Television News, Real-World Cues, and Changes in the Public Agenda

This paper examines the interrelationships between real-world cues, television news coverage, and public concern for the issues of energy, inflation, and unemployment. On the basis of longitudinal data, the authors show that media agenda setting is indeed unidirectional-television news influences public concern and not vice versa. Lead stories are significantly more powerful than ordinary stories in shaping the public's agenda. Prevailing conditions and events affect public opinion both directly and indirectly, by determining the degree of news coverage accorded issues. Roy Behr is a Ph.D. candidate in political science at Yale University. Shanto lyengar is Associate Professor of Political Science at the State University of New York at Stony Brook. The authors are grateful to Bill Adams for providing research facilities at the George Washington University Library, and to Steven Rosenstone who contributed significantly to the preparation of this paper. An earlier version of this paper was presented at the 37th Annual Meeting of the American Association of Public Opinion Research, Hunt Valley, Maryland, May 20-23, 1982. Public Opinion Quarterly Vol. 49:38-57 C) by the Trustees of Columbia University Published by Elsevier Science Publishing Co., Inc. 0033-362X/85/0049-38/$2.50 This content downloaded from 207.46.13.129 on Mon, 27 Jun 2016 07:19:48 UTC All use subject to http://about.jstor.org/terms TV NEWS AND AGENDA SErTING 39 mediately before the election may have increased the importance voters assigned to foreign policy as a criterion for evaluating the presidential candidates. The election was thereby transformed into a referendum on the Carter Administration's foreign policy, with disastrous consequences for Carter's candidacy. Experimental studies tend to corroborate this argument; individuals exposed to more news about a particular issue come to assign greater weight to this issue when evaluating the incumbent president (see Iyengar, et al., 1984), and when formulating their voting choices (Iyengar and Kinder, 1984; Behr, 1984). In this article we rely on longitudinal data spanning seven years to investigate the public's concern for inflation, unemployment, and energy. Our analysis extends agenda-setting research into two hitherto unexplored areas. First, we trace public concern to both television news coverage and to prevailing conditions or events. Second, we investigate the impact of television news on public concern as well as the reciprocal impact of public concern on levels of news coverage. Limitations of Past Research Given the political ramifications of agenda-setting and the preeminence of television as a credible and trustworthy source of political information (see Bower, 1983), it is not surprising that a good deal of research has been directed at the agenda-setting effects of television news programs. These studies report a kaleidescope of findings. Some researchers claim that television news coverage has no impact on the audience's perceptions of issues (Patterson and McClure, 1976); others suggest that television lags behind newspapers as an agenda-setter (Benton and Frazier, 1976) and that the effects of television on perceptions of issue importance are limited to the least educated and informed segments of the citizenry (McCombs, 1976). Confusion over the agenda-setting power of television may simply be a product of methodological limitations. The cross-sectional sample survey favored by most researchers is hardly a powerful means of testing a dynamic process such as agenda-setting. A more appropriate strategy is to search for media effects over time, as news coverage and public concern evolve. Longitudinal studies of agenda-setting, though few in number, find that changes in the level of media attention do indeed produce changes in public concern for national issues (see MacKuen and Coombs, 1982). Similarly, experimental work demonstrates that network newscasts possess an uncanny capacity to shape viewers' political concerns (Iyengar, et al., 1982). In short, it is premature to dismiss television as a medium with little agenda-setting clout. Most research on agenda-setting is not only methodologically weak, it This content downloaded from 207.46.13.129 on Mon, 27 Jun 2016 07:19:48 UTC All use subject to http://about.jstor.org/terms 40 ROY L. BEHR AND SHANTO IYENGAR also suffers from conceptual limitations. Researchers ignore the effects of direct experience on individuals' political concerns (two exceptions are MacKuen and Coombs, 1982; Erbring, et al., 1980). While there can be no denying that citizens are highly dependent upon the media for public affairs information, personal experience too is a sufficiently credible source of information. Many national issues impinge on large numbers of individuals; some issues, including crime, civil rights, and unemployment have profound personal significance. Explanations of the issueattention cycle must therefore include both mediated and direct experiences; ignoring the latter may lead to notably exaggerated estimates of the effects of media. Prevailing circumstances and events can also affect individuals' political concerns indirectly by determining what the media pay attention to. As unemployment rises, the media may devote more time to unemployment and the public's concern for the issue rises. If this is the case, then agenda-setting represents the media alerting citizens to current realities. On the other hand, despite the claims of network executives that their newscasts are mirror images of current realities, news coverage of national issues may be quite indifferent to prevailing conditions. Media agendas may be determined instead by idiosyncratic editorial, organizational, or commercial imperatives (see Epstein, 1973; Altheide, 1976; Gans, 1980), thus diverting the public from the real problems facing the nation. In either event, it is imperative that indicators of national conditions be brought to bear on the relationship between news coverage and issue salience. Real-world indicators serve two purposes: first, to assess the sensitivity of television news coverage to current conditions and events; second, to distinguish between the effects of news coverage and realworld conditions on public concern for issues. Finally, agenda-setting researchers have generally ignored the critical question of causality. It is taken for granted that news coverage is the driving force and that agenda-setting is a unidirectional or recursive process. The possibility of a feedback effect, namely, that public concern itself spawns news coverage, is ignored. The networks may choose to broadcast stories that are of current interest to their viewers. Our analysis is sensitive to this possibility. We attempt to estimate the impact of news coverage on public concern as well as the impact of public concern on