Campaign Advertising and American Democracy

3. Measuring Exposure to Campaign Ads

C H A P T E R T H R E E Measuring Exposure to Campaign Ads T HE DISPARATE CLAIMS about campaign advertising and its effects on American citizens, explored in the last chapter, reflect in part the myriad methodological approaches that have been brought to bear on the subject. The methodological pluralism represented in the literature is, by some measures, a good thing; however, at times it seems as though the field has been left with more questions than answers. Ultimately, the study of campaign advertising—and its impact on citizens—has been hampered by a dearth of good empirical data and an insufficient understanding of how advertising is crafted and deployed in political campaigns. In a perfect world, scholars seeking to describe the nature of a political ad war and its effects would have precise information on exactly which ads were aired (and how many times) and exactly who saw them (and how many times). In this perfect world, scholars also would have detailed information on the demographics, attitudes, and political predispositions of those exposed to particular ads. Of course, in a perfect world, political strategists also would randomly deploy their advertising efforts, thereby allowing scholars to attribute any differences in advertising exposure purely to the advertising. In the laboratory, perfect information and randomization are easy to come by; researchers know precisely what ads subjects have been exposed to. Sadly, in the real world, matters are a good bit more difficult. The absence of comprehensive data on the content, timing, volume, and targeting of political advertising has limited what scholars can report about the strategies employed by campaigns and the balance of advertising in particular contests. More important , the lack of comprehensive data has made it difficult for scholars to measure citizens’ exposure to advertising and to study the effectiveness of these communications. 26 C H A P T E R T H R E E We begin this chapter with a detailed critique of several approaches to the study of political advertising. Some of these approaches focus on measuring the advertising environment of a particular campaign through the use of aggregate campaign spending data, archival collections of television commercials, and logs from the public files of television stations. Other approaches—experimentation, self-reports by survey respondents, and proxy measures of exposure—attempt to measure the effects of advertising. Each of these methods has advantages, but each also has weaknesses that make it difficult for scholars both to characterize the information environment and to infer how campaign messages influence the attitudes and behavior of citizens. Campaign Spending One common proxy for the campaign information environment is candidate spending. By measuring campaign expenditures, scholars have sought to make descriptive inferences about the relative volume and impact of candidate messages . For instance, some scholars have examined the relationship between expenditures and election outcomes (Green and Krasno 1990; Jacobson 1992; Gerber 1998); others have explored money’s impact on voter knowledge and affect toward candidates (Coleman and Manna 2000; Coleman 2001). While candidate spending may be a reasonable “quick and dirty” proxy for the intensity of campaign communications, the measure can be far removed from the actual messages that voters receive and to which they respond. Indeed, researchers have recognized this mismatch, referring to a “black box” through which campaign money is translated into electoral outcomes. As Coleman and Manna acknowledged, “campaign money must work through campaign strategy, advertising content, advertising frequency, and other intermediaries” (2000, 759). There are few places left in American where money can directly buy votes. Rather, money affords candidates the means by which to spread their messages or bring their supporters to the polls, two activities designed to increase the candidates ’ vote shares. The important point, as Ansolabehere and Gerber noted, is that “total campaign spending may not be a good measure of expenditures devoted to actual campaigning” (1994, 1107). Making use of Fritz and Morris’s (1992) comprehensive analysis of Federal Election Commission spending reports , Ansolabehere and Gerber separated campaign expenditures into three types: “direct communications with voters,” such as radio or television commercials ; other campaign activities, such as polling, office expenses, or the hiring of a consultant; and spending unconnected to a candidate’s own campaign, such as a donation of money to another candidate. Ansolabehere and Gerber find that House challengers devote, on average, only 58 percent of total expenditures to campaign communications. For House incumbents, the comparable figure is just 42 percent (1994, 1110). M E A S U R I...