Over the past half-century, voter turnout in the United States has generally declined. As political pundits decry our flagging sense of civic duty and campaign managers scramble to turn out the vote, many psychologists also are trying to figure out why some people vote and others don't--and how to get more of the electorate to the polls.
"Watching and understanding how and why people vote is an important way to monitor the civic health of our country," says social and political psychologist Jon Krosnick, PhD, of Stanford University.
Many factors can help predict the "how and why" of voting, says Krosnick, who will soon become co-principal investigator of the National Election Studies (NES), a 50-year-old research project that surveys voters before and after every election to follow trends in American politics.
The predicting factors, he says, fall into two categories: stable attributes and factors specific to a particular election. Stable attributes include things like basic demographic factors. On the whole, for example, older people are more likely to vote than younger people, and people with more education are more likely to vote than those with less education.
The election-specific factors, Krosnick adds, include how--and how strongly--a voter feels about the candidates in a particular race. In one study, for example, published in the American Journal of Political Science in 2001, he and his colleagues analyzed NES data from the 1972 through 1996 presidential elections and found, among other things, that vehemently disliking at least one candidate is a strong predictor of turnout.
"If you like both candidates, you'll be happy either way," Krosnick says--and thus less likely to vote at all.
One election-specific variable that could conceivably affect this month's election is the perceived closeness of the race. In our swing-state-centered times, voters in all but a handful of prime locations, such as Florida or Missouri, might be inclined to think that election results in their states are more or less predetermined.
But does that mean they'll be less likely to show up at the polls? Answers to this question are mixed: The same 2001 study by Krosnick and his colleagues found that whether a person believed that an election would be close did not predict their likelihood of voting. But other studies by psychologists and political scientists have found that election closeness is related to voter turnout, including one study of Canadian elections by political scientist James Endersby, PhD, and his colleagues in the May 2002 Journal of Politics (Vol. 64, No. 2).
Finally, convenience plays a role too. For example, a study by Henry Brady, PhD, a political scientist at the University of California, Berkeley, has found that people who live near polling stations are more likely to vote than those who live farther away--even a quarter-mile can make a difference.
Getting out the vote
Other psychologists, meanwhile, are less interested in understanding exactly why people do or don't vote, and more interested in figuring out how to persuade them to get to the polls.
For example, an interdisciplinary team of researchers at the University of Minnesota's Center for Political Psychology, led by psychologist Mark Snyder, PhD, and political scientist John Sullivan, PhD, conducted a study that piggybacked on a "natural experiment" run by the youth movement Rock the Vote during the 1996 presidential election.
In December 1995, Rock the Vote began asking newly registered voters to fill out pledge cards and then mailed the cards back to the voters two weeks before election day in 1996. At first, the cards simply asked voters to sign a pledge that they would "rock the system by exercising my right to vote on November 5, 1996." Partway through the campaign, however, Rock the Vote staff added a sentence to the cards that read "I will vote because _____" and allowed potential voters to fill in their own reasons.
The researchers tracked down the cards' recipients, and found that people who received the personalized pledge cards were significantly more likely to vote than people who received the generic ones.
The researchers suggest that the psychological concept of attitudinal advocacy--the idea that making a written or oral commitment increases the likelihood a person will actually perform a behavior--could underlie their results. They suggest that people who received the card with the sentence prompt might have felt a stronger sense of commitment than those who received only the generic card.
Cognitive dissonance theory--the idea that people try to avoid having inconsistent or dissonant thoughts--could also play a role, the researchers say.
Other psychologists have also found that attitudinal advocacy can help motivate potential voters. On the eve of the 1984 presidential election, social psychologist Anthony Greenwald, PhD, now at the University of Washington but then at Ohio State University, and his colleagues called 60 Ohio State students who were registered to vote in the local precinct. All the students were reminded of the election the next day, but half were also asked whether they intended to vote--and all of those asked said yes. Greenwald and his colleagues found that the students who had been asked whether they would vote were 25 percent more likely to vote than those who had not been asked.
Meanwhile, other research suggests that any kind of personal, authentic communication and encouragement can help induce people to vote. Donald Green, PhD, a political scientist at Yale University who often collaborates with psychologists and has published research in the Journal of Personality and Social Psychology, found in a 2003 study of six door-to-door canvassing experiments that nonpartisan students and community members who met face-to-face with potential voters were successful in increasing the voter turnout in the neighborhoods they visited. He's also found that impersonal methods like e-mail, leaflets and perfunctory phone calls are not effective.
"People are not foolish," he says, "so quality matters. Personal, authentic communication with voters does make them more likely to vote."
Letters to the Editor
- Send us a letter