, students shown a picture of a library subsequently spoke quieter than students shown a control image.

In another study, people left a lab slower if they had been exposed to words such as "bingo" and "Florida" rather than to controls.

Presumably the subjects associated bingo and Florida with often slower seniors, experts say.

"Similarly, voting in a school may activate school relevant norms that one should support public education," study co-author Marc Meredith of the Massachusetts Institute of Technology said in an email.

(Related: "'Competent' Face Helps Win Elections, Study Suggests" [June 9, 2005])

School Influence

Berger, Meredith, and colleague Christian Wheeler at Stanford University analyzed data from the 2000 Arizona general election.

Statewide about 54 percent of voters approved an education-funding initiative to increase the state sales tax to 5.6 percent from 5 percent. The extra money boosted teacher salaries, among other things.

The researchers looked at the approximately 80 percent of the precincts where they could obtain data on the type of polling location and how people voted.

They found that 56.02 percent of the people who voted at schools supported the initiative whereas 53.99 percent approved the initiative in non-school locations.

A statistically significant difference held up when controlled for factors such as political views and demographics.

In a second experiment, subjects exposed to images of schools, such as rows of lockers and classrooms, were more likely to support a school-funding initiative than people exposed to control pictures such as office parks.

In post-experiment interviews, none of the subjects believed the pictures had anything to do with their decisions.

In the paper, Berger and colleagues write that "these findings illustrate that consequential real-world decisions can be influenced by subtle environmental features even outside awareness."

Jaime Molera, a political consultant with The Molera Alvarez Group in Phoenix, Arizona, was the campaign director for the 2000 education initiative. He agreed school polling stations may have helped win the election.

"But it also helped that we had over U.S. $2 million to help sell it in our state," he added.

Election Impact?

According to co-author Meredith, election officials should refrain from hasty decisions based on the study.

Further experiments are required to see if the same type of effect occurs in churches and other settings. Besides, he added, all voting systems have "unappealing consequences."

Recent studies have shown, for example, that voter turnout hinges on how close the polling stations are to their homes. Anecdotal evidence also suggests parking availability at stations affects turnout.

According to Berger, using multipurpose rooms devoid of religious symbols and student-related items should help minimize any effect of the voters' surroundings.

"Given that this could influence a close election, I think it's worth paying more attention to these types of issues and making sure that manipulation by interested parties isn't going on," he said.

Molera, the political consultant, said campaign strategists may take note of this study as they search the landscape for advantages.

"You try to give yourself as many edges as possible," he said. "But at the same time, I think the bottom line is having the resources to get your message out."
 

© 1996-2008 National Geographic Society. All rights reserved.