Challenges from an “alternative facts” era

Politicians have lied throughout history, but today’s political climate presents particular challenges to fact-checkers — and that was the concern of journalists and others gathered at our summit earlier this year.

Throughout the 2016 election cycle and into the new White House administration, some politicians’ habit of making “demonstrably false claims” has kept fact-checkers exceptionally busy. (Andy Borowitz, a humor writer for The New Yorker, has joked that President Trump has created “10 million jobs for fact-checkers.”) But while fact-checkers have tried to hold those in power to account, the current administration has complained about fact-checkers’ work.

The Trump administration has publicly repudiated journalistic fact-checking and dismissed critical news coverage as “fake.” It is a deliberate strategy, media and communications experts say, and one that appears to resonate with some (mostly) conservative audiences who were already primed to mistrust the media during the 2016 campaign. The verbal attacks on reporters during the campaign, particularly at rallies where journalists were kept in “pens” and taunted by Trump supporters, went largely unaddressed by media leaders and continue today.

For fact-checkers, a key concern is how to cover the administration’s prevaricators and supporters of this “dishonest media” theme, without appearing to justify administration claims that the news media are “the opposition party” or “enemies of the people.”

Consider some carefully fact-checked claims from the first month of the new administration. There were administration claims about massive voter fraud in the November election that had no basis in fact; misinformation about the nation’s murder rate; and inflations of the inauguration crowd size. From the president’s official Twitter account came statements about “any negative polls” — that is, those that are not favorable to his immigration policies — being “fake news.” At a news conference, the president said his electoral vote total beat every president since Ronald Reagan, even though every subsequent president but one had higher totals. When questioned on NBC about the conflicts between administration claims about the inauguration size, a counselor to the president explained the White House was simply offering “alternative facts.” The instantly immortal phrase was widely critiqued.

Perversely, the power and reach of the president’s bully pulpit is demonstrated most amply when he says something that is demonstrably untrue. In a series of tweets on March 4, Trump alleged that President Obama had ordered a wiretapping of Trump’s office while Trump was still a candidate and a private citizen. Trump offered no evidence and his claim has since been denied by an Obama spokesman, Obama’s national intelligence director, members of Congress (including Republicans), and the current directors of the National Security Agency and the FBI. Nonetheless, a CBS News poll released March 29 found that 74 percent of Republicans surveyed believed it was “very likely” or “somewhat likely” that Trump’s office had been wiretapped during the campaign.

The incidents also contributed to a debate among fact-checkers that began well before election day: whether to call such comments “lies.” The New York Times called Trump’s comments about voter fraud a lie on its front page; NPR, by contrast, has stated that it does not want to ascribe lies to Trump without knowing for certain his intent was to deceive.

Fact-checkers at the conference made it clear they want their role to be understood as sifting through truth as finely as they can, not seeking to undermine any particular politician or party. Still, they’re aware that’s not a message that resonates in this politically charged post-election climate.

Neil Newhouse, a Republican pollster and a summit participant, told the group, “I don’t think you’ve got a fact-checking problem as much as you have a problem with the idea of the news business being impartial observers. Americans simply don’t trust the news media in general.”

Newhouse cited numerous polls that demonstrate a declining faith in news outlets. A Gallup poll released last September, for example, found that Americans’ trust in mass media “to report the news fully, accurately and fairly” had plunged to a new low. Only 32 percent of those surveyed said they had a “great deal” or “fair amount” of trust in the media, which was a drop of 8 percentage points from 2015. The numbers were even lower among Republicans, with only 14 percent saying they had faith in the media. In an American Press Institute survey last year, only 6 percent of people said they have a great deal of confidence in the press.

In October, Quinnipiac University released a poll that found 55 percent of likely voters believed the media were biased against Trump, including nearly 90 percent of Republicans.

“Your industry is losing the public opinion battle,” Newhouse told the group. “Americans believe that reporters are biased and try to help the candidates that they support win.”

This poses a challenge for democracy, Newhouse added: If Americans can’t agree on what facts are, how can they agree about how to respond?

Lack of trust in the media unquestionably undermines the credibility of fact-checkers, noted Angie Drobnic Holan, the editor of PolitiFact. “While we are clear in our minds about the differences between reporting and opinion, what’s clear to us is not at all clear to readers,” she said. “Fact checks come out of the reportorial tradition, but people see us as opinion [writers], because we’re weighing in.”

Readers don’t always distinguish between analysis — a considered judgment about how, for instance, a policy is likely to play out — and opinion. The push in recent times to offer more analysis as a means of distinguishing their coverage may have undermined the credibility of media outlets, suggested Greg Linch, a data developer at McClatchy.

Jessica Arp of WISC-TV in Madison, Wis., one of the pioneers in fact-checking at the local level, said reporters must work harder to engage their audiences in the story development process. That effort should help readers and viewers understand that reporters are not trying to promote their own agendas; rather, they are attempting to help people to sort truth from fiction themselves. “The more we can make fact-checking about trying to help you, maybe the more trust people will have,” Arp said. More examples of reader engagement are discussed below.

The context in which fact-checking often appears on digital platforms — draped within news coverage and opinion and even advertising — affects its credibility.

“One thing that’s important to consider is the role that shady sponsored content plays in diminishing trust in fact-checking and the press,” said Joyce Garczynski, a communications librarian at Towson University. “When you have fact-checking side-by-side with ads for less-than-reputable products and services, and users can’t tell the difference, it can only hurt news outlets’ overall credibility.”

Addressing partisan and cognitive bias

The potential pitfalls of journalists attempting to play referee have been inherent to fact-checking since it first emerged as a stand-alone pursuit more than a quarter-century ago, Tom Rosenstiel, executive director of the American Press Institute, said at the summit.

“We were assuming that people knew our role and that calling out falsehoods is something they’d appreciate,” he said. Instead, fact-checkers are sometimes accused of tilting their judgments for partisan reasons. “We’re playing a more pointed role, so we’re at the point of that spear” of media distrust, Rosenstiel said.

Political fact-checking encourages reporters to question politicians’ accounts of events, rather than simply recording what they said. Typically, a fact check is limited to one claim by one person; and if only one person’s veracity called into question, that can lead partisans to wonder whether the media’s deck is stacked against its candidates.

Accusations of bias against the news media are nothing new, but in an era of starkly polarized divisions there’s now an increasing awareness of how non-journalists — that is to say, citizens in general — also bring bias to the table.

Shortly after the January inauguration, two researchers conducted an experiment using photos from the Trump inauguration and from the 2009 Obama inauguration. Taken from the same vantage point, the photos showed that the crowd for Trump was less dense than the crowd for Obama.

When shown the photos, Trump voters were significantly more likely than Hillary Clinton voters or non-voters to give the wrong answer about which photo had more people. “Even when the photographic evidence was directly in front of them and the question was straightforward, one in seven Trump supporters gave the clearly false answer,” the researchers found.

This finding was in keeping with other social science research that has shown people are willing to skew their own beliefs in order to align with their party, such as believing the economy is doing better when their party holds the White House.

At the summit, Dan Kahan, a professor of psychology and law at Yale University, outlined studies he helped conduct that show a clear partisan divide when it comes to answers about questions on subjects such as gun control, evolution and fracking.

For instance, there’s a clear partisan divide when people are asked whether it’s true or false that “human-caused global warming will result in flooding of many coastal regions,” with conservatives more likely to say that is false. But if that statement is prefaced by the phrase “according to climate scientists,” there’s no partisan divide at all, said Kahan.

When tested on their scientific knowledge of climate science, liberals and conservatives are likely to offer the same (correct) answer, Kahan said. On the other hand, if the question is limited to one’s personal beliefs about climate change, respondents tend to follow their partisan inclinations. Conservatives know that the Republican Party, for the most part, is skeptical about climate change, and they want to represent beliefs that comport with the party’s position.

“People are trying to be who they are as members of a group,” Kahan said. “They know the answer. They’ve been told what scientists believe. There are ways in which they can use that information, but otherwise that’s overridden by their desire to be part of a community and have the collective identity.”

It’s not ignorance. Kahan’s experiments show that the more numerate or educated people are, the more likely they are to give wrong answers if they feel partisan fealty demands it. More educated people are more likely to find ways of rationalizing their incorrect answers.

“Our work shows that people not only use emotional evidence to fit their group’s position, they use slow, deliberative reasoning to do the same,” Kahan said. “They’re never missing it when the data supports their view. If [the data] doesn’t, they’re turning on the cognitive afterburners.”

This complicates the work of fact-checkers when their work casts doubt on positions held by party leaders and the public who follow them. What’s more, the sense of subscribing to a partisan identity can extend to people’s choices of news outlets.

“Several people noted that when fact-checking is associated with the mainstream press it is immediately suspect among people who depend largely on Fox and the Internet for their news,” said Tom Stites, founder and president of the Banyan Project, a nonprofit that helps support community news cooperatives. “An important tactic going forward will be finding ways to bring fact-checking to people in neutral packages.”

Even when people do change their minds about misinformation thanks to a fact check, they may dismiss the results, said researcher D.J. Flynn of Dartmouth College. They may believe corrective information about a particular economic fact but may subsequently downgrade the amount of importance they attach to that particular issue, because their new (correct) stance does not help their party’s cause.

Therefore, Flynn said, “If you correct a false claim, they might double down on their existing opinions.” In other words, persuasion is a moving target. Presenting people with evidence and discrete facts may not be enough to change how they think about an issue more broadly.

“Our fact-checking really only speaks to half of our brains,” said Jeff Sonderman, deputy director of the American Press Institute. “It treats people as rational actors: If we give people the right information, they’ll make the right decision. But we know we’re not rational people. Half of the brain is emotional and driven by other things.”

And fact-checkers, no matter how well-intended or neutral they consider themselves to be, must be willing to check themselves for bias, suggested Brad Scriber, deputy research director for National Geographic magazine. “How can fact-checkers be assured that we are keeping our own biases in check? How do various outlets calibrate to be sure that your staff is objective and remain so?”

A growing challenge: Misinformation and social media

Emerging from the summit discussions were two clear challenges: One, getting carefully checked information to the audiences who would benefit from it, which we will discuss later in this report. The other: Figuring out how to respond to the torrent of misinformation that is being spread rapidly via social media and other methods.

Fake news — stories that are invented either partly or fully, in order to push a narrative — became a phenomenon in 2016, with its practitioners enjoying substantial success in website traffic.

As one conference attendee noted, fake news is the “like” economy working at peak efficiency. Fake news is designed to have the hallmarks of content that spreads frictionlessly through social media: headlines that stick and generate lots of clicks.

It also works in terms of reinforcing group identity, encouraging people to embrace stories that reinforce their existing worldview. Although fake news stories that supported Trump received most of the attention in 2016, there are already indications that those on the left are sharing more fake news stories during the Trump presidency than they did during the campaign.

“People who think they’ve been pushed out of the political world as it is right now are going to be susceptible to misinformation — they’re going to focus on whatever makes them feel better,” Brooke Binkowski, managing editor of Snopes.com, recently told the Christian Science Monitor.

Fact-checkers must decide how much of their time and attention they can afford to devote to correcting information that is not only misleading but clearly false. They are aware that much of the misinformation during a campaign season flies in under the radar; viral emails, for instance, still play a role. And they may have to step up their games, as technology becomes more sophisticated. In the near future, for instance, fakers will be able to convincingly alter audio and even video in ways that will make it hard to distinguish fact from pure fiction.