Search :
Winter 1997

Research Magazine > ARCHIVE > Spring 96 > Article

by Jennifer T. Daly

They slung so much mud in the New Hampshire primary, some candidates took to wearing mud boots. "Before I'm through," said 1996 presidential hopeful Lamar Alexander, "I might need a pair of waders."

Such antics may move voters to reach for an antacid, but for a pair of University of Georgia advertising/public relations professors, it simply promises to be another exciting season of research.

Since 1982, Spencer Tinkham and Ruth Ann Lariscy have compiled one of the nation's most extensive studies of U.S. congressional campaigns, probing the sometimes not-so-gentle art of persuasion from the perspective of the candidates, the consultants and the voters.

And especially in the past few years, their research -- along with the rhetoric of the politicians themselves -- has focused on the negative.

"I think a lot of people would like us to find that, yes, negative advertising is a horrible thing and it's turning people away from their beliefs in government," Lariscy said. "But that's not the case. We've found that negative ads actually heighten interest and make people talk about politics even more."

Why candidates use negative ads isn't so difficult to explain: They are the recurrent weapons of winners. The focus of Tinkham and Lariscy's research is on how they work. And the most basic reason may simply be that they are memorable.

"If you get 10 compliments and then one person says really insulting things to you, what are you going to remember?" Lariscy said.

But that apparent simplicity belies the researchers' findings that negative ads are "cognitively more complex" than positive ads. Candidates whose messages are exclusively positive limit the criteria from which potential voters can judge them.

"The funny thing about it is you needn't say anything at all about yourself if you're attacking someone else," Tinkham said. "Conclusions are still reached about both the sponsor and the opponent"

The research shows that you can make yourself look better by making the other candidate look bad, but the reverse isn't true: You can't make the opponent look bad by making yourself look good in positive ads.

"If a candidate runs a positive ad that says, "I'm a Vietnam War veteran and therefore I will be the best person to represent you in Congress," the candidate typically believes that people will infer that his opponent is not a Vietnam veteran," Lariscy said.

"Many consultants believe this practice works," Tinkham said. "But in our study, we found positive ads simply do not work in the same way negative ads do."

Successful Strategies

By studying the reactions of more than 200 viewers of 10 test ads -- seven negative, three positive -- Tinkham and Lariscy have identified some principles of successful negative ads. Among them:

The news must be new. For example, in 1992, attacks on Bill Clinton regarding alleged sexual indiscretions were relatively ineffective because they were perceived as old news. "Old dirt is just that -- old dirt," Lariscy said. "But if people do not know about it, even if it happened a long time ago -- say that he smoked marijuana 20 years ago and it is just now revealed -- that would be considered new."

The message must be believable. "If a candidate places an attack that is clearly off the wall, it is not going to work," Lariscy said.

The ad should excite and entertain. While a candidate's stand on the deficit or health care may influence the election through debates or town meetings, the study shows ads that dwell on serious issues have minimal voter impact. However, the persuasive power of an attack is enhanced significantly if the ad is exciting and entertaining.

"We have found a strong, positive relationship between the attitude toward the ad and candidate preference," Tinkham said. "However, we also have found that simply measuring whether the ad is liked does not give a complete picture of the commercial's performance."

For example, the researchers caution that the message can get lost in the delivery. An exceptionally entertaining attack ad can, in effect, become independent of its sponsor: The voter may remember the ad, but may not remember the candidate.

And regardless of the entertainment quotient, there are several forms of attacks that voters rarely embrace. Some topics appear taboo, the researchers said.

"It happens every now and then, but usually ads attacking an opponent's children, wife or husband, sex life, etc. are not done," Tinkham said.

A case in point is the 1992 advertising exchange between Rep. Newt Gingrich, R-Ga., and challenger Tony Center, which was so harsh that -- even during an exceptionally negative campaign season -- it captured the attention of Newsweek's George Will.

Will recounts how Gingrich's campaign fired off this ad: "Meet Tony Center, a trial lawyer . . . Center actually sued to strip a 4-year-old and a 1-year-old of their child support to pay his own legal fee."

And from the Center camp: "It was Newt who delivered divorce papers to his wife the day after her cancer operation."

"It has to be fair. It can be nasty if it's real and fair," said Tinkham. "Going after a candidate's mother because she's in an insane asylum would be over the line."

Humor is another ticklish area. Most of the political consultants in Tinkham and Lariscy's survey advocated the use of humor; however, the experimental research shows that humor carries an inherent risk.

For instance, in the lab, participants were asked their opinions of a 1984 television ad in which challenger Mitch McConnell uses a pack of bloodhounds to illustrate a search through various vacation spots for incumbent Sen. Dee Huddleston, D-Ky., who McConnell charged was earning $50,000 in speaker's fees on the lecture circuit when he should have been in Congress.

Of the 10 ads viewed by the study group, the humorous bloodhound spot posted the lowest score -- even though, in the real world, McConnell won. Did the ad lead to the win? Tinkham and Lariscy's lab results suggest McConnell won in spite of the ad.

In fact, they have found that so many variables factor into a win or loss that it's difficult to measure the exact impact of attack ads. A negative campaign could fail, for example, because the candidate, in desperation, decided to "go negative" to turn around an already losing campaign.

"If both candidates go negative it levels the playing field a little bit," Lariscy said. "We do know that if one decides to go negative and the other doesn't, the one who doesn't usually ends up very damaged."

The study dispels what is known as the "boomerang" effect, the theory that voters will punish the candidate who goes on the offensive with an attack ad.

"Our research shows that while voters may be somewhat less positive toward a candidate who uses a credible negative ad, the net effect of its use will be positive," Lariscy said.

In Defense of Attacks

One of the most definitive findings from the study is that the candidate who is attacked must defend against the ad. Both laboratory studies and case histories have brought Tinkham and Lariscy to this conclusion.

"Negative information does not exist in isolation," Tinkham said. "If the opponent just lets it go under the assumption that nobody will ever believe this, the odds are he or she will lose."

"Candidates sometimes feel they are taking the high road -- you know, I'm not going to lower myself into the gutter and even respond to that kind of attack,'" Lariscy said. "But no defense is a losing strategy. We found that to not respond is very damaging, very dangerous."

Failure to respond now bears the portentous moniker the "Dukakis effect," which refers to the 1988 presidential campaign in which Michael Dukakis, to a large extent, failed to respond to attacks from opponent George Bush -- and lost handily.

"We believe that was one of the primary factors that hurt Dukakis," Tinkham said.

Early results from Tinkham and Lariscy's current research show that an effective defense should occur as quickly as possible after the attack.

For example, they point to a 1986 ad by Senate candidate Henson Moore of Louisiana chastising his opponent, Rep. John Breaux, D-La., for missing 1,083 votes in the U.S. House. Within 24 hours, Breaux responded with an ad -- mimicking Moore's attack ad -- that concluded "1,083 is the number of jobs lost in Louisiana every 10 working days due to Republican policies sponsored by Henson Moore.

This quick, creative response won over study participants and also may have contributed to Breaux's success at the polls.

The most effective response to an attack ad is made in the same medium where the attack occurred -- television to television, newspaper to newspaper.

"In local races, candidates often respond to a negative broadcast ad with a long endorsement ad in the newspaper," Lariscy said. "You know, one of those ads where 500 people sign their names. Our examination suggests that this is not really appropriate. If you're attacked on the radio, you should respond on the radio."

"If for no other reason than to reach the same audience the attack reached," Tinkham added. "Because audiences are different in different media."

The research also shows that too aggressive a counter-attack actually may be a weak defense.

"When candidates lash back as a form of rebuttal, it is much less effective than a rebuttal that is assertive that says, this is what my opponent said, and this is why it's not true,'" Tinkham said.

Beating an opponent to the punch -- called "inoculation" by political pundits -- is a relatively recent development in campaigns. Early research results show that while this approach reduces the power of future attacks, a post-attack defense is more effective. Candidate surveys on the same subject show politicians view the preemptive strike as a risky defense.

The Power of Money

The 1992 presidential candidacy of Ross Perot and Steve Forbes' bid in 1996 focused attention on the ability of big campaign coffers to sway an election. Tinkham and Lariscy's research confirms the importance of deep pockets -- especially in congressional races where the financial playing field typically isn't level between incumbents and challengers.

"In this respect, the incumbent is at a great advantage over the challenger," Lariscy said. In fact, in data from the 1990 congressional elections, Tinkham and Lariscy found that the power of incumbency, by itself, gives an approximate 16 percentage-point advantage over challengers.

But their study also found that money and incumbency, as in the cases of many losing Democratic incumbents in the 1994 congressional elections, are not everything. A thoughtful campaign strategy can give an under-funded challenger the edge against an incumbent or stronger financed campaign. For example, "using research wisely helps shape message strategy and direct more efficient spending of available dollars," Tinkham said.

Traditional campaign activities -- making personal appearances, distributing posters and canvassing neighborhoods with yard signs -- can also be effective tools. They work, not only for under-funded challengers, but also for incumbents and open-race candidates, the researchers said.

"In general, interpersonal communication has more impact than mass communication," Tinkham said. "It's inherently less efficient on a cost-per-thousand-reached basis, but there is nothing more persuasive than looking someone in the eye and asking them for their vote, contribution or volunteerism."

Surrogates also prove effective at the grass-roots level. These are typically relatives, friends or campaign workers who blanket communities and meet individual voters face-to-face in place of the candidate.

Despite the success of traditional campaign activities, Tinkham and Lariscy have recorded a significant decline in this form of campaigning. "We hate to see it die because it's very powerful," Lariscy said.

One of the common mistakes of an under-funded campaign is to compensate for lower advertising budgets by seeking out radio talk shows or news interviews for free air time. But Tinkham and Lariscy have found this approach to be one of the greatest risks a candidate can take because the candidate ends up with very little control, if any, over the message.

"Many of them rush to it because it's cheap, but it's very dangerous," Lariscy said. "How many interviewers will let the candidates know the questions in advance?"

What To Expect

Recent elections have brought some especially interesting changes for the researchers to study. Now, for example, many incumbent candidates are launching attack ads first -- a tremendous change from only a few years ago.

"It used to be that negativity was the bailiwick of the challenger trying to unseat the all-powerful incumbent," Lariscy said. "Now everyone is doing it."

The 1996 election season also has introduced some new forms of old campaign strategies. Instead of meeting voters door-to-door, many candidates are opting for a home-to-home page approach.

"We think interactive media might stimulate interpersonal communication because it allows the voter to respond and to have input into the process," Tinkham said.

"Of course, the problem with that is that such a small percentage of people currently use the World Wide Web," Lariscy said.

Direct mail advertisements are a current topic of extensive study by the researchers. Between 1982 and 1992, the use of direct mail in congressional campaigns jumped a dramatic 300 percent, they said. Candidates have embraced the approach because it allows them to tailor messages, especially attack messages, for a specific zip code area.

"Direct mail is quite negative because candidates can feel free to attack selectively and to target particular negative arguments to different voter segments," Tinkham said.

The researchers especially are interested in which medium carries the message most effectively. So far, television has led the way in negative campaign advertising. However, the tremendous increase in direct mail mud-slinging provides interesting comparative research, the professors said.

The duo also is considering a new challenge after the 1996 election data is collected. State legislative races appeal to them because an extensive research project, on the level of their congressional study, has not been done at the state level, they said.

While many Americans would prefer the flu over sifting through hours of negative ads, candidate statements and consultant opinions, Tinkham and Lariscy have made these often "muddy" studies their lives' work.

"A lot of times people turn on the TV and they say, Oh no, it's another political season,'" Lariscy said. "I know it's hard to believe, but we actually thrive on it."

For more information please e-mail rlariscy@uga.cc.uga.edu  or stinkham@uga.cc.uga.edu


Return to Spring 1996 Index

Research Communications, Office of the VP for Research, UGA
For comments or for information please e-mail the editor: rcomm@uga.edu
To contact the webmaster please email: ovprweb@uga.edu