Exposure to the Newer Media in a Presidential Primary Campaign

Abstract

Cable television news channels and online news sites appear to offer interested voters the ability to follow presidential election campaigns more closely than ever before. However, survey research looking at the extent to which Americans are taking advantage of these newer media is incomplete. Rarely is new media use adequately assessed in surveys, and no extant study has simultaneously examined exposure to contemporary news channels over the course of several weeks. The present study uses an aggregate-level analysis of naturally occurring news consumption behavior to determine whether public selection of broadcast news programs, cable news channels, and online news outlets follows the primary election schedule and fluctuations in voter interest in the election. The results suggest that people turn to cable news and online political content during key political events (i.e., the Super Tuesday primary period) but less so when the political stakes are much lower. In addition, the data reveal that news reading at local news sites during key events takes on a more local character than does reading at other times. In sum, the study demonstrates that aggregate-level use of the newer media is responsive to changes in the political environment. Audiences seem willing to take advantage of a growing number of options for finding information about politics.

Keywords:

• campaign news
• Internet
• cable television
• presidential primary
• new media
• news audiences

Funding for this study was provided by the UIUC Campus Research Board at the University of Illinois at Urbana-Champaign. I thank Larry Oathout for his assistance in data coding, the CNN Research Department for some of the data used in this project, and Scott Althaus for comments on an earlier draft.

Notes

∗p < .05; ∗∗p < .01.

^† p < .10; ∗p < .05; ∗∗p < .01.

^aThe New York and California primaries occurred on Super Tuesday.

^bMichigan held its primary in February.

∗p < .05; ∗∗∗p < .001.

^aThe Nando Times was an online unit of the McClatchy Company, a newspaper publisher. The site was closed in May 2003.

^bFor site viewers, this represents unduplicated viewers.

1. It seems sensible to also hypothesize that the primary schedule will be more strongly correlated with both cable television exposure and online political news use than with network television news exposure. However, as will shortly become apparent, the data for testing this suggestion were not available for this project.

2. For all analyses reported here, the two components were combined.

3. The Pew Center survey data suggest that CNN.com was the most popular online news site for campaign information in 2000. Twenty-four percent of those who reported going online for campaign news in 2000 named CNN.com as the destination they visited most frequently; 59% reported visiting that site at least once during the campaign (Pew Center, 2000a). A good number of campaign news readers online also reported visiting Web sites of national television networks (45%) and national newspapers (33%) at least once during the campaign. Other popular destinations were news features of major portals such as AOL and Yahoo (57% reported going there at least once) and MSNBC.com (52%; Pew Center, 2000a).

4. A second analyst coded at least 500 page views from each site. The unweighted average kappa for the content coding was .88.

5. For example, the last 3 days of May are covered by June 4 survey data that also include the first few days of June.

6. The index scores ranged from 18 to 22 in April (not shown). The index remained in that region until the party convention period, when it again briefly rose above 40.

7. Looking at the components of the voter involvement index individually yields almost identical results.

8. The analysis in Table 2 used seasonally adjusted figures to compensate for the weekly news reading cycle. Because only Mondays, Tuesdays, and Wednesdays are examined in this analysis, unadjusted data are used.

9. The Louisville Courier-Journal was omitted from this analysis because it did not experience any political news page views during this period.

10. The overall pattern was similar across the sites; they were combined to increase statistical power.

11. A search of the Vanderbilt television news archive for the word “primary” in March returned 19 news reports on the three broadcast networks' daily news programs. A parallel search in May failed to detect a report.

12. I am indebted to an anonymous reviewer for suggesting this possibility.

13. Similarly, it is important to note that comparisons between television viewing and Internet use are hampered by the fact that these are very different media. While simple audience counts are used for all of the media, the duration analysis made use of minutes of exposure to cable and views of Web pages. It is possible that the differences between these two metrics may account for some of the superiority of the former as a correlate of the May primary schedule. That comparability issue is a problem inherent to medium comparisons and must be kept in mind when examining the results here.