DISSERTATION ABSTRACT

There are countless problems in the world, yet the public agenda is limited. The New York Times, in particular, has about eight front-page articles a day.  The policy issues captured in these articles send critical cues to politicians and citizens alike about which issues are important and, by exclusion, which are not. We know media attention can influence opinion and policy, but we know relatively little about the mechanisms by which this attention is distributed.  How do issues become front-page news?

I develop a theory of media dynamics to explain patterns of front-page attention.  The media, as an institution, does not process the onslaught of daily information about real-world problems perfectly or efficiently.  With too much information, not enough agenda space, and a multitude of constraints directing the news-selection process (such as the need to sell newspapers and incentives to appease powerful policy entrepreneurs), the media processes information disproportionately.  That is, it does not pay attention to problems in proportion with their severity, nor does it shift gradually over time to accommodate new problems as they develop and phase out old problems as they are solved.  Instead, most of the time media attention gets “stuck” in equilibrium, with the agenda looking more or less the same, day after day.  Meanwhile, however, information continues to unfold in other issue areas.  When the information build-up becomes too large to ignore, attention lurches to the new problem, completely overhauling the previous status quo. 

The result is that the front-page agenda is skewed in two important ways. First, attention is disproportionately distributed across issues; a few issues get the majority of news coverage, while most receive close to none.  Second, changes in attention are disproportionately distributed; the agenda does not shift incrementally, but rather displays periods of relative stasis punctuated at intervals by dramatic change.

Against this foundation of media information processing dynamics, I identify eight key variables that directly affect which issues become front-page news, how much attention each issue receives, and how the agenda changes over time: real-world events, prior attention, front-page congestion, scope of discussion, journalistic obligations and norms, entrepreneurial activity, public opinion, and political context.

I test my theory of how media attention is distributed using an original dataset of all Times front-page articles coded by issue, from 2000 to 2005 (some 18,000 stories in all).  I document the disproportional distribution of attention across policy issues and the disproportionate nature of how the agenda changes over time.  And I develop a statistical model of front-page attention that confirms my expectations about the effects of specific variables on the news-selection process.

I give special examination to the scope of discussion—that is, the degree to which a policy debate is concentrated on a narrow number of dimensions or spread across a wide range of dimensions—and the relationship of this variable with front-page attention.  My analysis suggests that the scope and attention have a mutually reinforcing influence on one another.  Increasing attention paid to an issue offers more “room” for the issue to expand in scope.  At the same time, when the scope of a debate widens, the additional perspectives makes the issue more salient to a greater number of people (citizens, policy entrepreneurs, and journalists alike), and the issue draws more attention.   Under the right conditions, the reinforcing link between an expanding scope of discussion and an increasing amount of attention can lead to a cascade of attention, completely displacing the previous status quo.

Taken together, the findings of this study show how front-page attention, while impossible to predict on a day-by-day basis, exhibits strong and predictable patterns over time.  Most of the time, institutional constraints leave media attention in a holding pattern, fixating on the current “hot” topics.  But this equilibrium will always be broken by change—it’s just a matter of when, and by what issue.  Most importantly, when the agenda changes it does so explosively, often fueled by changes in specific event and non-event variables in the system, including the scope of discussion.  What all these findings mean is that front-page attention is very, very difficult to control.  Policy entrepreneurs hoping to manage the spin of an issue will be sorely disappointed.  But for policy advocates wanting to raise awareness of a public problem, the best strategy is clear: instead of saying the message louder, diversify the message itself.