This article examines the influence of television news on viewers' attributions of responsibility for political issues. Television's systematic reliance on episodic as opposed to thematic depictions of political life elicits individualistic attributions of responsibility for national problems such as poverty and terrorism. These attributions emphasize the actions of private rather than governmental actors. By obscuring the connections between political problems and the actions or inactions of political leaders, television news trivializes political discourse and weakens the accountability of elected officials.
Get full access to this article
View all access options for this article.
References
1.
Valerie S. Folkes , “Consumer Reactions to Product Failure: An Attributional Analysis,”Journal of Consumer Research, vol. 10 (1984).
2.
Shanto Iyengar , Is Anyone Responsible? How Television Frames Political Issues (Chicago: University of Chicago Press, 1991).
3.
Donald R. Kinder and Roderick Kiewiet, “Economic Discontent and Political Behavior,”American Journal of Political Science, vol. 23 (1979).
4.
4. Stephen Ansolabehere and Shanto Iyengar, Going Negative: How Political Advertisements Shrink and Polarize the Electorate (New York: Free Press, 1995).
5.
Richard Thaler , “The Psychology and Economics Conference Handbook,” in Rational Choice: The Contrast Between Economics and Psychology, ed. Einhorn and Hogarth.
6.
Howard Schuman and Stanley Presser, Questions and Answers in Attitude Surveys: Experiments on Question Form, Wording, and Context (New York: Academic Press, 1982).
7.
7. Iyengar, Is Anyone Responsible? chap. 3.
8.
William A. Gamson , “News as Framing,”American Behavioral Scientist, vol. 33 (1989).
9.
9. Of course, this is a probabilistic argument. It is possible that despite the use of randomization, experimental conditions will still differ in composition. It is standard practice in experimental research to verify that random assignment of participants achieves the desired effect. In none of the framing experiments described here did the conditions differ significantly on any relevant background characteristic.
10.
10. The nine separate experiments were administered between June 1985 and September 1987. The number of individuals who participated in each experiment ranged from 40 to 244. While the participants constituted a reasonable approximation of Suffolk County, New York, they were, in comparison with the American electorate, more affluent, more Catholic and more Jewish, more educated, and more likely to have voted in the last election. For further details on the sample participants, see Iyengar, Is Anyone Responsible? chap. 3.
11.
11. The instructions stated, “Today television news is the major source of information for Americans. This study is about how people evaluate, understand, and interpret television news stories. We are particularly interested in `selective perception.' Do people's opinions about politics and government influence how they react to news? Do Republicans and Democrats really see the same news?”
12.
12. Although these responses were unwieldy and coding intensive, they have the advantage of nonreactivity; unlike closed-ended questions, open-ended questions do not cue respondents to think of particular causes or treatments. Two coders read each questionnaire and classified each response. Despite the large number of raw categories, intercoder agreement indicated acceptable levels of reliability.
13.
13. The experimental procedures adhered fully to the American Psychological Association's guidelines on the conduct of experimental research.
14.
14. To guard against the possibility of confounded variables, the experimental tests of framing were always replicated with an entirely different set of news stories. To the degree that the observed framing effects appear across different sets of manipulations, the possibility that idiosyncratic differences between episodic and thematic reports are at work is minimized.
15.
15. Crime elicited the largest number of causal and treatment attributions—2.7 and 2.1, respectively—presumably because of its relatively obtrusive nature.
16.
16. The “black crime-episodic” condition described a violent gang-related confrontation in Los Angeles. The “black crime-thematic” condition described the extent of crime in inner-city neighborhoods in several major cities. The “white crime-episodic” condition focused on the well-known shooting incident in the New York City subway in which Bernhard Goetz, a white passenger, shot at four black teenagers who allegedly approached him in a menacing manner. The “white crime-thematic” condition detailed the growing importance of organized crime.
17.
17. The distinctiveness of unemployment may stem from the unusually high level of prominence accorded the issue during the period of this research. Thus people may have regarded unemployment as a particular instance of national economic problems.
18.
18. Each respondent's percentage of individualistic attributions was subtracted from the percentage of societal attributions to arrive at a net responsibility score.
19.
Philip E. Converse , “The Nature of Belief Systems in Mass Publics,” in Ideology and Discontent, ed. David Apter (New York: Free Press, 1964).
20.
20. For experimental evidence that demonstrates this connection, see Franklin Gilliam, Shanto Iyengar, and Adam Simon, “The Intersection of Race, Crime and Television News” (Paper delivered at the annual meeting of the Midwestern Political Science Association, Chicago, 1995).
