Vanishing Voter Poll Reflects Low Interest in Political Party Conventions

Contact: Doug Gavel
Phone: 617-495-1115
Date: July 22, 2004

CAMBRIDGE, MA -- The television audience for the national party conventions has declined in size in recent elections, and preliminary indicators suggest that 2004 will not reverse the trend. In the Kennedy School of Government Shorenstein Center's Vanishing Voter Project national poll of July 14-18, only 31 percent of the respondents said they intended to watch some or most of the upcoming Democratic convention. During the same week of the 2000 campaign, only slightly fewer—28 percent—said they anticipated watching some or most of the upcoming convention.
This finding is surprising in that election interest is much higher this year than it was in 2000. In our recent poll, 50 percent of the respondents said they had paid at least some attention during the past week to the presidential campaign, up from 28 percent during the comparable week of the 2000 campaign.
Heightened election interest is also clear from other indicators. In our recent survey, 46 percent said they had read, seen, or heard a campaign news story within the past 24 hours. In our survey during the same week in 2000, only 28 percent made this claim. Compared with 2000, Americans were also more likely to say (28 percent versus 19 percent) they had talked about the campaign within the past day and to say (63 percent to 30 percent) they had spent time in the past day thinking about the campaign.
Nevertheless, the Democratic convention is a remote part of Americans’ thinking about the election. When the respondents in our national poll were asked when the Democratic convention would be held, 66 percent said they were not sure and 11 percent placed it a month or more away. Only 23 percent said it would be held in the next week or two.
This lack of anticipation undoubtedly owes partly to the fact that the national conventions no longer have a deliberative role. They serve to ratify decisions made in the presidential primaries and caucuses.
Nonetheless, the conventions continue to be significant events. They retain a purpose for which they were invented in the 1830s—the rallying of the party faithful. They also serve a more modern purpose. They boost interest in the campaign and heighten citizens’ understanding of the candidates. In an age of 10-second soundbites and 30-second ads, the conventions stand alone as an opportunity for the public to hear at length why each party and its nominee should be entrusted with the presidency. Studies have found that, aside from the presidential debates, no events do more to interest and inform the voters than do the party conventions.
However, this effect has diminished because of the declining size of the televised convention audience. In 1976, the conventions had an average television rating of 28 points, meaning that 28 percent of all households with a television set were watching the convention during the average minute. The typical household in 1976 watched more than 11 hours of convention coverage. In 2000, the conventions had an average television rating of 15 points—the lowest yet— and the average household watched only 2 hours of coverage.
The 2004 conventions could set a new low. However, two factors—one of which will act to increase the audience and the other of which will serve to depress it—will ultimately decide whether the convention audience in 2004 will fall below that of 2000. The significance of these two factors results from the fact that the overall size of the convention audience is contingent on the number of "inadvertent viewers." Unlike "deliberate viewers" who turn on their television set with the intention of watching a convention, inadvertent viewers are those people who come across the convention coverage while watching television and decide to stay tuned.
The level of election interests affects the number of inadvertent viewers. Viewers who are interested in the election are more likely than the uninterested to watch the televised convention when coming across it while changing channels. Thus, the fact that election interest is higher in 2004 than it was in 2000 bodes well for the size of the convention audience.
However, the fact that the broadcast networks are reducing their convention coverage to 3 hours from the 5 hours provided in 2000 bodes poorly. The size of the inadvertent audience depends on whether the convention is being televised on cable only or whether the major over-the-air broadcast channels—ABC, NBC, and CBS—are also carrying it. Roughly a fifth of U.S. television households do not have cable or satellite television. Moreover, viewers with cable or satellite do not ordinarily check out all the programs on the channels available to them. Whether they have 50 or 500 channels, most viewers routinely monitor only 6 to 10 of them. ABC, NBC, and CBS are usually among these channels. Thus, the likelihood that a viewer will actually come across the convention while watching television is directly related to whether a broadcast channel is carrying it. In 2000, inadvertent convention viewers were three times more likely to be captured through a broadcast channel than through a cable channel.
Thus, cutbacks in broadcast coverage will adversely affect the size of the convention audience. As recently as 1976, each network carried well over 20 hours of coverage per convention. By 1992, the average had fallen to 12 hours. It was but 5 hours in 2000 and, as indicated, will fall to 3 hours this time, which will reduce sharply the size of the inadvertent audience.
The 2000 conventions reveal just how important the level of broadcast coverage is to the size of the audience. During the hours of broadcast coverage, the television audience was never less than 15 million television households and ranged as high as 27 million. During the hours when cable coverage only was available, the audience never reached as high as 10 million households.
The broadcast networks’ decision to cut their hours of convention coverage in 2004 will effectively reduce Americans’ election interest and information. The impact will be most pronounced among younger adults. Age is closely associated with inadvertent convention exposure. In 2000, more than three-fourths of adults under 30 who watched the conventions did so inadvertently, as compared with only half of those older than 30. At this point in the 2004 campaign, younger adults are far less informed about the candidates and issues than are older adults. The reduction in broadcast coverage of the convention means that many fewer younger adults will be exposed to John Kerry’s message (and a month later, to George Bush’s message) than would be the case if the networks had retained their full public-service commitment to live convention coverage.
The broadcast networks argue that their convention participation is less critical today than in the past because gavel-to-gavel coverage is available on cable. This claim is clearly only partly accurate. The networks also say that the conventions now lack suspense and therefore are no longer newsworthy. By that standard, they could also justify slighting the State of the Union Address or, for that matter, President Ronald Reagan’s funeral, which received more than twice the live broadcast coverage than the 2004 Democratic convention will get. Polls indicate that an unusually large number of Americans believe that this year’s choice between Kerry and George W. Bush is critical for the nation’s future. Many of these Americans will watch broadcast television next week without knowing that other channels are carrying convention programming that could help them decide which direction they would like the country to go.
The results reported here are from a nationwide telephone survey of 518 adults conducted July 14-18, 2004. The survey has a sampling error of ±4percent.
Click here to see tables that could be used in news reports accompanying this material.


John F. Kennedy School of Government 79 John F. Kennedy Street
Cambridge, MA 02138
617-495-1100 Get Directions Visit Contact Page