Fact Checking and Embedded Links

From Green Policy
Jump to navigation Jump to search

Fact Checking

Correcting 'Falsehoods' and Lies, Disinformation & Misinformation

Algorithms, Big Data, AI / Artifical Intelligence ... Looking into Future of Media

Data & Society studies the social implications of data-centric technologies, automation, and AI

Through empirical research and active engagement, our work illuminates the values and decisions that drive these systems — and shows why they must be grounded in equity and human dignity.

Tips of the News Trade via Poynter, St. Petersburg, Florida

PolitiFact Project Goes International

Disinformation - Online - Dangerous

Tracking the 'Big Picture' and Reporting Disinformation & Misinformation Online

NewsGuard targets Disinformation / Misinformation: Check out the NewsGuard online services


The Professional World of Journalism Steps Up to Confront 'Bad Actors'

Bots, Conspiracy Theories, Dis/Mis Information, 'Troll Farms', Political Manipulation

Digital Rights Tag Cloud from GreenPolicy360.net.png

As embedded fact-checking grows, reporters need to show their sources

Increased transparency will allay the suspicions of a skeptical public.

By: Joel Luther and Bill Adair | Poynter.org

Five years ago, the trend began as a trickle.

A few reporters started adding a bit of extra information to their stories, declaring when a politician said something false. At the time, this was a bold move. Political stories were traditionally for news and the he-said/she-said of politics and policy; fact-checking was considered a separate genre. The Duke Reporters’ Lab, dubbed it “embedded fact-checking”. This fact checking continued to grow and by 2019, there were dozens of examples... this was progress, although nearly all were about Donald Trump.

Last year, the practice grew even more. As Trump and his supporters repeated false claims about the election, reporters at The Associated Press, the New York Times, the Washington Post and many other outlets used thousands of embedded fact checks in their news articles.

We are encouraged by this progress but concerned that journalists often do not reveal the source of their conclusions. As the trend grows, they need to be more transparent to show why they concluded a statement is false.

To track this new phenomenon, we counted the use of words and phrases such as “baseless,” “falsely claimed” and “incorrectly said,” which are typically used when reporters embed a fact check, in stories in The Associated Press, The New York Times and in transcripts from CNN. Beginning in the late 1970s, these phrases appeared fewer than 100 times every month.

Usage began to tick up in 2017, coinciding with the start of the Trump presidency. “Mr. Trump falsely claimed on Twitter that ‘millions of people’ had voted illegally,” wrote The New York Times. “Trump claimed falsely that the crowd for his swearing-in stretched down the National Mall to the Washington Monument and totaled more than 1 million people,” said The Washington Post.

The big surge came last year. The word “baseless” appeared 646 times in The Associated Press, 925 times in The New York Times and a whopping 1,004 times in transcripts from CNN.

We found the same spike last year with the phrase “falsely claimed.” It appeared 120 times in CNN transcripts, 124 times in The Associated Press, and 265 times in The New York Times.

The sharpest increase came after the Nov. 3rd U.S. election.

“Major Networks Cut Away From Trump’s Baseless Fraud Claims,” read a Nov. 5 New York Times article. “Trump and allies spread falsehoods to cast doubt on election,” wrote The Associated Press on Nov. 4. Again and again, false claims were debunked right in the news stories — not by trained fact-checkers, but by political reporters willing to call out a falsehood.

The reporters usually did not show their sources, however. When we examined a sample of those stories, we found that few provided evidence why the claims were false or baseless. Reporters were surely on solid ground — a host of fact-checkers, other journalists and courts had debunked claims about election fraud and other voting problems by Trump and his supporters. But the articles didn’t provide links or attribution.

That lack of sourcing is troubling. If reporters are going to make a bold claim that a politician’s claim is baseless, they need to show why with authoritative reporting the same way that fact-checkers do.

So while we are encouraged by the growth in embedded fact-checking, we don’t want it to allow shortcuts or sloppiness. We have a few suggestions:

- It’s not enough to just label something “baseless” or “false.” You’ve got to explain why, which can often be accomplished with one additional sentence. So after you say a party leader made a false claim about nonresidents voting illegally in a state, add a sentence that says, “Fact-checkers examined that claim and found no data to support it.”

- Transparency is key. Link to fact checks or other authoritative, independent sources to show reassure readers it is not the writer’s opinion.

As we wrote in 2019, embedded fact checks shouldn’t just be about Trump, even though he is likely to continue his falsehoods as an ex-president. While his penchant for misinformation helped launch the practice into the mainstream, journalists need to expand this trend and hold other politicians accountable, too.

Embedded fact-checking is a powerful new way to get more accurate information to more readers and viewers. But it needs to be done right to reassure a skeptical audience that it’s based on facts.


Joel Luther is an associate in research at the Duke Reporters’ Lab and manager of the ClaimReview Project.

Bill Adair is the Knight Professor for the Practice of Journalism and Public Policy at Duke University and the founder of PolitiFact. The fact-checking site is owned by the Tampa Bay Times, which is owned by Poynter.

St. Petersburg, Florida