The Tampa Bay Times announced last week that Bill Adair, the newspaper’s Washington bureau chief and the founder and editor of the award-winning factchecking website PolitiFact, is resigning to take a position as the Knight Professor of Computational Journalism at Duke University’s Sanford School of Public Policy (where I went to graduate school). He will remain a contributing editor at PolitiFact.

As a political scientist who studies misinformation and the former editor of a factchecking website, I have known Adair for several years and followed his work closely since PolitiFact was founded. He and I discussed his decision and the state of the factchecking movement in an online exchange on Friday. What follows is a lightly edited transcript.

Brendan Nyhan: Bill, congratulations on your appointment. Could you tell us why you decided to take this position and what you hope to accomplish? Your predecessor at Duke, Sarah Cohen, helped develop Reporter’s Lab, a computational journalism initiative, before leaving for The New York Times. Do you envision pursuing a similar initiative centered around factchecking or will you have a broader portfolio?

Bill Adair: I plan to build on Sarah’s great work and use the Reporter’s Lab to experiment with new forms of journalism.

I think we need to try new approaches. We’re living in a digital age when the Web and mobile devices have given us a new canvas to create different story forms, but journalism is largely stuck in the 20th century, relying on old-media approaches like the inverted pyramid news story. We need to try some new ways of telling stories and conveying information that really harness the power of our digital age.

There are a few evangelists who are doing this, such as Reg Chua at Thomson Reuters. But too much of our journalism is still being told the old way.

At PolitiFact, we created two new story forms—the Truth-O-Meter fact-check and the campaign promise update for our Obameter. I want to explore creating other new forms that news organizations can use, particularly for coverage of government.

BN: That sounds very promising, especially given that the inverted pyramid structure often highlights misinformation rather than corrective information. Do you have any specific ideas that you want to explore about how to present factchecking or cover misleading claims more effectively? My research suggests, for instance, that graphical corrections are more effective than text.

BA: I’m looking more broadly than just factchecking because I think there’s a great need to try new ways of presenting information. But I remain very interested in what works and what doesn’t work in presenting factchecking information, and I’d love to explore ways that we can make factchecks more effective.

BN: Now seems like a good time to look back at what you’ve accomplished at PolitiFact, which has become the most influential factchecker in the business, including winning a Pulitzer Prize. What are you most proud of about your work there? And what you think the next steps are for the factchecking movement in general and specifically for PolitiFact?

BA: I’m really proud of the way we were able to expand to the states. We have 10 state PolitiFact sites in key places such as Ohio, Florida, Wisconsin, and Virginia, where reporters for our partner news organizations do Truth-O-Meter fact-checks.

Letting someone else use our prize-winning brand was a risk. But we applied lessons from the fast food industry (conduct lots of training, provide good manuals, and do quality control checks) —and it’s worked well. Last year, the factchecking by our partners at the Cleveland Plain Dealer was really important in the U.S. Senate race. (Editor’s Note: CJR wrote about the PolitiFact Ohio’s campaign coverage here and here.)

I think PolitiFact’s next step is to keep expanding. We want to find partners in the remaining states and see if we can expand internationally. We are partnering with a respected Australian journalist to launch PolitiFact Australia later this spring. If that goes well, we’ll consider other countries.

BN: What are the larger lessons that you’ve learned from this process about how best to train journalists in factchecking? I’m thinking of both media organizations that want to scale up their use of the approach internally as well as journalism training in academic institutions. In both cases, one obstacle seems to be shifting reporters’ mindset and approach away from traditional “he said, she said” reporting.

Brendan Nyhan is an assistant professor of government at Dartmouth College. He blogs at and tweets @BrendanNyhan.