Survey finds attack ads work, though better on some voters than others


If the early campaign-season barrage of negative advertising is any indicator, the 2012 election is going to be a decidedly uncivil one. According to the Wesleyan Media Project, 70% of advertising in the current presidential campaign has been negative – as compared to just 9% at this stage four years ago.

But just how well do these negative ads work? Historically, political science research has had a hard time uncovering much evidence for their effectiveness. But recent research is more and more finding that they do indeed move voters.

A recent survey by Arizona State professors Kim L. Fridkin and Patrick J. Kenney finds that the more “uncivil” advertisements voters see about a candidate, the more negatively they evaluate that candidate, as long as the ads are also seen as relevant.

Additionally, the researchers find that some voters are more sensitive to attack ads than others, and that attack ads are generally more effective in deflating challengers than incumbents. Their article, “Variability in Citizens’ Reactions to Different Types of Negative Campaigns,” was published last year in the American Journal of Political Science and makes for some fascinating reading.

To assess the impact of negative advertising, the scholars surveyed 1,045 citizens across 21 different 2006 Senate races both before and after the election.

They also had a team of researchers code the ads in those races on two dimensions: civility and relevance. They found that overall, 47% of the ads were “uncivil” and 12% were “irrelevant.” They defined civility as “an explicit use of harsh, shrill, or pejorative adjectives.” (e.g. “After all these years, can’t he offer more than smears and distortions?”) Relevance described whether the ads focused on the candidate’s record (e.g., “The Senator voted to give tax breaks for companies that move overseas”) or not (e.g. “My opponent parties with Playboy playmates”).

In general, the more voters were exposed to ads that were both uncivil and relevant, the more their evaluation of the candidates declined.

“What is impressive,” Fridkin and Kenney write, “is that the relevance and civility of advertising exert a strong impact on candidate evaluations, even controlling for party and ideological proximity.”

The researchers also found that negative press coverage independently reduced the favorability for incumbents, though not as clearly for challengers (presumably because challengers don’t get as much press coverage).

Unsurprisingly, the majority of respondents said they do not like negative ads. Overall, 82% of respondents either agreed somewhat (35%) or strongly (47%) that “some negative advertisements are so nasty that I stop paying attention.”

But some were less bothered by attack ads than others. In particular, the researchers found that the following traits were all independently associated with a higher tolerance for negative political advertisements:

  • strong partisanship (both parties)
  • following campaigns closely
  • conservative political beliefs
  • being a man
  • being young
  • a lack of political sophistication (i.e. inability to adequately place both parties on an ideological scale)

Interestingly, the voters who dislike the ads the most are also the most likely to be swayed by them, while those less bothered are also less likely to be affected. Those who have more tolerance are also less likely to see the ads as irrelevant and uncivil in the first place.

One encouraging sign, however, is that the research did find some support for a backlash effect. “When incumbents stray away from relevant messages and produce and disseminate irrelevant and uncivil messages,” Fridkin and Kenny wrote, “citizens react by lowering their evaluations of these incumbents.” In other words, voters get mad when incumbents unleash irrelevant attacks. So enough may be enough. But then again, they don’t punish challengers as harshly, according to the research.

Then again, with super PACs now around to run negative ads, candidates can worry less about the backlash. Research by Deborah Jordan Brooks and Michael Murov of Dartmouth (which I wrote about last week) finds that attack ads are much more effective when done by independent groups, precisely because they shield candidates from the backlash

If Fridkin and Kenny are correct, then there is one antidote to the effectiveness of attack ads: a bunch of highly partisan, active voters without much political sophistication, especially young conservative males. These individuals appear to have built up a resistance to attack ads. But then again, imagine an electorate filled with them.