When voters say when: the most effective get-out-the-vote techniques

A vote in South Dakota costs $70.50. At least that is what candidates, parties and interest groups spent per vote cast in 2002 to communicate their message in the state.

But during perhaps the most crucial weeks before the election voters can get sensory overload because of all of the direct mail pieces they receive and all of the television ads they view.

This year the final weeks of the campaigns will not be unique, campaigns will make their final push for voters support but research shows they should be careful not to communicate too much.

Research on campaign communication conducted by the Center for the Study of Elections and Democracy (CSED) at Brigham Young University during the 1998, 2000 and 2002 election cycles studies how voters respond to saturated information environments and how campaigns can effectively communicate in such a setting.

The volume of advertising in battle-ground presidential states and competitive congressional races is likely to be higher this year than in 2002.

The increase in advertising could be due to a competitive presidential contest with amply funded campaigns, party committees and outside groups. This means an increase in the flow of money to the system over the mid term election in 2002. Also, the provisions of the Bipartisan Campaign Reform Act (BCRA) that prohibit issue advocacy broadcast communications paid for with union or corporate treasury funds within 60 days of the general election do not apply to the ground war. This could encourage so-called 527 groups, which depend on soft money, to shift to non-broadcast communications in the closing days, while groups that have adequate hard-money reserves can continue to broadcast during the closing weeks of the campaign.

You do not need to look back too far for an example of too much information in a campaign environment. After a bruising U.S. Senate campaign in South Dakota two years ago, candidates, political parties and a variety of interest groups spent at least $24 million--about $70.50 per vote cast.

"I cannot make an informed decision about a candidate from advertising ... the TV ads and mail ads don't tell me anything," one South Dakota voter said in the survey.

Another said, "The length of time spent campaigning and the amount of money spent was absolutely ridiculous ... The TV ads were so numerous and repetitious, you eventually tune them out entirely."

There are three important facts to to remember when communicating with voters. First, in competitive federal or statewide races, voters are confronted with a large volume of campaign communications, especially during the closing weeks of the campaign. Second, many of those voters develop election fatigue and begin to tune out much of the information. Third, the selectivity of voters during the final days of the campaign favors some campaign strategies over others.

In 2002, CSED sponsored two surveys to study campaign communication. The first was a panel survey conducted in four battleground U.S. House and Senate contests along with a national baseline sample. This compared campaign dynamics in four competitive races with the nation as a whole. The battleground races were the Connecticut 5th, the Colorado 7th, the Missouri Senate and the Arkansas Senate race.

The second survey was a voter log in which we asked a random sample of registered voters in Minnesota, South Dakota, Arkansas and New Mexico to collect and send us all of their political mail as well as keep a log booklet of all of their telephone and face-to-face contacts with political campaigns during the last three weeks before Election Day. Further details on both surveys are available in "The Last Hurrah? Soft Money and Issue Advocacy in the 2002 Congressional Elections" (Brookings Institution Press, 2004).

High Volume of Communications

Voters receive frequent campaign communications during the closing days of the campaign. In the voter log survey, voters in the middle of two of the most watched and most competitive U.S. Senate races in 2002, Minnesota and South Dakota, received more than 18 pieces of political mail during the last three weeks. South Dakota voters received more than seven pieces of mail on average for the U.S. Senate race with the maximum being 32 pieces of mail on the Senate race alone in the last three weeks.

The story from the panel survey is similar. From late August to mid-September, survey respondents nationally reported seeing 2.7 television ads on an average day, while respondents in the Arkansas Senate race had more than twice as much television exposure, reporting seeing nearly 6.9 ads on an average day. By the last week of the election, the average number of TV ads viewed per day had jumped to 7.6 nationally, but was double digits in all of the competitive races except Connecticut District 5. There were 11.9 in Arkansas Senate, and 12.5 in both Colorado 7 and the Missouri Senate race. Similar trends occur in questions about the volume of mail, telephone calls and direct personal contact in these same campaigns.

How Do Voters Respond to the Onslaught of Information?

Voters express fatigue in dealing with the volume of campaign activity in intensely fought campaigns. In the panel survey, we asked respondents' views on whether or not the amount of mail and advertising was too much, about right or too little. Nationally, 35 percent said there were too many campaign ads or too much political mail in their U.S. House race. Not surprisingly, in our four sample races, between 57 and 75 percent gave this response, 57 percent in Arkansas and Connecticut 5, 61 percent in Missouri, and 75 percent in Colorado 7.

With the barrage of information they were receiving, many voters simply stopped paying attention to political communications. In the last wave of our survey, we asked respondents: "Which of the following statements best summarizes your experience during the election campaign? After all the advertising and mail, I stopped paying attention; or there was a lot of advertising and mail, but I kept paying attention throughout." In the national baseline sample, 56 percent of respondents said they stopped paying attention. However, in our four competitive races, the number of respondents saying they stopped paying attention was even larger. The numbers for Arkansas, Colorado 7, Connecticut 5 and Missouri were 62, 77, 70 and 67 percent respectively.

In a saturated campaign, contacts from trusted sources have greater impact, and voters turn to trusted, neutral sources as they seek to cut through the barrage of campaign information. We asked respondents in each wave of the panel survey the following open-end question: "In an election with conflicting messages it is sometimes hard to know where to go for information. What is the most trusted or important source of information you use in deciding your vote for US Senate/House of Representatives?" Responses were coded into more than 20 categories. One noticeable pattern is the extent to which the respondents shifted towards specific newspapers in their district and state as the campaign progressed and became more intense.

For example, in the Arkansas and Missouri Senate races in the first wave in late August/early September, only 5 percent of respondents gave a specific newspaper as the most trusted source of information.

By the second wave, the proportion of respondents in these two senate races relying on a specific newspaper had risen to about 30 percent, and this stayed steady through the end of the campaign. In Colorado 7, 9 percent saw the specific local newspaper as the most important source of information; the number rose to 39 percent in the last week of the campaign. In Connecticut 5, a specific newspaper started off in the first wave with a relatively high 26 percent, then increased to 46 percent by the end of the campaign.

While the overall volume of campaign communications and the subsequent voter fatigue are out of the control of individual consultants, voter reactions to the campaign volume in a competitive environment suggest some strategies to ensure that campaign messages penetrate the noise.

If resources permit, one option to consider is to communicate early in the cycle before voter fatigue sets in. Another option, particularly useful for issue groups, is to focus on group members who are likely to receive and respond to your message. The orange postcards used by the National Rifle Association (NRA) to communicate its endorsements to members are simple and cheap, but are likely to be received and trusted by members.

Another strategy is to produce messages that get noticed because of unusual features or size. In the California 27th Congressional District in 2000, the League of Conservation Voters (LCV) produced two attention-getting pieces of mail. One was purple and was a 5-inch-by-17-inch card when it arrived in the mailbox. It opened to 10 inches by 17 inches and then opened again to 20 inches by 17 inches. This larger format stuck out in the mail, and its color, unusual for political mail, also generated visual interest. The second LCV mailer was a standard size, but when opened had a pop-up feature like a children's pop-up book.

Finally, our research points to the importance of free media in a competitive campaign. Voters turn to newspapers and television news for information, and so an effective strategy for pursuing positive free media exposure will help ensure that campaign messages will get through to voters.

J. Quin Monson is an assistant professor of political science and assistant director of the Center for the Study of Elections and Democracy (CSED) at Brigham Young University (http://csed.byu.edu).

David Magelby is dean of the College of Family, Home and Social Sciences, distinguished professor of political science, and a senior research fellow at CSED.

Kelly Patterson is an associate professor of political science and director of CSED.

COPYRIGHT 2004 Campaigns & Elections, Inc.
COPYRIGHT 2004 Gale Group