Do Campaigns Still Have the Power of Persuasion?

New research reinforces the notion that persuasion as a campaign tactic is dying. 

Since 2004, many consultants have warned candidates off trying persuasion tactics on the shrinking slice of the electorate considered up for grabs each cycle. Instead, these practitioners advocate for budgets going into voter identification, grassroots targeting and mobilization, and GOTV. In other words, base turnout. 

Now, California-based researchers have determined that the tactics popularized by the Bush-Cheney campaign still hold water. 

"Most of the time persuasion efforts by campaigns are actually not very effective," said Joshua Kalla, a researcher at the University of California, Berkeley.

Together with Stanford University's David Broockman, the researchers surveyed 40 field experiments and conducted nine of their own with campaigns from the federal level to ballot propositions. The original research was done while serving as unpaid consultants to Working America, an AFL-CIO-affiliated grassroots advocacy organization.

"Average effect, over all these studies, was pretty much zero," Kalla told C&E.  

Some consultants didn’t need new research to tell them that. In an interview with PBS after President Bush's reelection, Matthew Dowd summed up the campaign’s base strategy like this: “[I]ndependents or persuadable voters in the last 20 years had gone from 22 percent of the electorate to 7 percent of the electorate in 2000. And so 93 percent of the electorate in 2000, and what we anticipated –93 or 94 percent in 2004, just looking forward and forecasting –was going to be already decided either for us or against us.

You obviously had to do fairly well among the 6 or 7 [percent], but you could lose the 6 or 7 percent and win the election, which was fairly revolutionary, because everybody up until that time had said, ‘Swing voters, swing voters, swing voters, swing voters, swing voters.’”

Still, Kalla’s research did show that voters were perusable in some circumstances. In the 2008 Oregon Senate campaign, there’s an example of a pro-choice group “correcting” voters who backed then-Sen. Gordon Smith with a mailer informing them the Republican wasn’t in fact pro-choice.

That same sort of policy-education persuasion worked with a canvass of black voters in a North Carolina during a recent gubernatorial race (white voters reacted negatively to the same material), according to the research paper, recently published in American Political Science Review. 

From these 49 studies, the bulk of them were mailers and door-to-door canvassing, only a handful looked at TV and digital. Moreover, the campaigns and partner groups did the creative and targeting while the academics did the analysis.

Kalla said a takeaway is for candidates is to be weary of consultants’ claims of closing a large polling gap.

“Clearly voters change their minds, but if a consultant comes to you and says, 'we're going to persuade 20 percent of the electorate to vote for you using this TV ad or this mailer,’ I think people should be skeptical," he said. "When it comes to thinking about campaign budgets and tactics and win number, have realistic expectations."

There are ways to persuade people, he added. Kalla encouraged consultants to keep ”doing research and development and learning what campaigns can from political psychology, marketing, social psychology, from different kinds of tactics, ideas and approaches that they currently might not be using. 

“If your starting point is, 'hey, this might not be working,' then they might be open to more of that risk taking and development," he said.