Share this:

Tyler Johnson 80x108Paul Kellstedt 80x108How do variations in the consumption of media across Americans affect policy attitudes at the individual and aggregate levels? Tyler Johnson and Paul Kellstedt examine how newspaper readership and television viewership choices shape liberalism or conservatism towards spending decisions made by government from the mid-1970s to the mid-2000s. They find that infrequent newspaper readers exhibit greater variance in opinion and, as a result, their opinions move in less parallel fashion over time compared to those who sometimes or always read newspapers. Television viewership rates, however, do not appear to affect opinion at the individual or aggregate level.

Americans are avid consumers of media. As print fades, television evolves, and the internet and social media rise, how this consumption takes place is changing rapidly, and, given technological advancements, apparently will continue to do so well into the future. Despite such changes, the central role of the media as informer and entertainer in the lives of most Americans cannot be denied.

While the media may be the primary source of information for most Americans, we know that not all Americans consume media at the same rate. Data from the General Social Survey (GSS) show that some citizens continue to subscribe to and read newspapers on a daily basis, while others do so a bit more haphazardly throughout the week. There are also those who rarely if ever pick up a copy from a newsstand let alone their front porch. Some Americans turn on the television for upwards of 6-8 hours a day or more, while others watch a few hours a day if that. Some even report living a television-free lifestyle altogether.

Media effects research conducted across the past five decades has taught us a great deal about how individual exposure to different types of content produced by media has the potential to shape American policy attitudes, illustrating how the media set agendas, prime evaluations, and frame the components of policy debates. Such research has spoken very little though to the potential explanatory power for public opinion of the variation that exists across Americans when it comes to media consumption. Our research shows such variations do affect public opinion, specifically variation in newspaper consumption.

Using GSS data from 1984 to 2004, we examine the extent to which individuals took liberal or conservative positions on spending issues related to the environment, mass transportation, social security, highways and bridges, welfare, military and defense, racial and ethnic minorities, education, big cities, health care, and space exploration. Looking at any single individual’s positions on these issues overall rather than issue by issue allows us to make a broader statement about said individual’s liberalism or conservatism in policy (giving us an individual-level metric akin to James Stimson’s concept of Policy Mood, which captures the American public’s overall liberalism or conservatism on hundreds of policies in total at any given point in time). It also offers insight into any individual’s belief system.

Credit: Daniel R. Blume (Creative Commons BY SA)

Credit: Daniel R. Blume (Creative Commons BY SA)

We discover that those who read newspapers daily have more stable and coherent attitudes on government spending than those who sometimes or never read the newspaper. The more one reads a newspaper, the less uncertainty one has when it comes to their beliefs in total on whether or not the government should spend more, less, or about the same amount on important policy areas tackled by elected officials on a regular basis. This consistency in consuming newspapers may arm individuals with the information they need to develop and maintain a belief system on what the federal government should and should not do financially. Unlike with differences in newspaper readership across individuals, differences in television viewing rates from one person to the next have no impact on opinion stability.

The aggregated opinions of those who read newspapers less than once a week also move a bit differently over time from those who read newspapers several times a week or those who read newspapers every day. This is revealed by examining American opinion on the policy areas mentioned earlier as well as on affirmative action, busing, capital punishment, how courts deal with criminals, income inequality, gun laws, assistance to African Americans, government intervention, helping the poor, helping the sick, housing, and income taxes from the early 1970s through the mid-2000s. Once again, rather than analyze these policy areas individually, we combine them into a single measure akin to the aforementioned Policy Mood.

When those who read the newspaper every day become more liberal or conservative on policy, so too do those who read the newspaper a few times a week; in fact, the opinions of these groups correlate at 90 percent over the 30 year period studied. Those who read the newspaper once a week or less act similarly, but not as similarly; the extent to which they are liberal or conservative on policy correlates with the daily and multiple times per week newspaper readers at 74 and 71 percent respectively. Were we to ask ourselves “who moves Policy Mood?” across all Americans, we could argue that since the Mood of those who rarely if ever consume newspapers moves a bit less glacially than others, these individuals are slightly more likely to account for overall changes in liberalism and conservatism on policy than those who more regularly read the newspaper.

Similar conclusions cannot be drawn when it comes to the Policy Mood of different types of television viewers over the same thirty year period. The aggregated policy attitudes of individuals who watch 0-1 hours of television per day look incredibly similar to those who watch 2-3 hours of television per day or 4 or more hours of television per day. All three series are correlated at 87 percent or above. Policy attitudes of these three subgroups are also driven by causal dynamics and economic indicators like unemployment and inflation in similar fashion.

The findings here fit into a broader tradition of research in recent decades that the power of the media to shape opinion is not powerful or minimal, but rather subtle. Not all media consumption shapes opinion (as evidenced by our lack of findings on television consumption), and primarily it is those who rarely if ever read newspapers whose opinions are most effected by their choices as to what to consume and when.

This article is based on the paper “Media Consumption and the Dynamics of Policy Mood” in Political Behavior.

Please read our comments policy before commenting.

Note:  This article gives the views of the author, and not the position of USApp– American Politics and Policy, nor of the London School of Economics. 

Shortened URL for this post:


About the authors  

Tyler Johnson 80x108Tyler JohnsonUniversity of Oklahoma
Tyler Johnson is an Assistant Professor in the Department of Political Science at the University of Oklahoma. His research interests lie at the intersection of American political behavior and institutions, using elite activity and norms to explain aggregate/individual opinion and media coverage.


Paul Kellstedt 80x108Paul Kellstedt – Texas A&M University
Paul Kellstedt is an Associate Professor at the Department of Political Science at Texas A&M University. His current research projects include work on economic evaluations in the mass public and an examination of the nature of policy mood.