Election Campaigns and Media Effects
ELECTION CAMPAIGNS AND MEDIA EFFECTS
For most people living in established democracies and societies that are in transition to democracy, election campaigns are primarily experienced through the media. Politicians know that far more people turn to the media for information than turn out for political rallies in local town squares. The daily campaign activities are thus primarily designed to meet the constraints and deadlines of the major news outlets. Therefore, there are two important contexts to consider when thinking about the effects of the media in election campaigns. One is the context of the campaign or the potential media effect on the campaigns of candidates, which can be described as the institutional level of media effects. The other is the context of the potential media effect on individual voters or citizens, which can be described as media effects at the individual level.
Institutional Contexts and Effects
The institutional effects of the media on the campaigns of candidates may vary depending on the type of electoral system, the rules and regulations governing campaign coverage, and other institutional characteristics of the political and media systems. In the United States, for example, where the race for the presidency begins a year prior to the election with candidates declaring their candidacy and then moves into the primary season when Republicans and Democrats vote in each state to select the ultimate nominees for the parties, the media play a very important role in shaping expectations and judging outcomes. In the year leading up to the actual election, the media pass judgement on the viability of the candidates based on the indicators that the media decide are important.
One of the most important indicators has been the amount of money a campaign has raised, and another is the professional background of the candidate's campaign managers. These two factors have led to some candidates withdrawing from the race even before the primary season begins, so voters are never even given a chance to pass judgement on those particular candidates. In the 2000 race for the presidency, money raised was used as the major indicator of the viability of George W. Bush, and any challengers on the Republican side were considered to be marginal until the primary season began. The surprise victory of Senator John McCain in the 2000 New Hampshire Republican primary was all the more powerful because it exceeded expectations, and momentum provided by that win generated more than six million dollars in campaign donations via the Internet. The Internet has made campaign donations much faster, and its use in campaigns may further fuel the momentum provided by unexpected outcomes. In other countries where money is not an important indicator of candidate viability because of different campaign finance rules, or where the professionalism of electioneering is a more recent phenomenon, there may be less opportunity for media to have an effect on the campaigns of parties and candidates.
The shaping of expectations is very important—sometimes more important than actual outcomes. In the New Hampshire primary in 1992, for example, Bill Clinton finished second after Paul Tsongas. That fact is difficult for most people to remember because the media coverage of that primary named Clinton the real winner because he did much better than expected. As this shows, one does not have to win a primary in order to be labeled the winner.
Media coverage, of course, is not determined by journalists alone. It is a product of the efforts of politicians and their advisors, the so-called spin doctors who talk with journalists. To what extent do politicians have control over the news agenda? A comparative study of news coverage of elections in Great Britain and the United States in the 1980s and how it was produced showed that British politicians had considerably more opportunity to influence television news coverage than U.S. politicians did and that U.S. television journalists exerted considerably more discretion in shaping the news agenda than their British counterparts did. Holli Semetko, Jay Blumler, Michael Gurevitch, and David Weaver (1991) provided evidence for this conclusion with a variety of content analysis indicators. These indicators included the following:
- the amount of space used in the main evening news program for coverage of election news (more in Great Britain than the United States),
- the amount of news devoted to politicians' "soundbites" (considerably more in Great Britain),
- extent to which the main topics of news stories were initiated by politicians or journalists (more party-initiated news in Great Britain, more media-initiated news in the United States),
- the proportion of political stories in which politicians or parties were the main focus (greater in Great Britain than the United States), and
- extent to which reporters offered evaluations of political participants (more in the United States than in Britain).
Whereas British reporters were more likely to offer only descriptive comments on politicians' activities on the campaign trail, U.S. reporters were more likely to evaluate candidate performance. The only instance in which politicians in both countries were on equal footing in terms of their ability to influence the news agenda was in the domain of visuals. Politicians in both countries initiated the majority of key visuals in election news stories, and the vast majority of these visuals were favorable. In the United States, however, positive visuals were far more likely to have been accompanied by critical voiceover commentary by reporters; in Great Britain reporters were more likely to describe the scene in a neutral way. A look at television coverage of elections in the two countries in the 1990s suggests that while British reporting may be moving in the direction of the U.S. coverage, there is still some gap between the two.
Other institutional contexts of importance include the balance between public and commercial (private) broadcasting, the political autonomy of broadcasting from government and political parties, the rules and traditions that surround party access to broadcasting, and the extent of partisanship in the printed press. In theory, television (whether the channel is public or private) is expected to provide impartial coverage of politics, and this is deemed to be of particular importance at election time. Research by Semetko (1996) has shown that in practice, the meaning of "balance" in election news varies not only across countries but also across news organizations within a particular country.
In the United States and other countries, for example, parties of government and the president or prime minister continue to conduct the business of government during the official election campaign as well as in the weeks preceding it. It is up to reporters and journalists, as well as political partisans, to label any such event or activity as "campaigning." U.S. television reporters have been more ready than their colleagues elsewhere to label as campaigning any incumbent activity at any time during an election year, regardless of the gravity of the event or situation. For example, Jimmy Carter's "Rose Garden strategy" in the final weeks of the 1980 election during the U.S. hostage crisis and George H. W. Bush's 1988 visit to Florida (a "key" state in that presidential campaign) to provide government relief to the victims of Hurricane Andrew were both connected by journalists to vote-getting strategies. As a contrast, in the final days before the 1990 election in Germany, the first national election after the fall of theBerlin Wall and the unification of the eastern and western parts of the country, television news coverage failed to mention that Chancellor Helmut Kohl's meetings with heads of state were "photo opportunities" or in any way connected to enhancing his image as a leader.
These cross-national contrasts in reporting styles are changing, however. David Swanson and Paolo Mancini (1996) have argued that most countries had by the 1990s moved in the direction of the United States with respect to campaigning techniques and strategies, as well as with respect to an increased number of television channels and thus more competitive media markets. This brings with it a tendency for reporting to become more ratings-dependent and star-oriented coverage. One indication of this trend was the decision in 1999 to have Klaus-Peter Siegloch anchor the main evening news on Germany's Second German Television channel (ZDF), one of the country's two public service broadcasters. Siegloch was a well-known figure because of the reports that he had been filing from Washington, D.C., during the previous five years. As a result, he brought his personal credibility to the program, along with a more American style of anchoring, and he incorporated many of the format features found in U.S. evening television news. The result was increased ratings for the program.
One very popular form of broadcast access to election information is the debate between party leaders. A debate is arguably the single most important unifying event of a campaign, if only because millions of electors share the experience. While debates are a tradition in U.S. presidential elections, they are less common in other countries. Debates can be the centerpiece of the campaign, and they can have important effects on the candidate image, perceptions of who is winning, and ultimately voter participation. Although debates can produce more voter involvement in elections and are therefore quite welcome, there is no guarantee that debates will take place. For example, debates between the party leaders were common in Germany in the 1970s and 1980s, and Peter Schrott (1990) has shown that the perceived winners of the debates won more election votes than the perceived losers of the debates. However, there were no debates in the Bundestag elections of 1990, 1994, and 1998 because the incumbent chancellor, Kohl, was running for re-election as the leader of the Christian Democratic Party (CDU) and was not interested in participating in debates.
Individual Contexts and Effects
Media effects on individuals may be short-term or long-term effects. They may be cognitive (i.e., effects on political knowledge), attitudinal (i.e., effects on political opinions) or behavioral (i.e., effects on actual voting). Despite the fact that television has transformed the electoral process, it has proven difficult to isolate exposure or attention to television as a significant variable in determining vote choice. In the 1990s, a number of studies explored media effects in one or another of these domains in the national elections of a variety of countries. Richard Johnston and his colleagues (1992) studied Canada. For Great Britain, William Miller (1991) studied the 1987 general election, John Curtice and Semetko (1994) studied the 1992 general election, and Pippa Norris and her colleagues (1999) studied the 1997 general election. Semetko and Julio Borquez (1991) studied the effects of the media in the 1988 presidential election in France, while Semetko and Klaus Schoenbach (1994) studied the first national election in German after the 1990 unification.
The United States has the longest history of election research, and it dates back to the 1940s. The media became a central focus of some election research in the 1960s and 1970s, and they have been the central focus of many more studies since then. Some important examples include Thomas Patterson (1980, 1994), who has written seminal books on media effects in the 1976 presidential election and has conducted a long-term study of changes in news coverage of presidential campaigns, Marion Just and her colleagues (1996), who have studied how people interpret election information, and Russell Dalton and his colleagues (1998), who have studied the way in which metropolitan newspapers mediated the 1992 campaign. Two of the most well-known concepts in media effects research on political attitudes, agenda-setting and priming, provide a way of looking at questions of effects of news and information on public opinion and the influence of politicians on news content.
The notion of an all powerful media—with direct effects injected as if by hypodermic needle—was an important part of mass society theorists' explanation of the experience of Nazi Germany. The propaganda model of the 1920s later led social scientists in the United States to study the power of the media in democracy and the electoral process. However, the early empirical evidence suggested only a limited ability of the media to influence the public's political attitudes and voting behavior. Empirical research into media effects on partisan preferences in U.S. elections in the 1940s brought the so-called "reinforcement" model of media influence into fashion. According to this model, exposure to news during the campaign did not change vote choice for most people; it simply reinforced preexisting partisan preferences. By the 1970s and 1980s, the news media became major players in the presidential selection process. As a consequence, a broader view of media influence, known as the "limited effects" model, emerged.
The concept of "agenda-setting" refers primarily to the process by which issues in the news become important in public opinion. It brings scholarship back to the notion of a powerful media, but one which does not have electoral outcomes or the vote as its primary focal point. It has become one of the most important concepts in public opinion and media effects research. Since the term itself was first coined by Maxwell McCombs and Donald Shaw (1972) in their community study of media agenda-setting in the 1968 U.S. presidential campaign, hundreds of empirical studies have been published on the subject (see Protess and McCombs, 1991). The majority of these studies focus on the effect of news agendas or media agendas on public opinion. As in the year-long study by David Weaver and his colleagues (1981), many of these studies involve data collected during election campaigns. Taken together, these studies provide a substantial body of evidence that the news media can and do influence public perceptions of the importance of issues.
A number of studies have also failed to support the agenda-setting hypothesis, however. The study conducted by Miller (1990) of the 1987 British general election and the study conducted by Norris and her colleagues (1999) of the 1997 general election found little or no significant agenda-setting effects over the four-week campaign period. Similarly, in Germany's 1990 election, Semetko and Schoenbach (1994) found that the most visible issues in the news were not those that were most important with regard to public opinion. These studies provide important evidence to suggest that agenda-setting, as it was originally and narrowly defined in terms of media effects on issue salience, was not operating. However, the studies also found evidence of other significant media effects, particularly in the domain of public evaluations of political parties and top candidates. Agenda-setting therefore should not be taken as the sole or the primary indicator of a powerful news media. The absence of evidence to support the agenda-setting hypothesis in an election does not mean that other important effects on opinions are entirely absent.
A related concept is priming, which refers to the ability of what is emphasized in the news to alter the standards by which citizens evaluate political leaders. By emphasizing some issues and by ignoring others, the news media may "prime" the public to think about those issues when judging the performance of politicians (see Iyengar and Kinder, 1987; and Iyengar, 1991). Jon Kros-nick and Donald Kinder (1990) show that, indeed, exposure to news about Attorney General Ed Meese's announcement concerning U.S. involvement in the Iran-Contra Affair directly and immediately led to changes in the issues that were used by the public to evaluate President Ronald Reagan's performance.
One of the earliest findings of agenda-setting research established variation in effects; not all of the people are influenced all of the time. One of the most important developments in agenda-setting and priming research has been the identification of the contingent conditions under which influence can occur. Two of the most commonly discussed conditions are interest and knowledge. A number of studies have established that effects can be modified by the public's interest in information or knowledge about a subject. In these studies, the more knowledgeable people are distinguished from those who have little or no knowledge. Generally speaking, those who are least susceptible to agenda-setting and priming effects are those who have some independent store of knowledge. This knowledge enables people to argue against what they see in the news. These studies therefore offer a rather disturbing conclusion from the perspective of democracy. Shifts in public opinion about the president most commonly occur in those who are the least informed or knowledgeable, as those who know the least and have weak or no attachments to political parties are most likely to be influenced by the news.
Political Advertising and Media Effects
Although scholars and practitioners alike agree that political advertising is important for election campaigns, there is no clear agreement on the effects of political advertising on electoral outcomes. Political advertisements on television and radio count for much more in U.S. elections than in many other countries such as Great Britain, for example, where the purchase of broadcast advertising is prohibited and the forms of television advertisements are regulated. There are far more advertisements in U.S. elections than in elections abroad, and as Lynda Lee Kaid and Ann Johnston (1991) have shown, the percentages of negative advertisements in U.S. election campaigns has increased over the years. Negative advertisements take various forms; at the core they involve criticism of a candidate, a policy position, or past performance. An overview of research by Kaid (1999) has shown that exposure to advertisements does influence public perceptions of the candidates. However, a meta-analysis (i.e., an empirical study of all the studies published to date specifically on the effects of negative advertising) led Richard Lau and his colleagues (1999, p. 851) to question "why negative political advertisements have become so popular in practice when there is so little evidence that they work especially well."
Despite these doubts on the effectiveness of such advertising, debate continues over the question of whether negative advertising mobilizes or demobilizes the electorate. For example, it has been argued that "going negative" actually discourages people from going to the polls to vote and diminishes confidence in the political system (Ansolabehere and Iyengar, 1995; Ansolabehere, Iyengar, and Simon, 1999), but analysis of similar data resulted in the view that such conclusions cannot be sustained (Wattenberg and Brians, 1999). Research by Steven Finkel and John Geer (1998) on the effects of attack advertisements have also cast doubt on the idea that they demo-bilize the electorate.
Research methods are often at the core of the debate, although different campaign settings, for example, whether it is a presidential, congressional, or local election, can also influence conclusions about the power of negative advertisements and negative information. Kim Kahn and Patrick Kenney (1999) showed that in the 1990 U.S. Senate elections, for example, voters were able to distinguish between "mudslinging" and "legitimate criticism," and when the latter (but not the former) increased, citizens were more likely to vote. The effect was especially strong for those who had low interest in politics, little knowledge about politics, and lacked attachments to the main parties or described themselves as independents.
Much of what is now known about the media in election campaigns comes from research conducted in the United States. There is a considerable amount of scholarship in Germany on media content and its uses and effects in elections, as well as a growing body of literature in Canada, Australia, Great Britain, Italy, Spain, Scandinavia, and The Netherlands. However, data remain extremely limited for many other advanced industrial societies. It has only been relatively recent that the topic has become the focus of scholarship in Latin America, largely because of the rise of television as a major source of political information, candidates' strategic use of the news media, and the growth of public opinion polling in that region. In Russia, Eastern Europe, the new republics, and other societies in transition to democracy, research on elections and the media is still in its infancy.
The institutional contexts of elections in these other countries can be quite different from the United States. The main challenge for research on individual-level effects is to identify the contingent conditions under which effects occur. In other words, researchers need to identify the specific characteristics of media contents and media audiences that lead to specific types of effects, and they need to determine how the institutional contexts enhance or diminish these effects.
Ansolabehere, Stephen, and Iyengar, Shanto. (1995). Going Negative: How Political Advertisements Shrink and Polarize the Electorate. New York: Free Press.
Ansolabehere, Stephen; Iyengar, Shanto; and Simon, Adam. (1999). "Replicating Experiments Using Aggregate and Survey Data: The Case of Negative Advertising and Turnout." American Political Science Review 93:901-909.
Curtice, John, and Semetko, Holli. (1994). "Does ItMatter What the Papers Say?" In Labour's Last Chance?, eds. Anthony Heath, Roger Jowell, and John Curtice. Aldershot, Eng.: Dartmouth.
Dalton, Russell J.; Beck, Paul A.; and Huckfeldt, Robert.(1998). "Partisan Cues and the Media: Information Flows in the 1992 Presidential Election." American Political Science Review 92:111-126.
Finkel, Steven E., and Geer, John G. (1998). "A Spot Check: Casting Doubt on the Demobilizing Effect of Attack Advertising." American Journal of Political Science 42:573-595.
Iyengar, Shanto. (1991). Is Anyone Responsible? Chicago: University of Chicago Press.
Iyengar, Shanto, and Kinder, Donald R. (1987). News that Matters: Agenda-Setting and Priming in a Television Age. Chicago: University of Chicago Press.
Johnston, Richard; Blais, Andre; Brady, Henry E.; and Crete, Jean. (1992). Letting the People Decide: Dynamics of a Canadian Election. Stanford, CA: Stanford University Press.
Just, Marion; Crigler, Ann; Alger, Dean; Kern, Montague;West, Darrell; and Cook, Timothy. (1996). Crosstalk: Citizens, Candidates, and the Media in a Presidential Campaign. Chicago: University of Chicago Press.
Kaid, Lynda Lee. (1999). "Political Advertising: A Summary of Research Findings." In The Handbook of Political Marketing, ed. Bruce I. Newman. Thousand Oaks, CA: Sage Publications.
Kaid, Lynda Lee, and Johnston, Ann. (1991). "Negative Versus Positive Television Advertising in US Presidential Campaigns, 1960-1988." Journal of Communication 41:53-64.
Kahn, Kim F., and Kenney, Patrick J. (1999). "Do Negative Campaigns Mobilize or Suppress Turnout? Clarifying the Relationships Between Negativity and Participation." American Political Science Review 93:877-889.
Krosnick, Jon A., and Kinder, Donald R. (1990). "Altering the Foundations for Support for the President Through Priming." American Political Science Review 84:497-512.
Lau, Richard R.; Sigelman, Lee; Heldman, Caroline; and Babbitt, Paul. (1999). "The Effects of Negative Political Advertisements: A Meta-Analysis Assessment." American Political Science Review 93:851-877.
McCombs, Maxwell, and Shaw, Donald. (1972). "The Agenda-Setting Function of the Mass Media." Public Opinion Quarterly 36:176-187.
Miller, William L. (1991). Media and Voters: The Audience, Content, and Influence of the Press and Television in the 1987 General Election. Oxford: Clarendon Press.
Norris, Pippa; Curtice, John; Sanders, David; Scammell, Margaret; and Semetko, Holli A. (1999). On Message: Communicating the Campaign. London: Sage Publications.
Patterson, Thomas. (1980). The Mass Media Election.New York: Praeger.
Patterson, Thomas. (1994). Out of Order. New York:Vintage.
Protess, David, and McCombs, Maxwell, eds. (1991). Agenda-Setting: Readings on Media, Public Opinion and Policymaking. Hillsdale, NJ: Lawrence Erlbaum.
Schrott, Peter R. (1990). "Electoral Consequences of'Winning' Televised Campaign Debates." Public Opinion Quarterly 54:567-585.
Semetko, Holli A. (1996). "Political Balance on Television: Campaigns in the US, Britain, and Germany." Harvard International Journal of Press/Politics 1:51-71.
Semetko, Holli A.; Blumler, Jay G.; Gurevitch, Michael; and Weaver, David. (1991). The Formation of Campaign Agendas. Hillsdale, NJ: Lawrence Erlbaum.
Semetko, Holli A., and Borquez, Julio. (1991). "Audiences for Election Communication." In Mediated Politics in Two Cultures: Presidential Campaigning in the United States and France, eds. Lynda Lee Kiad, Jacques Gerstle, and Keith Sanders. New York: Praeger.
Semetko, Holli A., and Schoenbach, Klaus. (1994). Germany's "Unity Election": Voters and the Media. Cresskill, NJ: Hampton Press.
Swanson, David, and Mancini, Paolo, eds. (1996). Politics, Media, and Modern Democracy. New York: Praeger.
Wattenburg, Martin P., and Brians, Craig L. (1999). "Negative Campaign Advertising: Demobilizer or Mobilizer?" American Political Science Review 93:891-899.
Weaver, David H.; Graber, Doris A.; McCombs, Maxwell E.; and Eyal, C. E. (1981). Media Agenda-Setting in a Presidential Election: Issues, Images, and Interest. New York: Praeger.
Holli A. Semetko