Donald Trump talks with reporters before boarding Air Force One in February, 2020. For this study, researchers presented participants with three different kinds of stories based on a claim made by the president and then asked them to rate the accuracy of the claim. (AP Photo/Evan Vucci)
News stories presented as fact-checks could help readers update their beliefs more accurately than conventional news articles, but such stories are also perceived to be more biased, new research suggests.
The study, published June 1 in Mass Communication and Society, explores whether explicitly labeling a story as a "fact-check" could help people form a more accurate understanding of information in the article. This label served as a signal to study participants that "this isn't just a story about politics; it's a story about the truth," Michael Wagner, a professor at the School of Journalism and Mass Communication at the University of Wisconsin–Madison and the study's senior author, told The Academic Times.
"We found that literally calling out that the purpose of a story is to engage in a fact-check does help — people get more accurate," Wagner said. "But it comes at a cost, which is that people find the story to be biased or the journalist to be biased."
Previous research established that the inability to accurately update our beliefs is partly the work of motivated reasoning, a mental mechanism in which we take new information and make it support what we already believe. According to this framework, people are either motivated to be accurate or motivated directionally — in other words, to support their political team.
"If you're a Green Bay Packers fan, and you're watching a Packers game, you're directionally motivated when the referee makes a close call; you want it to go the Packers' way," Wagner explained. "But if you're watching the Raiders play the Falcons, you're probably more accuracy-motivated: You look at the replay and say, 'Oh, yeah, he fumbled,' whereas you're less likely to do that when it's somebody on your team."
For the first of a pair of studies, the researchers used Qualtrics, a third-party survey administrator, which recruited 510 people for a sample intended to reflect the demographics of the U.S. The participants were shown corrective information in three formats: a conventional news story, a fact-check with no rating and a fact-check with a rating. All three stories pertained to a claim made by former President Donald Trump about gun violence in Chicago. Then, participants were asked to rate the accuracy of the claim and the level of bias they detected in the stories.
The fact-checking label moved the accuracy of people's understanding of news stories from below to above the midpoint on a five-point scale, an effect that was not moderated by news source or partisan identity. Despite their best efforts, the researchers couldn't "knock down" the effect with alternative explanations, Wagner said.
"What if it's from the [Associated Press] or Fox News or another source? What if we account for people's strength of partisanship? Those things still don't diminish the fact that a fact-check helps people become more accurate in assessing the verifiable truth in a news story," he said.
Wagner acknowledged that asking about the accuracy of claims made by a highly polarizing figure such as Trump could create an effect not specific to people's perceptions of fact-checking, but he argued that it's important to incorporate real-world conditions in experiments.
"We wanted to show people a real fact-check," he said. "This is a claim that Trump made that had been fact-checked by numerous organizations. We also thought it was a really high bar: If we can find an effect where people who are strong supporters of Donald Trump saw a fact-check that corrected him, and then they updated their attitude about what the truth is, that would be a strong finding, and we could be pretty confident that it wasn't just a Trump effect."
For a second study involving 652 participants who were also recruited through Qualtrics, the researchers used a similar story about gun violence in Chicago that shifted the focus from Trump to congressional leaders. The second study replicated the findings of the first, showing a boost in understanding but also an increased tendency to believe the news source was biased.
Overall, the researchers found that providing a rating, such as PolitiFact's Truth-O-Meter, did not have an effect on updating beliefs — a surprising finding for Wagner and his colleagues, who thought that ratings like "Pants On Fire" would encourage motivated reasoning.
"If I'm a liberal and see that Trump has a Pants On Fire, I really want to keep believing that. Or if I'm a conservative, I might want to do all that I can to push back against that rating," he said. "But we didn't find that was statistically discernible from any other rating."
Even though news consumers came away from fact-checks with a more accurate understanding across the two studies, they also perceived stories in the fact-check genre — and the journalists who wrote them — to be more biased than conventional news articles. Basically, hostile perceptions increased if the story was labeled as a fact-check.
Researchers have known for years that strong partisans perceive neutral stories as being hostile to their point of view, in what's known as the hostile media effect. Studies have also shown evidence of a relative hostile media effect, in which news content that favors one side or another — even if it's accurate — is perceived as more biased by people whose viewpoints tend to conflict with that side than by people who tend to agree with the messaging.
"We thought a fact-check goes one step beyond that; it's not journalism that goes, 'Here's what we can tell you about the verifiable truth, thank you very much.' This is somebody making a claim, and [reporters] are telling you whether it's true or not," Wagner said. "We thought people would interpret this through the lens of the relative hostile media effect, and that is what we found."
"It's sort of depressing to see that fact-checking can come at this kind of cost," he added.
Given the rise of political polarization and the proliferation of online disinformation, Wagner believes it's important to show whether it's possible for people who don't benefit politically from a fact-check to accept it as the truth, and how news consumers perceive organizations that publish such stories.
"Perhaps now more than ever, or at least more than for the last 50 years, having a good sense of how people decide what's true is really important for the health of democracy," Wagner said. "We have lots of leaders making competing claims about basic fundamentals: The veracity of elections — are they free and fair? Is an attack on the Capitol building while votes are being counted a problem or not? — these are really the barest of requirements for a democracy."
The study, "The power of a genre: political news presented as fact-checking increases accurate belief updating and hostile media perceptions," published June 1 in Mass Communication and Society, was authored by Jianing Li, Omar Dumdum and Michael Wagner, University of Wisconsin–Madison; and Jordan M. Foley, Washington State University.