The Two Levels of Agenda-Setting: How Media Dictates Public Attention and Perception

A citizen consuming news for 45 minutes a day encounters roughly 150 distinct headlines, yet the global information apparatus generates over 3 million articles in that same 24-hour window. The mechanism that filters those 3 million potential realities down to the 150 that reach the screen is not random; it is a structural sorting process known as agenda-setting. By allocating finite space and time, editorial boards establish the boundaries of public attention.[6]

The foundational mathematics of this mechanism were established in 1972 by researchers Maxwell McCombs and Donald Shaw. Analyzing undecided voters in Chapel Hill, North Carolina, during the 1968 presidential election, they measured the frequency of issues covered by local media against the issues voters identified as the nation's most critical problems. The resulting correlation was +0.967—a near-perfect statistical alignment between the press's priorities and the public's concerns.[1]

This initial discovery formalized the first level of agenda-setting theory: object salience. The press, as political scientist Bernard Cohen observed, "may not be successful much of the time in telling people what to think, but it is stunningly successful in telling its readers what to think about." The sheer volume of coverage dictates an issue's perceived importance.[1][6]

By allocating finite resources—front-page placement, broadcast minutes, and typographic weight—editors signal the relative importance of an event. If a television network dedicates 14 minutes of a 22-minute evening broadcast to a border dispute, the audience structurally absorbs the dispute's primacy, regardless of their personal stance on the border itself.[5]

However, media analysis over the subsequent decades revealed that the 1972 model was incomplete. The press does not merely hand the public a blank object to consider; it hands them a painted one. This realization birthed the second level of agenda-setting: attribute salience.[7][8]

Second-level agenda-setting dictates how to think about the object by emphasizing specific traits while obscuring others. When covering a tax policy, a news organization might highlight its impact on the federal deficit, which represents an economic attribute, or its effect on working-class families, which represents a social equity attribute.[2][7]

"The media not only tell us what to think about, they also tell us how to think about it, and consequently, what to think," researchers noted in the Journalism & Mass Communication Quarterly. By selecting which attributes define the narrative, the media primes the audience's evaluative criteria before the audience has independently processed the event.[7]

This attribute selection overlaps heavily with framing theory. As outlined in the Journal of Communication in 2007, framing involves constructing a narrative that promotes a particular problem definition, causal interpretation, and moral evaluation. If a protest is framed through the attribute of "civil rights," it generates a fundamentally different public response than if it is framed through the attribute of "public disruption."[2][5]

The architecture of agenda-setting has fundamentally shifted with the digitization of news. Historically, the agenda was set by a concentrated oligopoly of broadcast networks and national newspapers. Today, the process is driven by "intermedia agenda-setting," where news organizations continuously monitor and replicate each other's coverage priorities to maintain competitive relevance.[3][8]

A 2019 study published in the International Journal of Communication examined this intermedia dynamic during a major policy reform debate. The researchers found that elite national outlets still act as the primary nodes, establishing the initial object salience. Regional and digital-native outlets then adopt those exact priorities within a 24-to-48-hour window.[3]

This homogenization was quantified in a 2024 analysis of the 2016 and 2020 U.S. presidential elections, presented at the International AAAI Conference on Web and Social Media. The data revealed that despite the proliferation of independent digital publishers, the core agenda remained highly centralized across the ecosystem.[4]

The AAAI researchers tracked millions of URLs and found that partisan media ecosystems, rather than setting entirely different agendas, largely adopted the same first-level objects—the events themselves—but applied radically divergent second-level attributes.[4]

This creates a fractured public sphere where opposing political factions agree on what the crisis of the day is, but possess entirely incompatible frameworks for evaluating it. The shared reality of the 1968 Chapel Hill study has been replaced by a shared syllabus with contradictory textbooks.[1][4][6]

The introduction of algorithmic distribution further complicates this mechanism. Social media platforms do not generate original reporting; they re-weight the salience of existing reporting based on engagement metrics. This creates a feedback loop where the audience's reaction to an attribute dictates its future visibility across the network.[8]

Consequently, the power to set the agenda is no longer the exclusive domain of the editorial board. It is a negotiated settlement between the publisher's initial attribute selection, the algorithmic amplification of outrage or affinity, and the intermedia network's willingness to replicate the narrative. The structural power remains, but the actors wielding it have multiplied.[3][8]