A new way of looking at trust in media: Do Americans share journalism’s core values?

The deep divides over trust in the news media are usually portrayed as largely ideological. Democrats are seven times more likely than Republicans to say they trust the mainstream media, and independents are four times as likely.[ref The General Social Survey (GSS) has been tracking confidence in the press since 1973. In 2018, only 6% of Republicans say they have “a great deal of confidence” in the press, compared with 12% for independents and 21% for Democrats. The GSS is a project of the independent research organization NORC at the University of Chicago, with principal funding from the National Science Foundation. In addition, Gallup has been tracking trust in news media since 1972. Its annual Governance poll found that only 10% of Republicans say they have even “a fair amount of trust” in the news media, compared with 36% for independents and 73% for Democrats. Megan Brenan, “Americans Remain Distrustful of Mass Media,” Gallup.com, September 30, 2020.] But the argument over media trust often has the feel of people talking past each other—many journalists denying they slant the news to help one party over another, while many of their critics, especially on the right, scoff at that denial.[ref In another Media Insight Project study, results showed that 38% of adults believed that the personal biases or political opinions of a journalist influence their decision about if or how to cover a story.] Still others, particularly on the left, question whether some basic notions of journalistic independence and open-minded inquiry are a delusion and the press should become more strictly partisan.[ref In a Knight Foundation/Gallup study, 43% say journalists’ biases are so overwhelming it can be difficult to sort out the facts.]

A major study released today by the Media Insight Project, a collaboration of the American Press Institute and The Associated Press-NORC Center for Public Affairs Research, opens up a new way of looking at the issue of media trust and may offer new avenues to address it.

The study finds that not all Americans universally embrace many of the core values that guide journalistic inquiry. And uneasiness with these core values of journalism is more connected to people’s underlying moral instincts than to politics.

When journalists say they are just doing their jobs, in other words, the problem is many people harbor doubts about what the job should be.

The study tested public attitudes toward five core values of journalistic inquiry that many journalists consider fundamental. We identified these principles based on previous studies and input from a group of journalists. These core journalism values include such ideals as it’s vital for a free society to monitor the powerful to keep them from misbehaving, and the press should be a voice for the less powerful in society.

In all, only 11% of Americans unreservedly embrace all five of the journalism principles tested and these people tend to be politically liberal. However, most Americans don’t fully endorse these journalism principles, and the distrust goes beyond traditional partisan politics.

How people view those core values of journalism, moreover, is closely associated with deeper feelings they have about what moral values are important generally.

People who put more emphasis on the moral values of loyalty and authority, for example, tend to be more skeptical of some of the core values journalists try to uphold, or at least worry that these values could be taken too far. People who put more emphasis, by contrast, on the moral values of fairness for all and caring for the less fortunate tend to be more aligned with core press values. These differences persist even when we control for a person’s political partisanship and ideology.

These moral differences also influence what kinds of news stories people think the press should emphasize and how stories should be framed.

The differences are subtle and cannot be dismissed as another case of political or ideological divide. We find, for instance, some people often associated with having more liberal political views (such as Democrats, women, or people of color), are hesitant about some core journalistic values. And there are some core journalistic values that do resonate among conservatives. Education level also correlates with how people respond to some of the values journalists hold dear.

Rather than distrust toward the media being tied only to the perception of partisan bias, the problem at the heart of the media trust crisis may be skepticism about the underlying purpose and mission journalists are trying to fulfill in the first place.

The results of the study shed light on why the debate over trust in the news media has long seemed so intractable, with journalists believing they are just doing their jobs and critics seeing clear signs of political leaning and the denials of journalists as proof of dishonesty.

The findings also point to some changes journalists can make in the way they report that could help rebuild trust. Journalists may be able to win the trust of skeptical audiences by reexamining some basic notions of what is important, the story mix, what themes stories touch on, by broadening how those stories are framed, and what values are emphasized in headlines.

As an example, the research suggests stories that are focused largely on care and fairness are more likely to appeal to people with liberal instincts. Stories that talk about heroes and loyalty, meanwhile, are more likely to resonate with people who identify as conservatives. At the same time, people who define themselves as conservatives also like stories that touch on themes of care and fairness, but often not as enthusiastically.

In tests we conducted, small changes that added in different values and perspectives in addition to care and fairness made news stories appeal to a much wider range of audiences.

The findings build on a growing body of social and cultural research called Moral Foundations Theory. At its core is the idea that different people instinctively respond more strongly to certain moral values than others—such values as care, fairness, loyalty, authority, and purity.

Separately, our study identified five core principles that journalists value highly, such as the importance of spotlighting problems in society as a way to solve them, or the notion that more facts are likely to get us closer to the truth.

Some critics blame the trust crisis on the growing ecosystem of far-right media outlets that market themselves as an antidote to a mainstream press they deride as biased and liberal. And indeed, the rise of talk radio in the 1980s and later ideological cable news coincided with the long-term decline in media trust among conservatives. But the findings here suggest the roots of the dilemma go deeper and connect to some of the verities the press embraced as it split from partisanship more than a century ago.

The study also did not attempt to probe how or why people believe information that has been repeatedly established by evidence or even courts to be false. This is not, at heart, a study of people’s media behavior, but an attempt to get at underlying attitudes that have made the crisis in trust seem so intractable, the media landscape so fragmented, and the prospects of a public square with a common set of facts so at risk.

Among the findings:

• One journalism principle—factualism—generally wins more public support than the others tested. The idea that the more facts we have in society the more likely we are to solve problems was fully embraced by 67% of Americans. In contrast, social criticism, the idea that shining a spotlight on problems is the best way for society to solve them is the least endorsed (29% of Americans support it).
• The trust crisis may be more rooted in people’s moral values than their politics. We found the participants in our study fell into four distinct clusters based on similar moral values and attitudes toward core principles of journalism. Only one of those clusters leans clearly partisan—a liberal, mostly Democrat cluster of people who strongly endorsed journalism values. The other three groups are composed of a mix of party affiliations and ideologies, and have varying degrees of hesitation about some core journalism principles.
• The one group of Americans that showed the strongest support for the core journalism principles we tested was the smallest. This group—whom we call Journalism Supporters—make up only 2 in 10 of those surveyed. That number suggests that some of the traditional framing journalists bring to stories, and many of the traditional marketing appeals journalism organizations use that trumpet traditional journalism values, will only reach so far in rebuilding trust or winning new subscribers. The people in this cluster tend to put a higher emphasis on two of the five moral values: caring for others and fairness, values that closely align with what the press considers its core professional values. A great majority of this group think the news is accurate (83%) and a majority think the news media are trustworthy (58%). But even here there are reservations. Less than half of this group thinks the media are moral (26%) and only a quarter believe they care about people like them (24%). This group is also the most liberal of the four clusters.
• People who put more emphasis on authority and loyalty tend to be more skeptical about fundamental journalism principles. These people—whom we call Upholders—put a high value on respect for leaders and groups. They worry that some of the things journalists believe in can be intrusive and get in the way of officials doing their jobs. This group would like to see more stories about what works, not just what is going wrong. Upholders however, actively seek out and consume a lot of news. Most of them (60%) also think the news they consume is accurate. At the same time, they evince some deep reservations about the press and its values. Only 33% believe the news media in general are trustworthy. Even fewer (15%) think the press cares about them, or is moral (13%). Ideologically, Upholders are evenly split between conservatives and moderates (43%); another 14% call themselves liberal. Politically, half are Republicans, 3 in 10 are Democrats, and 2 in 10 are independents.
• Another group, people who care deeply about all five moral values, are generally supportive of journalism principles, but that support is not unqualified. This group, whom we call the Moralists, score higher than any other group on all five moral values. They are also a political and ideological mix, predominantly moderate and slightly more Democrat than Republican. They are also the most diverse group by age and race. And while the majority of Moralists think the news is trustworthy (51%) and accurate (74%), Moralists share some deep suspicions about the press. Only 2 in 10 think the news media care about people like them (20%) or are moral (22%). Only a third believe the press protects democracy (35%).
• The study also points to ways journalists can rebuild trust. If stories are rewritten to broaden their moral appeal, they become more interesting to people in all groups—both those more trusting of media and those more skeptical. We took some basic news stories and wrote each of them two different ways. The revised versions edited the lead sentence and headline to emphasize different themes of the story that highlight the moral values of authority or loyalty (e.g., calling out leaders or ties to the local community). The revised versions also included an additional paragraph that emphasized a different moral angle of the story in addition to the frames included in the original. In all other ways, the two versions contained the same information. In some instances, the revised stories were more appealing to all types of people. For example, significantly more people considered a revised version of a story about election security to be balanced (62% versus 44%). More also considered the revised story trustworthy (78% versus 70%). And even people who already trust the press tended to like stories more when those stories were revised to broaden their appeal.
• To woo subscribers, the media will need to vary its messaging beyond traditional appeals about journalism being a watchdog. The survey also tested different messages asking respondents to financially support a local news organization. The findings suggest people’s moral leanings definitely influence what kind of messaging about journalism they find appealing. People who most emphasize care or fairness, for instance, were more motivated by a message that highlighted the outlet’s commitment to protecting the most vulnerable through their news coverage. People who emphasized authority and loyalty preferred a message about the outlet’s long-term service to the local community.

How we conducted the study

The study builds on moral foundations research led by social psychologist Jonathan Haidt, which tests how people respond to different moral principles. Research has found that some people put a higher moral value on caring about others’ well-being and ensuring fairness for all. Others also value giving deference to authority and loyalty. Typically, liberals relate most to the values of care and fairness. Care and fairness speak to conservatives as well, but conservatives place additional import on loyalty, authority, and purity.

Starting with this moral foundation framework, this study asked people for their reactions to a group of basic journalistic values[ref The five journalistic principles we tested build upon values identified in The Elements of Journalism, past surveys of journalists such as those conducted by The Project for Excellence in Journalism, and feedback from a diverse group of journalists and researchers who are listed in Appendix III.] to see the relationship between people’s moral foundations and their attitudes about core principles of journalism. Respondents were asked a series of questions to measure their scores on the typical moral foundations values as well as common journalistic values,[ref Respondents answered each item from 1-Strongly agree through 6-Strongly disagree. Half of the items are favored to one side of a value or the other and are reverse coded, so that all range from 1-low in value to 6-high in value. Each of the four items measuring a value are averaged, excluding missing items. The final average of items for each value is the final item score.] such as the importance of transparency in public institutions, monitoring powerful people, and giving voice to the less powerful in society. For each of the values, people were asked about whether they also worry about the antithesis of those values (for instance, that too much focus on problems might make them worse). They then read common news headlines and the opening paragraphs of stories that reflected these journalism values. We measured feelings and reactions to each story and people told us how interested they were in them.

In a follow up to the initial survey, we gave the same respondents additional news stories to read. There were two versions of each story. Version A was written with a common frame and version B included additional viewpoints aimed at people valuing authority and loyalty. Respondents were randomly assigned to receive different versions of each story. Following each story, respondents answered a series of questions about their assessments of the story and their engagement with the material.

For journalists who aspire to traditional journalistic values, such as intellectual independence and separating news from opinion, the findings suggest potentially unexplored ways to garner broader public support without sacrificing these values. By adding additional framing to stories, including additional perspectives or even just adjusting a headline, the same common news stories can appeal to conservatives as much as they do to liberals, gaining a broader audience. And in the cases we tested, liberals who already had favorable views of the press didn’t like the revised stories any less; in some cases, they liked them more.

The first nationwide survey was conducted with 2,727 American adults from October 22 to November 15, 2019. The survey featured 1,413 interviews from the AmeriSpeak® Panel, the probability-based panel of NORC at the University of Chicago, and 711 interviews from the Dynata nonprobability panel. The second survey was conducted from August 18 to 24, 2020, and featured interviews with 1,155 AmeriSpeak® panelists who completed the first survey.[ref The second survey was conducted in summer 2020 before the presidential campaign was in full swing. The partisan gap in trust toward the media widened during the Trump presidency, but Republicans have been significantly less trusting of the media than Democrats for the last few decades, regardless of the party in the White House.]

Continue reading: Chapter 2: What we set out to do

A more detailed methodology, along with a PDF of the full report and toplines of the study’s two surveys, are available here.