Agenda Diversity and Agenda Setting From 1956 to 2004

Abstract

Using aggregate time-series data, this study tests four research hypotheses. First, we examine the long-term trend in the issue agenda diversity of the public as measured by the Gallup Most Important Problem question from 1956 to 2004. Second, we test whether the agenda-setting effect between the media agenda and the public agenda has become weaker over that time. Finally, with multivariate autoregressive integrated moving average (ARIMA) modeling, this study investigates the causal relationships among the longitudinal changes in public agenda diversity, the New York Time's content diversity, and the New York Times' agenda-setting effect on public opinion. While no significant trend in the agenda-setting effect was found, we found a significant quadratic trend in public agenda diversity and significant causal relationships among these three time-series measures. In short, increased public agenda diversity decreased the New York Times' public agenda-setting effect, and the issue agenda diversity of the New York Times has decreased over time, but the overall agenda-setting effect between the New York Times and the public has not become weaker over time.

KEYWORDS:

• agenda diversity
• agenda-setting effect
• public opinion
• New York Times
• ownership concentration

Notes

1. A flat or even distribution is considered most diverse because it indicates that all the categories are equally well represented.

2. Zhu (1992) makes explicit the theoretical reasoning and mathematical derivation for the relationship among the agenda-setting effect, issue competition, and one-way distraction.

3. According to Jones and Baumgartner's (2005) description, the sample is derived by coding the first entry on every odd-numbered page of the Index.

4. Originally, the New York Times' content coding system had 27 content categories (listed earlier). Seven issue categories from the New York Times' content coding system are not included in the MIP coding system. They are: (1) state and local government administration, (2) weather and natural disasters, (3) fires, (4) arts and entertainment, (5) sports and recreation, (6) death notices, and (7) churches and religion. They are recoded into the category named “other topics” because they are irrelevant to the MIP question. After the recoding, the two systems become identical. However, when measuring the agenda diversity of the New York Times, we used the original 27-category system to make the measure as accurate as possible.

5. The 20 issue categories are: (1) macroeconomics, (2) civil rights, (3) health, (4) agriculture, (5) labor, immigration, and employment, (6) education, (7) environment, (8) energy, (9) transportation, (10) law, crime, and family issues, (11) social welfare, (12) community development and housing, (13) banking, finance, and domestic commerce, (14) defense, (15) space science, technology, and communications, (16) foreign trade, (17) international affairs and foreign aid, (18) federal government operations, (19) public lands and water management, and (20) others.

6. Theoretically, the value of Simpson's D in this study can vary from 0 to 0.95 with 0 indicating that one issue completely dominates the media content and 0.95 indicating that all 20 issue categories receive an exactly equal share of the content (a flat distribution).

7. This is a very commonly used measure of association between two sets of rankings. It varies from −1.0 to +1.0.

8. The annualized or quarterized proportions were constructed by normalizing the percentage of responses in every major topic (e.g. Defense) by the total percentage of responses to a single poll, and then averaging these proportions across the multiple polls in any given year or quarter.

9. Interestingly, the mean agenda-setting effect reported here (0.51) is virtually identical to the 0.53 reported in Wanta and Ghanem's meta-analysis (2006, 45).

10. Further differencing may cause an inconvertible moving average without autoregression, so it is not recommended (Dickey and Fuller 1979).

11. Cable News Network (CNN) is a 24-hour TV news channel launched by Ted Turner in 1980.

12. During the middle of the Reagan presidency, which had been favorable to deregulating business, the Cable Communications Acts of 1984 is the first major federal act deregulating cable. Since then, cable has captured more and more prime-time audience from the broadcast networks.