Press, Polls, and the 1988 Campaign: An Insider's Critique

Abstract

On October 12, 1988, four weeks before the presidential election, Peter Jennings, the anchor of ABC's "World News Tonight," looked solemnly into the studio camera-and through it to millions of living rooms across the nation-to announce that this night's broadcast was about to do something that "has not been done before." Having grabbed the audience's attention with the promise that something unprecedented was about to happen, Jennings went on to reveal... the results of the latest ABC/Washington Post presidential poll. As the press secretary for the campaign of Michael Dukakis, watching the newscast from the hotel suite in Los Angeles where we, the campaign workers, were preparing for the final debate the next evening, I understandably had more than a passing interest in what one of the major networks was reporting. But I think it's fair to say that even those Americans who were more interested in the World Series than the presidential election would have agreed with my initial reaction to Jennings' hyperbolic buildup to his newscast: Devoting campaign coverage to an opinion poll was hardly unprecedented; on the contrary, throughout the long campaign of 1988 it seemed that poll reporting was campaign reporting. Polls were everywhere. You couldn't escape them. Every time you turned on a television set, listened to a radio or opened a newspaper, someone had a poll telling you, in effect, that you had already voted. Everyone in the media business-from the tiny radio station in Iowa that asked people to flush their toilets when they heard the name of their favorite candidate land thus measured support by the amount of water flow at the public works plant, with an undetermined "margin of error" caused by residents who were simply answering the call of nature at the wrong moment to such giants as CBS, NBC, ABC, CNN and major metropolitan papers-was in the poll business in 1988. In focusing on yet another poll, ABC's report that night in October, then, was not something that "had not been done before." (CBS, for instance, was reporting the results of its own poll the same evening.! Quite the opposite, it was more of the same, part of a growing pattern of a surfeit of polls in campaign coverage. But it was noteworthy in one respect: it took the media's obsession with political polls to a new level. For the first twelve-and-a-half minutes more than half the entire broadcast when you add the commercial breaks-the only "news" was the poll. It showed that George Bush was leading Dukakis 51 percent to 45 percent in the popular vote, but in a state-by-state breakdown Bush had "firm" leads in 21 states (to Dukakis' 3) and "leaning" leads in 15 more (to four for Dukakis). A map of the United States was prominently shown, displaying Bush's states in shades of red and Dukakis' in blue-the same visual image we have become accustomed to seeing on election nights. There were reporting segments exploring various aspects of the poll, explanations of how the electoral college works, analyses from political experts on what Bush had done right and Dukakis had done wrong up to this point, and then reactions from workers of the two campaigns on what they thought about the poll results. I am not a pollster or even an expert on polling, and this is not a story to complain about the accuracy of polls, but it is worth noting that in various campaign postmortems a number of respected pollsters have questioned this particular poll's methodology and accuracy. I raise the incident instead to make a larger point about campaign coverage, because I think it is a telling example of one of four trends in campaign reporting moving in the wrong direction. The principal problem with ABC's coverage of its poll that night was the excessive attention the network gave to it: more than half that night's news. Think for a second. What other event in the world would get half of a broadcast devoted exclusively to it? The death of U.S. Marines from a terrorist bomb in Beirut? Elections in the Soviet Union? A new cure for cancer? If Bush had admitted he had personally delivered arms profits to the contras or Dukakis had revealed he was secretly planning a major increase in the income tax, would that have gotten as much time? Maybe so. But in this case, nothing had actually happened-except that a network had gone to considerable expense to conduct its own opinion poll and like every other news outlet during the campaign, differing only in the extreme degree of its presentation, decided that this was the most important news it had to offer.