[131] ANOTHER LOOK AT THE AGENDA-SETTING FUNCTION OF THE PRESS

AUTHORS ’ NOTE: An earlier draft of this article was presented at the 1973 meeting of the Theory and Methodology Division- of the Association for Education in Journalism at Ft. Collins, Colorado. The authors thank Jay Blumler, John Bowes, and Steven Chaffee for their reading of the earlier The agenda-setting hypothesis asserts that the media have an effect indirectly by choosing certain issues for emphasis, thus making those issues more salient to the audiences. The hypothesis, stated in such general terms, presents formidable conceptual and methodological difficulties that are dealt with in this article. A controlled study of the audiences of two newspapers with differing content emphases was conducted during the 1972 presidential campaign. The results show only moderate support for the agenda-setting hypothesis; the honesty in government issues, given heavy play in one of the two newspapers, failed to generate much enthusiasm among readers of either paper. In addition the results suggest agenda setting is not a broad and unqualified media effect. Predicted differences mainly were restricted to the less involved and less motivated partisans who were heavily dependent on the newspapers for their political news. Finally, the importance of studying issue saliences apart from political attitudes was illustrated by the relatively strong relationship between such saliences and voter turnout and direction. a quarter century has passed since Lazarsfeld, Berelson,~~N and Gaudet (1948) published their classic study of communication and voting behavior. Despite the fact that the major inferences about communication were based on only 483 respondents in only one county in Ohio, that the chosen Erie County voted strongly Republican in that Democratic election year of 1940, and that a pretelevision campaign was necessarily involved, the results of that study have remained largel