How Science Can Help Get Out the Vote
Research offers several proved strategies for boosting turnout on Election Day
Only about half of the people who could vote in the 2012 U.S. presidential election actually did so (53.6 percent of the voting-age population). This puts turnout in the U.S. among the worst in developed countries. By way of contrast, 87.2 percent of Belgians, 80.5 percent of Australians and 73.1 percent of Finns voted in their last elections. In a nation quick to defend democracy both within its borders and beyond, why are more Americans not exercising what is arguably their biggest democratic right?
Certainly there are political and mechanical obstacles within the American voting climate that make it difficult for people to even get to the polls, such as onerous voter ID laws or a shortage of polling stations in some locales. The absence of automatic voter registration (as in Finland) or mandatory registration (as in Australia) also limits turnout.
But beyond these structural hurdles, most theories that examine the mindset of those who do not vote speak to disengagement from electoral politics or disbelief in government’s ability to affect progress. Solutions that aim to address these problems typically inform people about the importance of their vote in electing a government that works for them. Yet this tactic does not appear to sway many. Despite such efforts, turnout has consistently hovered around 50 percent for the past nine U.S. presidential elections—the highest being 56.9 percent in 2008.
Behavioral science might explain why these informational interventions fall short. A substantive body of evidence indicates that the environment in which we make decisions can fundamentally alter them. For example, what we think others are doing, how voting makes us feel about ourselves, and what we need to do to vote all affect whether or not we participate on Election Day. So instead of simply telling Americans to vote, the science suggests we need to think about the context in which citizens decide to cast their ballots.
Always Have a Plan
A number of traditional mobilization efforts are directed at getting citizens to agree they will vote come election time. But just as many of us intend to exercise, eat healthy and save for retirement, people often fail to act on their intentions. As a 2015 review by researchers at Harvard University and the University of Pennsylvania concluded, making concrete plans can help people translate goals into actions across a number of domains.
In a field experiment conducted among 287,000 would-be voters in Pennsylvania during the 2008 Democratic primary election, researchers tried to see if voter turnout could be increased by helping people make a concrete plan to implement their intentions. One to three days before the November 2008 election, behavioral scientists David Nickerson, now at Temple University, and Todd Rogers of Harvard asked one group of would-be voters about their intentions to vote and a second group about their intentions and also about when, where and how they would accomplish the goal of voting.
Voter records showed that making a plan was more than twice as effective as simply asking people about their intentions. Overall there was a 4.1 percentage point increase in the likelihood of voting by making a plan relative to people who did not receive a phone call. (The average effectiveness of commercial phone banks, assessed from dozens of studies, is about one percentage point.)
Everyone Else Is Doing It
Conventional wisdom (and practice) suggests that we could convince people to vote by stressing that their particular ballot is very important because not many other people are voting. Yet findings in behavioral science indicate that most of us are motivated by the desire to conform to the social norm—meaning we are more likely to do what most people are doing.
Two get-out-the-vote field experiments during the 2005 general election in New Jersey and the 2006 primary election in California tested these hypotheses. They found that individuals were much more motivated to vote when they believed lots of other people were voting compared with when they thought relatively few others were voting.
In another field experiment run by researchers at Yale University and the University of Northern Iowa during the 2006 primary election in Michigan, potential voters received direct mail noting that both they and their neighbors would be informed of who had voted after the election. Amazingly, this led to an 8.1 percent increase in turnout—one of the most successful get-out-the-vote tactics studied to date. Conventional direct-mail reminders, in contrast, yield just a 0.162 percent increase in turnout on average, according to a 2013 estimate based on 110 studies.
If most of us vote, then being part of the truant few who do not feels like we are shirking a social contract. Publicizing voting records may therefore increase the salience of this social obligation and possibly bring shame on nonvoters. Following through, however, allows them to maintain their self-identity as contributing members of society.
All about Identity
Some of the largest-ever experimental effects on voter turnout come from an experiment that used people’s desire to shape or conform to a worthy self-identity, that is, the identity of “someone who votes.”
In a study published in 2011, psychologists at Stanford University and Harvard presented would-be voters in the 2008 presidential election in California and in the 2009 gubernatorial election in New Jersey with a preelection survey that framed voting as either an expression of self-identity (“How important is it to you to be a voter?”) or simply an activity (“How important is it to you to vote?”). In both cases, participants completed the survey the day before or the morning of the election.
Being “a voter,” one might argue, is about who you are as an upstanding citizen—a part of your identity that feels good to embrace and act on. The act of voting is simply that, an action, and one that anyone can, in principle, take. The results showed a remarkable 10.9 percentage point increase in turnout among people in the “voter” identity condition.
Such an increase nationally could have historic consequences. Indeed, it would bring American voter turnout up to 64.5 percent—ahead of both Canada and the U.K., lifting the nation from 31st to 19th place out of 34 developed countries in a Pew Research Center analysis.
To Vote or Not to Vote
Although tackling political barriers to voting remains critical, the great strength of these behavioral interventions lies in their ability to overwhelm obstacles by catalyzing citizen motivation. And for people who do not vote because they believe one person’s ballot cannot change election outcomes, behavioral science also offers a reason why voting is important for individuals.
Research has found that in addition to signaling who we are to others, our actions tell us something about ourselves—shaping our own preferences and beliefs. From this perspective, people who do not vote are not merely abstaining from the democratic process in one instance. They are also “telling” themselves: “I don’t care about politics.” Moving forward, they may also become less interested in civic rights, local governance, foreign affairs, and so on. And for those who do vote, participation is not just an expression of interest in current politics but also a seed that could grow into an active political life.