Wednesday, November 11, 2009

MSM Agenda-Setting in the 2008 Election

What is it, what role did it play in the 2008 presidential election and where is it going?

Agenda-Setting in the Media

Maxwell McCombs and Amy Reynolds (2002) note that Bernard Cohen sums up the idea of media agenda-setting best “with his observation that the news media may not be successful in telling people what to think, but they are stunningly successful in telling them what to think about (Cohen, 1963)” (p. 1). To that end, agenda-setting theory states that the amount of news coverage a topic gets is associated with its “salience” among the news-consuming public.

The question is frequently asked of communication theorists: Is society the way it is because of the media, or are the media the way they are because of society? Which is the preexisting condition? In the earlier stages of mass communication research, the prevailing wisdom certainly seemed to suggest the former – that societal opinion was formed in response to the media. McCombs and Reynolds (2002) note the following:
(Walter) Lippman’s opening chapter in Public Opinion, which is titled “The World Outside and the Pictures in Our Heads,” summarized the agenda-setting idea, even though he did not use that phrase. His thesis was that the news media, our windows to the vast world beyond our direct experience, determine our cognitive maps of that world. Public opinion, argued Lipmann, responds not to the environment, but to the pseudoenvironment constructed by the news media” (p. 2).
We can say of the agenda-setting model that a few things seem to be quite clear. When it comes to information which cannot be obtained autonomously, we can certainly argue that our knowledge of things with which we have no direct experience is entirely limited to the information garnered through various media. For example, it is impossible for a person who has never been to (or had any direct association with) Vietnam to form uniquely autonomous opinions about Vietnam.

Every opinion that such a person might form is entirely dependent on information provided to that person via a second or third party source. In this way, we can conceptualize the validity of Lippman’s point – that public opinion of an issue is inherently tied to the media portrayal of that issue, particularly when it is fundamentally impossible for individuals to autonomously ascertain first-hand information about that event. Hence Cohen’s point – that the media tell the public what to think about.

Perhaps the most concise counter-argument to agenda-setting theory is the natural history account. McCombs and Reynolds (2002) note the following regarding natural history:
Commonly, the measures are the total number of news stories about the item and the percentage of the public citing an issue as the most important problem facing the country. This perspective is named natural history because the focus typically is on the public agenda in the rise and fall of a single item over time (p. 7)
The natural history account of media agenda-setting simply suggests that the media cover what they do because the public have a preexisting interest in the topic. A natural history theorist might say, for example, that the media covered the economy more during the 2008 presidential election because that is the issue about which the general public cared most. The agenda-setting theorist would argue that the media focused on the economy, and so the public focus was directed to the economy as a result.

Phyllis A. Anastasio, Karen C. Rose and Judith Chapman (1999) contend that, “By showing only a tiny and unrepresentative portion of the world through its window, the media may help to create the very world it seeks to reflect” (p.152). It’s clear from this statement, that Anastasio et al. share Walter Lippman’s view of media agenda-setting. But Anastasio et al. (1999) add an interesting caveat to the agenda-setting model:
Neutral media coverage of a controversial event, such as an election, often results in members of both sides of the controversy perceiving the media as hostile to their own group. Because coverage of both sides of an issue tends to emphasize differences between sides, the perceiver's own group membership is made salient and thus sets in motion the motivation to perceive the in-group as superior and the out-group as inferior. Thus, neutral coverage of the in-group is perceived as unfair and hostile in comparison with the inflated perceptions of the correctness of one's in-group.
The assertion here is that, even in the instances where media coverage is objectively neutral, consumers of that news perceive it as biased because they think it misrepresents/ inadequately represents their worldview. Anastasio et al. (1999) refer to this phenomenon as “social identity.”
Social identity is a powerful sculptor not only of perceptions, but of opinions as well. Research has shown that opinions are often influenced by other members of the in-group. Even when an in-group member presents an opinion that is unpopular and goes against one's natural inclinations, the in-group member still remains a persuasive force, much more so than any out group member.
This suggests that the source of information invariably decides how well that information is received within a specific social identity classification. For example, Americans might trust an American broadcaster more on the Afghanistan War than someone working for Al Jazeera.
Anastasio et al. (1999) also contend there are two major routes of persuasion through which attitudes and opinions are changed: the central and peripheral routes. The central route is one were an individual weighs all sides of the debate before coming to a conclusion. Anastasio et al. (1999) define the peripheral route in the following way:
The perceiver lacks either the motivation or the ability (e.g., because of time constraints or other pressing issues that drain cognitive resources) to fully process much of the message's information. When this is the case, any number of peripheral cues contained within the message may provide "mental shortcuts" that the perceiver can use to arrive at an opinion or decision.
Anastasio et al. (1999) set out to answer whether coverage emphasizing intergroup differences and intragroup similarities fueled the tendency to side with one's in-group. They found that the homogeneity of coverage greatly influences opinion to coincide with that coverage, but heterogeneous coverage results in heterogeneous opinion. The implication of their findings suggests that the media, when acting in unison, have an overwhelming influence on public opinion. Anastasio et al. conclude:
Not only do the media bias people's perceptions by offering an unrepresentative view of the world at times, but it may also facilitate biased processing of accurate information by presenting that information with an emphasis on intergroup differences… In summary, on the one hand, multiple news broadcasts that dissect the world into distinct social categories and emphasize group differences have the ability to perpetuate actual differences. On the other hand, news that obscures intergroup boundaries may have an equally great potential to diminish group differences and forge necessary connections. The media, which disseminates information and creates social norms, most likely has the power to build bridges as well as destroy them.
The implication for agenda-setting is that the influence of media coverage is relative to the consumer’s sense of self – how (and with whom) one self-identifies. In turn, the media can either feed or diffuse the divisiveness of perception through the relative divisiveness of coverage.
David Domke, Dhavan V. Shah and Daniel B. Wackman (2000) posit a slightly different view of agenda-setting which plays off of social identity, that political candidates and news media, through selection and emphasis of certain values and issues in an electoral campaign, are likely to influence which cognitions are activated as voters evaluate a political environment. Through their research, Domke et al. (2000) found:
The pattern of evidence, then, suggests that even after accounting for participants' issue positions, issue importance, age, party affiliation, and the total number of candidate attributions, discussion of issues in terms of rights and morals by politicians and news media not only increased the likelihood of attributions about candidate morality but also significantly altered the weight that participants placed on these appraisals in candidate choice.
The implication here is, as Lippman suggested, that not only do the media tell people what to think about (as Cohen asserted), but they influence how people think about specific issues. Here, Domke et al. take a slightly more coverage-centric approach to agenda-setting than Anastasio et al. – who focus primarily on who is delivering or receiving the message. The two theories can coexist with relative ease, and both seem to assert the validity of agenda-setting theory – that the media lead public opinion.

The 2008 Presidential Election

If we take the 2008 election as a case study, we can say with some certainty two things. First, that media coverage generally leads (it does not follow) public opinion. Second, the preexisting opinions and priorities of the general public don’t necessarily set the media agenda.
Media Tenor International ran a content analyses on all of the 2008 election coverage and found that, not only did the media give more favorable coverage to Barack Obama than to John McCain, but that the coverage, in many instances, seemed to spur public opinion (2008):

Significant in Media Tenor’s graphic above is the TV Tone excluding horse race issues. Naturally, coverage which focused on public opinion polls would follow or concur with opinion polling, but, with the exception of a few spikes on coverage, two things seem clear from Media Tenor’s content analysis. First, media coverage and public opinion very rarely fluctuated in opposition to one another (i.e. negative coverage did not increase for one candidate without that same candidate’s poll numbers dropping). Second, negative coverage for Mr. Obama seems to have preceded drops in his poll numbers whereas the opposite appears to be true for his spikes in opinion polling.

Interestingly, Media Tenor also found there was a disproportionate amount of coverage on horserace issues (polling):

Within the agenda-setting context, what can we say about this data? Or, more poignantly, what were the media trying to influence the public to think about and did that coverage have a causal relationship with public opinion?

First we should perhaps note that the Pew Research Center also found that the media gave more favorable coverage to Mr. Obama than to Mr. McCain. “In all, 36% of stories about Obama have been positive, vs. 35% that have been neutral. And 29% have been negative… only 14% of stories in which McCain was a significant factor were positive, while 57% were negative. The rest, another 29%, were neutral or mixed” (2008):

Second, Pew also found that the topics covered by the media tended to be those which favored Mr. Obama:

At the same time that this Pew study was released, Gallup Polling (2008) noted the following:
(The) top voter issue this year is the economy. The relevance of the economy has intensified since Sept. 15 with the extraordinary crisis on Wall Street and deteriorating consumer confidence. Gas prices, Iraq, healthcare, and terrorism remain important, but are taking second seat to the economy. Obama's perceived strengths: domestic issues, compassion, empathy, bringing about change. McCain's perceived strengths: experience, international issues, terrorism, viewed as capable commander in chief. All in all, at least in the short term, Gallup data make it clear that the uptick in negativity about U.S. economic conditions has benefitted Obama.
The conclusion we can draw from this is that the media were clearly covering those topics which favored Mr. Obama. What we can’t conclusively derive is if the media were covering these topics because of a preexisting public interest in them, or if the media coverage lead the public interest – which is the crux of the agenda-setting debate.

It may be the case that American mainstream media are so expansive and vast in their reach that doing an accurate content analysis-to-opinion poll comparison is impossible. Media Tenor certainly quantified the 2008 coverage data relative to polling, but their findings suggest that the coverage of the polling skews any connection we might make regarding the effects of the media on public opinion. Moreover, in a 24/7 news cycle where all news is not created equal (in audience or credibility), any comparative content analysis between news organizations is virtually impossible – as is finding any causal link to public opinion.

An interesting fact to note, however, is that - despite the Left-leaning coverage of the media and the eventual result of the election – 2008 CNN exit polls suggest that more Americans identified themselves as conservative (34%) than liberal (21%); these figures are basically identical to those of CNN’s 2004 exit polling. In the same vein, a recent Gallup poll found that Conservatives are the largest ideological group in the country – “40% of Americans interviewed in national Gallup Poll surveys describe their political views as conservative, 35% as moderate, and 21% as liberal” (2009).

The implication here is that more Americans lean Right than Left, but the media coverage was decidedly Left-leaning in tenor and content, and - perhaps as a result - the democratic candidate won the election. These findings seem to contradict the suggestion that media reflects rather than affects public opinion. One would expect that a mostly self-professed conservative population, if left to its own devices, would vote for the more conservative candidate. Clearly, that is not what happened.

Is it the case that looking at the media coverage before the election could have lead to an accurate prediction of the election? While it seems clear from the data that the media affected public opinion, that causal link does not appear to be the exclusive motivating factor of voting habits. Were we to take the average media depiction of the pre-2008 election environment to make a prediction about the results of that election, as Media Tenor notes, we would have concluded that Mr. Obama would win more decisively than he actually did (2008).

So, we can conclude that the media affect public opinion and that they perhaps affected public opinion enough in 2008 to make the difference in the election. But we cannot accurately say that agenda-setting serves as an accurate predictor in all cases. It seems to be the case, rather, that public opinion was just split enough in 2008 for media coverage to tip the scales in favor of Mr. Obama.

The Future of the Agenda-Setting Model

One diluting factor of the classic agenda-setting model is the increasing application of media uses and gratifications. Uses and gratifications theory contends that people use various media relative to the gratifications they seek from those media. In other words, people use media which reinforce their preexisting worldviews – which is a stance supported by the findings of Anastasio et al. Pew (2004) notes:
People who pay close attention to hard news express a preference for news that suits their point of view. Among those who follow international, national, local government, and business news, 43% say they like news with their point of view.
With an uptick in niched information sources, primarily driven by the internet and social media, Americans are increasingly presented with opportunities to consume information presented to them in a light which is favorable to their preexisting beliefs. The implication here for the classic agenda-setting model is that the mainstream media will likely experience a decrease in influence.

To a large extent, the main stream media still very much continue to frame the major debates. What they cover, by in large, still gets the most attention. But people are no longer restricted to the Walter Lippman pseudoenvironment constructed by mainstream news media. If a person really wants to ascertain information about a given topic, the number of sources available to them today is exponentially larger than in Lippman’s time. And, more than ever before, citizen journalism is beginning to drive the news. Hot topics of discussion on Twitter are making nightly newscasts, pedestrian camera phones now capture and upload current events faster than the media can even arrive on the scene and blogs are rivaling news websites in traffic.

It is increasingly the case that the traditional media are losing their monopoly on topic selection. And so we find ourselves at odds even with Bernard Cohen’s account of agenda-setting. The media are entirely dependent on ratings/circulation for profit (and, as a consequence, existence), and competition for the service they provide (information) is becoming fiercer and more niched. It’s not so much the case anymore that the media tell us what to think about… it’s more like we tell the media to show us why our worldview is correct – which feeds into the findings of both Anastasio et al. and Domke et al.

Mainstream media’s widespread influence will persist because they, by definition, have the best reach. It will always be the case that a medium with a larger base of consumers has more influence. That is much is intuitive. What remains to be seen for the agenda-setting model is how it can be reconciled with an increasingly polarized consumer base. It may be the case that the future of agenda-setting theory will focus on the affects of specific media within specific demographics. For example, quantifying the effect of New York Times coverage of specific issues and the corresponding/resulting salience of those issues within the democratic voter base.

To take it a step further, agenda-setting will play an increasing role in determining which media affect the salience of issues for which other media. Like, for example, determining how the New York Times selects the topics it covers and then quantifying which media pick up those topics after they get New York Times exposure. In this way, we can think of agenda-setting as a means for creating a sort of objective rank-order of various media relative to specific consumer bases. We’ll be able to tell which media serve as the catalysts for salience for specific demographics – a formula which has amazing business potential in the form of targeted advertising.

Communication theorists will continue to play a large role in determining which factors lead to greater salience within given populations. Despite all this, the goal of communication theory remains the same – to determine who says what to whom through which medium with what effect – and agenda-setting theory still plays (and will continue to play) a role in achieving that goal.

Anastasio, P. A., Rose, K. C., Chapman, J. (1999). Can the Media Create Public Opinion? A Social-Identity Approach. Current Directions in Psychological Science, 8(5), 152-155. Retrieved from:

CNN Exit Polling (2004). Retrieved from:

CNN Exit Polling (2008). Retrieved from:

Domke, D., Shah, D.V., Wackman, D. B. (2000). Rights and Morals, Issues, and Candidate Integrity: Insights into the Role of the News Media. Political Psychology, 21(4), 641-665. Retrieved from:

Gallup (October 22, 2008). Gallup's Quick Read on the Election.
Retrieved from:

Gallup (October 22, 2009). “Conservatives” Are Single-Largest Ideological Group.
Retrieved from:

Media Tenor (July 11, 2008). Campaign Watch: Image vs. Issues.
Retrieved from:

Pew Research Center’s Project for Excellence in Journalism (June 8, 2004). News Audiences Increasingly Politicized: Online News Audience Larger, More Diverse.
Retrieved from:

Pew Research Center’s Project for Excellence in Journalism (October 22, 2008). Winning the Media Campaign. Retrieved from:

1 comment:

Anonymous said...

very good article im writing an essay in my communication theory class on agenda setting and this really helped me compare agenda setting theory to a recent presidential election, especially one that was so unique to this country

Brawler's Search