NEW YORK (AP) — Americans value press freedom — with some exceptions.
About three-quarters of U.S. adults said press freedom is “extremely” or “very” important to society's well-being, according to a newly released Pew Research Center study. But at least half said they also support government restrictions on reporting
in specific circumstances, such as protecting national security or preventing inaccurate information.At the same time, many Americans
said it is “never acceptable” for the U.S. government to give journalists access to public figures in exchange for positive coverage, or to revoke a news organization's broadcast license.
“People are not all or nothing,” says Katerina Eva Matsa, director of news and information at Pew Research Center, when asked what was striking about the findings. “When you give them options and scenarios … they are negotiating. That was a very interesting perspective that the data gave us.”
The study, which was based on two surveys conducted earlier this year, also found that Americans were divided about whether the media is completely free to report the news, with Democrats and Republicans expressing very different levels of worry about restrictions on the press.
The analysis comes as the relationship between the U.S. media and Donald Trump's second administration becomes ever more fraught as the president, in his second term, seeks to restrict press coverage not to his liking. He or his administration has attacked, sued or taken administrative action against media outlets that have drawn his ire.
Matsa said that tension was one impetus for doing the report now. Other reasons: the changing media environment and the evolution of what “the press” is now, as well as the emergence of tech companies as gatekeepers of the news.
“Overall, there’s a lot of public conversation — and it’s the 250th anniversary of our nation,” she said. “So, it kind of felt like a good moment to look at how Americans think about this.”
Most Americans, about 7 in 10, said the Trump administration and the U.S. news media have a “very” or “somewhat” bad relationship. That's a slight increase from 2025, when 64% of U.S. adults said this.
They're more divided on whether the press faces additional restrictions under Trump's second term, although few think the press is freer than it was before. About 44% of U.S. adults said the press has become less free under the second Trump administration, while about 26% said there has been no change. Only 12% said the press is more free, and 17% said they're not sure.
Despite U.S. adults' broad belief in the ideal of press freedom, the survey found, they're willing to contemplate restrictions in specific circumstances.
Most U.S. adults said the government should not be able to require a news organization to take down a story if it is critical of the government or clearly favors one political side. But about half say it should be able to force a retraction if a story contains false information, and about 6 in 10 said this about a story that poses a risk to national security.
The findings indicate most Americans are open to some form of government restriction on the media, despite their broad embrace of press freedom as a concept. At the same time, there is unlikely to be uniform agreement about what constitutes a national security risk, for example, or who gets to determine that — and the poll did not probe those distinctions further.
When asked whether it was more important for the press to be free to report news even if some reporting is inaccurate, or if preventing inaccurate reporting should be prioritized even if it limits press freedom, about half, 53%, of U.S. adults said it was more important to prevent inaccurate reporting, and 45% said press freedom was more important.
But they see limits for government too. Just over half of Americans say it is “rarely” or “never” acceptable for the U.S. government to force journalists to reveal the identities of unnamed government sources or take away a news organization's license to broadcast. About 2 in 10 say these actions are “sometimes” acceptable, and only about 1 in 10 say they are “usually” or “always” acceptable.
Perhaps not surprisingly in a highly politicized environment, views on press freedom differed sharply according to political leaning.
Democrats were more likely than Republicans to prioritize a free press to begin with, the study found. About 8 in 10 Democrats said a free press is “extremely” or “very” important for society's well-being, compared to about two-thirds of Republicans.
Republicans were also more likely than Democrats to say the government should be able to require a news story be taken down in cases of national security. About 7 in 10 Republicans said this should be allowed, compared to about half of Democrats. A similar gap emerged on the publication of false information.
The share of Americans who said the media was “completely” free to report the news fell slightly from 33% last year to 28% in the latest survey. About half — 48% — said the press was “somewhat” free and about 2 in 10 said it was “not very” or “not at all” free.
There were differences between the parties there, too.
—In the new study, only about 2 in 10 Democrats saw the media as “completely” free to report the news.
—Meanwhile, most Democrats, 64%, said they were “extremely” or “very” concerned about potential restrictions on press freedom, while only about 3 in 10 Republicans held similar worries. The reverse was true during the Biden era.
The report is accompanied by a knowledge quiz based on the surveys that users can take to test how much they know about the First Amendment. It has eight questions, all of which were included in the survey. Matsa noted that while many Americans understand the basics of the First Amendment, many don’t know the mechanics of who it protects — and from whom.
For example, the study found that while 58% of U.S. adults know the Constitution protects the press from interference by the U.S. government, only 33% know that it does not protect the press from interference by private companies, such as a social media company or a web-hosting service.
___
The Pew-Knight analysis was based on two polls conducted using samples drawn from the Pew Research Center's American Trends Panel, which is probability-based. One survey of 3,507 U.S. adults was conducted March 23-29, 2026, and the other was conducted among 3,488 U.S. adults from June 22-28, 2026. The margin of sampling error for the March survey of 3,507 adults is plus or minus 1.9 percentage points. The margin of sampling error for the June survey of 3,488 U.S. adults is plus or minus 1.8 percentage points.
