Advertisement

SKIP ADVERTISEMENT

Gray Matter

In Iowa, Voting Science at Work

Todd Rogers and

Credit...Oscar Bolton Green

OF the two winners of the Iowa caucuses, who’s the better behavioral scientist, Ted Cruz or Hillary Clinton? To judge from their campaigns’ respective “get out the vote” efforts, both politicians seem to have studied up on recent research in the field.

Let’s start with Mr. Cruz. His campaign sent a mailer to the homes of Iowans pressuring them to show up to their caucus locations. The mailer noted “low expected voter turnout” in their area, gave them a grade for their past voting participation and disclosed the grades that the campaign had assigned to the recipients’ neighbors.

This strategy was reminiscent of an influential field experiment involving more than 340,000 citizens that was published in 2008 in the American Political Science Review. The political scientists Alan Gerber, Donald Green and Christopher Larimer sent letters reminding the recipients of their own voting histories, informing them of their neighbors’ voting histories and indicating that their neighbors had received similar mailings and would learn in the future whether the recipients voted. The study found that the letters increased voting rates more than any other “get out the vote” tactic previously developed. Political consultants have employed variants of the strategy ever since.

But whereas the mailings in the 2008 study exerted a kind of neighborhood peer pressure, Mr. Cruz’s mailers added an alarmist and misleading message: a warning of a potential “voting violation.” Iowa’s secretary of state, Paul D. Pate, called the mailer a “false representation of an official act” that was “not in keeping in the spirit of the Iowa caucuses.” Professor Larimer, one of the authors of the 2008 study, has speculated that because of the negative tone of Mr. Cruz’s mailer, it could “elicit a negative response.”

Mrs. Clinton’s campaign, by contrast, appeared to hew more closely to proven behavioral science research. For one thing, her campaign’s “call script” (the prepared text for phone solicitations) asked Iowans if the location of their polling site was “close enough to walk” or if they would “have to drive,” and included information about when the caucus location opened and closed, effectively prompting them to devise a voting plan.

One of us (Professor Rogers) and the political scientist David Nickerson researched this tactic in a study published in Psychological Science in 2010. Our field experiment, involving more than 287,000 Pennsylvanians, revealed that asking people what time they would vote, where they would be coming from and how they would get to their polling place more than doubled the impact of typical “get out the vote” messages.

In addition, the call scripts that Mrs. Clinton’s campaign used in Iowa used phrases like “being a voter” repeatedly, as opposed to merely referring to “voting.” In 2011, one of us (Professor Rogers), with the psychologists Christopher Bryan, Carol Dweck and Gregory Walton, published research in the Proceedings of the National Academy of Sciences showing that using “noun phrasing” that emphasized a civic identity (e.g., “be a voter”) could increase voter turnout compared with “verb phrasing” that emphasized a specific action (e.g., “vote”).

The Clinton call scripts also noted that public records showed that recipients had voted in past elections, and thanked them for their civic participation. This sort of expression of gratitude, combined with a reminder that who votes is public record, was shown to increase voter turnout in research published in 2011 by the political scientist Costas Panagopoulos in The Journal of Politics.

Finally, the Clinton campaign’s call scripts told Iowans that “a lot of people” would be caucusing this year. This strategy has been studied by one of us (Professor Rogers) and Professor Gerber. The call scripts in our field experiments had a consistent structure: They included a (true) message conveying the impression of either high expected turnout or of low expected turnout. The results, published in 2009 in The Journal of Politics, showed that citizens were more motivated to vote when hearing about high expected turnout. (This is notable since, all else being equal, your vote objectively matters less when turnout is higher than when turnout is lower.)

Likewise, Mrs. Clinton’s campaign’s assertion that caucus attendance would be relatively high was true — and the behavioral science research suggests it was also motivating.

Our democracy is stronger when more citizens participate in our elections. Both of the Iowa caucuses’ winning campaigns tried to increase participation. Mr. Cruz’s efforts may have benefited from a refresher course in behavioral science (and perhaps in ethics as well), but Mrs. Clinton’s reflected a solid understanding of the science of human behavior.

Todd Rogers is an associate professor of public policy at the Harvard Kennedy School of Government, where Adan Acevedo is a research fellow.

A version of this article appears in print on  , Section SR, Page 12 of the National edition with the headline: In Iowa, Voting Science at Work. Order Reprints | Today’s Paper | Subscribe

Advertisement

SKIP ADVERTISEMENT