Every four years, presidential candidates join us -- regularly and repeatedly -- in our living rooms for a quick chat. They pop in, 30 seconds at a time, for virtual campaign stops in the commercial breaks of our favorite TV shows, amid the evening news, even during time outs of sporting events.
Think back, and notable campaign advertisements that have flickered across television sets may come to mind: the uncompromising Willie Horton ad that helped sink Michael Dukakis in 1988; Ronald Reagan’s nostalgic and optimistic “Morning in America” spot from 1984; and LBJ’s haunting “Daisy Girl” commercial from 20 years earlier, with its ominous countdown and nuclear blast.
But how, on the whole, have presidential campaign ads changed in their six-decade run on the airwaves? Have they addressed different issues over time? Have they begun to represent a more diverse audience? Are they more forward-looking, or reflective? Do party differences matter in issue choices and messages? And what might these trends portend for the 2012 presidential race and beyond?
To find answers to those questions, a team of UNL researchers has examined the evolution of national presidential campaign ads from 1952 to the present, analyzing, categorizing and coding hundreds of the general-election commercials over the last 60 years. Now, with an eye on the 2012 presidential election, they’re opening that work up to the public.
The University of Nebraska Campaign Ads Project has assembled a vigorous dataset that notes a number of consistent ad characteristics: visual images, myths, emotion, evidence, issues, demographics, and others. The group -- Dana Griffin, assistant professor of political science; Damien Smith Pfister, assistant professor of communication studies; Marty Nader, a political science Ph.D candidate; and Jessy Ohl, a Ph.D. student in communication studies -- recently published the entire dataset online.
“Our hope is that scholars across the country can draw on this dataset to further the study of political communication from an interdisciplinary vantage point,” Pfister said. “Building a dataset has long been thought of as merely a prerequisite to research, but scholarly norms now recognize that the process of constructing datasets itself is a valuable mode of research.”
Early observations of the years of data have unveiled some interesting tendencies. Project researchers recently presented a content analysis at the Midwest Political Science Association conference in Chicago showing that, over time, presidential candidates have relied on more retrospective than prospective appeals in their advertisements. Candidates also are increasingly likely to attribute blame to past events than they are to take credit, and to forecast grimly what would happen if the opponent were to win rather than describe what they themselves would do if elected.
Also, candidates are decreasing the frequency with which they take positions on issues in ads, the researchers found. Instead, they’re progressively using campaign spots as a space to attribute policy positions to their opponent.
“Historically, in any given ad, voters ran about a 60 percent chance of encountering information about a candidates’ own issue positions and about a 30 percent chance of seeing issue positions attributed to the opponent,” Griffin said. “This pattern has changed significantly in the last decade, with 2008 being the first time that position-attribution outpaced position-taking.
“This rise of ‘other-centered’ campaign ads represent an important development in American politics, and it remains to be seen what impact this has on voters,” she said.
Another early analysis from the research found that for the most part, both Democrats and Republicans have tended to talk about the same issues with roughly the same proportions. The issues on which the parties differ in their attentiveness are social programs, health care, education and employment. Democratic candidates talk about those issues more, but Republicans still give them considerable attention – differing in levels of attentiveness, which varies by election cycle.
In addition to being a source for scholarly research and publishing, the group expects the dataset to provide some perspective in this year’s campaign between President Obama and presumptive Republican nominee Mitt Romney.
The project weblog will analyze long-term campaign ad trends in the context of the 2012 election cycle and will serve as an interactive forum to discuss this year’s ads within a larger historical framework.
“When observers ask whether the substantive content, argumentative technique, visual style or other elements of an ad go beyond those used in previous campaigns, we'll be able to say whether it’s been done before or not,” Griffin said.