News flash: Candidates' ads actually match deeds in Congress

October 27, 2008

If you think candidates never keep their promises and will say anything to get elected, you're certainly not alone. And you're not right, either.

The perception is largely untrue, says Tracy Sulkin, a University of Illinois political scientist, who has conducted an extensive study, apparently the first of its kind, comparing the campaign ads and legislative records of recent congressional officeholders.

Candidates' words generally match their deeds, according to Sulkin. The issues candidates say are priorities in their commercials are likely ones they care about and will take action on through the introduction and co-sponsoring of legislation, she said.

Whether they are vague or specific on an issue doesn't matter, Sulkin found. "There turn out to be no differences in subsequent activity among people who just say they care about an issue and people who lay out a specific plan. ... Specificity, which we seem overly concerned about, isn't actually a signal that you care more about the issue," she said. Instead, what drives candidates to be more specific on issues is the closeness of the race.

When a candidate attacks an opponent on a given issue, however, it does not mean the attacking candidate cares about that issue or will act on it, Sulkin's research shows. "Negative appeals, appeals that attack the opponent, don't have much signaling power about what that candidate is going to do," she said.

In another striking conclusion, Sulkin said that being in a "safe" seat, apparently free from challenge by a candidate from the other party, does not seem to produce unresponsive legislators who feel free to do whatever they want.

"One of the things that jumps out of these findings is that if we compare relatively safe people to relatively vulnerable people, the relatively safe people actually seem to follow through on their promises more than the relatively vulnerable people," Sulkin said.

In fact, their promise-keeping appears to add to their safety, she said. "People who keep their promises are rewarded for it."

For her research, Sulkin drew from campaign advertising material collected by the University of Wisconsin Advertising Project, in coordination with the Campaign Media Analysis Group, on the 1998, 2000 and 2002 elections. As the result of their work, she had access to storyboards for all political ads run in the top 75 media markets for the 1998 and 2000 elections, and in the top 100 media markets for the 2002 election.

Those markets cover over three-quarters of congressional districts nationwide, Sulkin said. In all, the research covered ads for 391 winning candidates for the U.S. House and 84 winning candidates for the U.S. Senate over the three election years.

"We know all the ads they aired," she said. "We had a full picture of what their advertising strategies looked like in their televised ads."

Sulkin and her research assistants coded the ads for what they said and how they said it on 18 different issues. Later, the researchers sought out information on the bills that those legislators introduced and co-sponsored during the terms that followed and applied the same coding on the same 18 issues.

They then compared what legislators had said about those issues in their campaigns to what they had done in Congress.

Sulkin used different models to evaluate the results, but found that even using the strictest model, there was a significant difference in the level of activity between those who talked about an issue and those who didn't on 14 of the 18 issues.

Sulkin said she chose to look at what the legislators introduced or co-sponsored, rather than at their votes on final bills, because she thought it better represented the legislators' initiative on issues. Only a small portion of introduced bills ever reach a vote and only about 10 percent pass, she said. Also, votes come at the end of the process, when most legislators have little control over the final product.

Sulkin also wanted to avoid making political judgments about whether the legislation introduced or co-sponsored by legislators was a net positive or negative for that issue. "You can't really separate out whether a vote improved education, for instance, but you can look at whether somebody who said they wanted to improve education actually went to Congress and worked on the issue," Sulkin said.
Sulkin has written a chapter on candidates' promise-keeping for an upcoming edition of the book "Congress Reconsidered," published by CQ Press. She also is working on her own book on the study, "Promises to Keep: The Legislative Legacy of Congressional Campaigns," under advance contract to Cambridge University Press. She published a related paper, drawn from the same research, in the January issues of the Journal of Politics, which showed that the imagery in candidates' ads also has a strong link to their priorities.

University of Illinois at Urbana-Champaign is a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to