ROUNDTABLE
New Media Measures for the ANES?"

Althaus and Tewksbury's Reaction

MEETINGS
Calls for papers
Upcoming meetings

RESOURCES
Featured data resource
Books
Grants and resources

LINKS
Archived Issues
Related Links
Policies & Procedures

"The Challenge of Measuring News Consumption"

By Erika Franklin Fowler, Ken Goldstein, and Dhavan Shah

Scott Althaus and David Tewksbury should be applauded for their effort.  We share the concerns they express regarding any proposal to rely on knowledge measures as a proxy for assessment of the process of information acquisition.  In addition to the objections they rightly raise, a knowledge-based approach fails to attend to two major issues: (1) the underlying process that produces learning from news consumption only for some as indicated by work on the cognitive mediation model (see link) and (2) the possibility that consuming news leads to misinformation instead of learning as observed in a recent PIPA analysis (see link).  Because information seeking and political expression in online settings has grown, we also agree that there needs to be considerably more measurement of patterns of Internet usage (see link).

Nonetheless, the approach they propose is not free of other potential drawbacks and limitations.  For example, when it comes to gauging exposure to information on television, it is crucial to keep some differentiation of programs. Television remains the dominant medium through which individuals receive political information; yet there is a wide span of programs from which to receive this information.  More important, both the content and the audience for different televised news programs differ widely. The implication, therefore, is that survey questions that treat television news as a monolithic whole, lumping together viewership across the day, or across cable networks, without distinguishing who watches what and how often, will introduce vast measurement error into studies attempting to ascertain the effect of America’s most widely used news medium.

The reason for this is simple.  The amount of political information varies widely between programs.  Fifty-four percent of the public reports watching local television news regularly, and for those viewers who watch only local television news, the amount of election information they are likely to receive is anywhere from one minute 41 seconds of election coverage per broadcast in a midterm election to three minutes and 11 seconds per broadcast during a presidential year. Viewers of national news, however, are likely to see almost double the amount of election news per broadcast during midterm elections and two and a half times the amount.  The comparison is compounded if one considers foreign policy coverage, where network broadcasts devoted nearly 10 times the space to such coverage as their local news counterparts.

Table 1. Amount of political information by type of television program

Source: Fowler, Goldstein, Hale, & Kaplan (2007) and Hale, Fowler & Goldstein (2007)

An even greater complication to the approach suggested by Althaus and Tewksbury is that both local and network news broadcasts cannot compare to the continuous news cycle of cable outlets. As suggested by the PIPA study cited above, the source from which one receives national and international news may have dramatic effects on both what is learned and which policies are supported.  This may speak to the quality of coverage, which is a more contestable issue than we wish to tackle here. Nonetheless, measurement of cable news use should be a high priority.

The stark differences in content across different outlets would be less of a problem for measurement if the audiences for each type of program did not differ systematically from each other as well.  This, however, is clearly not the case. Regular consumers of television news programming differ on a variety of dimensions.  In particular, local TV news consumers are more likely to be from the lower end of the sophistication spectrum compared to other viewers.

Table 2. Characteristics of ‘regular’ news audiences (numbers in percentages)

Source: Pew Center for the People & the Press, Biennial Media Consumption Survey 2006. 

We believe the failure of TV news to contribute to political knowledge as observed by Althaus and Tewksbury may in part be due to the consequences of lumping all television news use – regardless of type of programming consumed – into a single question. When focusing on the news sources respondents say they primarily use, our analysis of the 2006 CCES study finds that cable news is positively related to political knowledge whereas local television news use is negatively related to learning.  It seems crucial to take into account the type of TV news use.

Finally, eliminating different types of television use has serious implications for scholars of political advertising.  The importance of campaign advertising continues to increase, as does spending on media placement and scholarly interest in its effects.  The failure to include some level of detail on exposure to different news – or entertainment programs – will not allow scholars to continue to estimate a respondent’s likely level of advertising exposure.  Although Althaus and Tewksbury acknowledged this potential limitation of their proposed approach, we would add that such questions also have value in distinguishing the audiences of the three types of news broadcasts.

We acknowledge the inherent challenges of measuring news consumption and reaffirm our appreciation of the thorough analysis that Althaus and Tewksbury have performed.  That said, we contend that the risks of altering the measurement in the ways they propose may outweigh the advantages for those interested in the differences in the content (both in paid and free media) of local, national, and cable news.

Erika Franklin Fowler is a Research Fellow, Health Management and Policy, School of Public Health, at the University of Michigan and Research Director of the University of Wisconsin News Lab. Ken Goldstein is Professor of Political Science at the University of Wisconsin and Director of the University of Wisconsin News Lab. Dhavan Shah is Maier-Bascom Professor of Journalism and Mass Communication and Professor of Political Science, University of Wisconsin.