American Government › Section H · The media › Topic H.4

The impact of the media

The media rarely tell people what to believe outright, but they strongly influence which issues seem important, how problems are understood and how leaders are judged. Knowing these effects, and the biases built into news, helps citizens read coverage critically.

Do the media change minds?

This topic asks what difference the media make. In the 1920s the journalist Walter Lippmann argued that people rely on the press for "the pictures in our heads" of a world they cannot see directly. Researchers have argued ever since about how strong that influence is. Their answers have moved from very strong, to weak, to a middle position: the media have real but conditional effects, which depend on what audiences already know and believe, and are strongest when people have little first-hand experience of an issue.

Three generations of effects research

Early observers feared a hypodermic effect, in which messages are injected straight into passive audiences; wartime propaganda seemed to support the idea. Studies of the 1940 election by Paul Lazarsfeld and colleagues found instead minimal effects: voters' views were fairly stable, and media messages often reached them through a two-step flow, filtered by opinion leaders such as friends and co-workers. From the 1970s, George Gerbner's cultivation theory proposed that heavy, long-term television viewing slowly shapes beliefs, for example leading frequent viewers to see the world as more dangerous than it is. Today most scholars stress three subtler effects: agenda setting, framing and priming.

Agenda setting

Agenda setting is the media's power to influence which issues the public considers important, by choosing what to cover and how prominently. The political scientist Bernard Cohen observed in 1963 that the press may not often tell people what to think, but it is remarkably good at telling them what to think about. Maxwell McCombs and Donald Shaw tested the idea among undecided voters in Chapel Hill, North Carolina, during the 1968 campaign. The issues those voters ranked as most important closely matched the issues local and national media had emphasised. Later studies found the same pattern in many countries and for many issues.

Framing

Framing is the way a story is presented: which facts, causes and images it stresses. The same event can be framed as an economic, moral or security story. Shanto Iyengar distinguished episodic frames, which focus on a single case or event, such as one family losing its home, from thematic frames, which place an issue in a wider context, such as trends in housing costs and policy. His experiments found that viewers of episodic stories more often blamed individuals, while viewers of thematic stories more often held government or society responsible. Word choice frames too: calling a tax an "estate tax" or a "death tax" invites different reactions.

Priming

Priming means that by emphasising some issues, the media influence the standards people use to judge leaders. In experiments reported in News That Matters (1987), Shanto Iyengar and Donald Kinder showed viewers newscasts stressing one problem, such as defence or inflation. Those viewers then gave that problem more weight when rating the president. Priming follows from agenda setting: if coverage makes the economy seem the main issue, voters are likely to judge a president mainly on the economy. This is one reason campaigns and presidents fight hard to keep attention on issues where they are seen as strong.

Worked example: one story, three effects

Over a month, a network leads its newscasts with factory closures in several states. 1. Agenda setting: viewers begin to name jobs as the country's most important problem, though unemployment has barely changed. 2. Framing: if reports show one laid-off worker's struggle, the frame is episodic and invites sympathy for that person; if they chart trade and automation trends, the frame is thematic and points towards policy. 3. Priming: when asked about the president, viewers now weigh his record on jobs more heavily than before. 4. Check: none of these effects tells viewers what to conclude, yet each can shift opinion. 5. Lesson: notice what is covered, how and how often.

Bias: partisan and structural

Many Americans see the media as biased, and views split sharply by party. A Gallup poll of September 2025 found that only 28 percent of adults trusted the mass media a great deal or a fair amount, a record low, including 51 percent of Democrats and 8 percent of Republicans. Some outlets openly take an ideological line, especially in opinion programming. Researchers also identify structural biases that affect outlets of every leaning: a preference for conflict, negativity, novelty and vivid personalities over slow-moving policy detail. These come from the need to attract audiences, not from any party loyalty.

Covering campaigns and government

Campaign coverage leans towards horse-race journalism, stressing who is ahead in polls and fundraising rather than policies. Sound bites have shrunk: Daniel Hallin found that the average quotation of a presidential candidate on network news fell from about 43 seconds in 1968 to about 9 seconds in 1988. Pack journalism, reporters following the same story at once, can magnify gaffes and momentum, especially in early primaries. In government, the president receives far more coverage than Congress and can go public, appealing directly to citizens to pressure lawmakers, a strategy described by Samuel Kernell. Members of Congress compete for attention through local media, cable appearances and social media.

Coverage and society

News coverage also shapes how groups are seen. Heavy local reporting of crime can leave audiences thinking crime is rising even when it is falling, consistent with cultivation theory. Martin Gilens found that in the second half of the twentieth century, news magazines illustrated stories about poverty with pictures of Black Americans far more often than their share of the poor, which he linked to more negative public views of welfare. Studies of women candidates have often found more attention to appearance and family than for men, though some recent research finds the gap has narrowed. Such patterns can affect who runs for office and which policies seem urgent. Online, misinformation (false claims shared in error) and disinformation (deliberate falsehoods) spread fast; a 2018 Twitter study by Soroush Vosoughi, Deb Roy and Sinan Aral found false stories travelled farther and faster than true ones.

Worked example: weighing a charge of bias

A viewer says a network's election coverage was biased against her candidate. 1. Define the claim: partisan bias, meaning coverage systematically favoured one side. 2. Consider the hostile media effect: in a 1985 study, Robert Vallone, Lee Ross and Mark Lepper showed the same news coverage to opposing partisans, and each side judged it biased against them. 3. Look for evidence beyond impressions: count the tone of stories about each candidate, the time each received and the sources quoted. 4. Rule out structural bias: negative coverage of a front-runner may reflect horse-race habits rather than partisanship. 5. Conclude carefully: bias claims are testable, and perception alone is not proof.

Common mistakes to avoid

First, assuming the media inject opinions directly into audiences; effects are real but conditional. Second, confusing agenda setting, which concerns what people think about, with framing, which concerns how an issue is presented. Third, mixing up priming with persuasion; priming changes the standards used to judge, not views of the facts. Fourth, treating every complaint of bias as evidence; partisans on both sides tend to see neutral coverage as hostile. Fifth, overlooking structural biases, such as the pull towards conflict, which affect all outlets.

Key takeaways

Media effects research moved from the hypodermic model to minimal effects and the two-step flow, then to cultivation theory. Agenda setting shapes what people think about (McCombs and Shaw, 1968); framing shapes how issues are understood, with episodic and thematic frames assigning blame differently (Iyengar); priming shapes the standards for judging leaders. Trust in the media is low and split by party (Gallup, 2025: 28 percent). Structural biases favour conflict, negativity and the horse race; sound bites have shrunk. Coverage can distort views of crime, poverty and candidates, and the hostile media effect colours perceptions of bias.

Flashcards

Flashcard

Hypodermic model

Media messages injected directly into passive audiences

Early fear, later rejected.

Test your understanding

Question 1 of 5

After weeks of heavy coverage of school shootings, more people name school safety as the country's most important problem. Which effect is this?

Question 2 of 5

A report on unemployment follows one laid-off worker and her family. According to Iyengar, how are viewers likely to respond?

Question 3 of 5

Supporters of two rival candidates watch the same news segment, and each group says it was biased against their side. What does this illustrate?

Question 4 of 5

News coverage stresses foreign policy for months, and voters then judge the president mainly on foreign policy. Which effect is this?

Question 5 of 5

What did Daniel Hallin find about sound bites on network news between 1968 and 1988?

← Previous topicNext topic →