In September, Nancy Barnes—the outgoing executive editor of the Houston Chronicle, who is headed to NPR—informed readers of serious concerns regarding the work of a veteran statehouse reporter, Mike Ward, who arrived at the paper in 2014 from the Austin American-Statesman. Ward was flagged by a colleague over concerns about whether the sources quoted in his stories were real people. Barnes ordered a review. The researchers on the case couldn’t find a number of people cited in Ward’s recent stories; when Barnes asked Ward for his notes, he said they’d been destroyed. Barnes escalated the investigation, hiring David Wood, who won a Pulitzer Prize in 2012 as a correspondent for HuffPost, to conduct an independent search. The results of that examination, which lasted two-months, were published by the Chronicle on November 8. They were not pretty.
Wood examined 744 stories written by Ward between January 2014 and August 2018. A team of three, Wood wrote in his report, chose 275 people quoted in those stories and made aggressive attempts to find them. Forty-four percent of them—122 people—could not be found. Another 50 people were entirely unreachable; Ward ignored Wood’s repeated attempts to contact him for guidance and comment.
The inability to find Ward’s sources was troubling, given that, as Wood points out, “in the age of online records, including property ownership and court filings, almost everyone can be found quickly.” For journalists, the Chronicle’s attempt at transparency may have come as a relief; here was a bad situation reaching resolution, out in the open.
But for a news organization, it’s still a nightmare scenario, especially in today’s environment, where trust in the media is chronically low and members of the press are regularly rebuked as “enemies” by the president and his supporters. Just last week, Jim Acosta, a White House reporter for CNN, had his press credentials revoked after an exchange with President Trump, live on television.
Barnes, who left her post at the Chronicle on Friday and will replace Michael Oreskes as the senior vice president of news at NPR (Oreskes was forced out after allegations of sexual harassment were brought against him), says that the widespread fabrication apparent in Ward’s articles was unprecedented, in her experience. “I’ve been an editor a long time and I have never seen anything like this, period,” Barnes tells CJR. “On occasion you might have an unidentified quote that you flag [in the editing process], but never anything where there are so many people we cannot track down.”
When concerns about Ward first surfaced, Barnes says, she turned to others in the industry for help figuring out what to do. “I talked to editors who had handled similar circumstances to determine what best practices were,” she recalls. “I decided we needed to do our own investigation. Some do one with their own staff, but I didn’t want to do that. We chose to hire an external investigator to do the work for us. In a situation like this it’s important to be transparent, so we investigated it the way we would anything else.”
Wood ran the names of Texans quoted in Ward’s stories through LexisNexis, a database of publicly available contact information familiar to most journalists, as well as other search engines that scanned property records, public voter registration rolls, hunting and fishing licenses, phone numbers, and social media accounts. He called law enforcement and elected officials; he visited businesses where quoted sources, if real, were likely to be found. In effect, he conducted a detailed, magazine-style fact check—something that lay readers may not realize doesn’t usually happen at daily newspapers.
The Chronicle announced in Wood’s piece that it would retract eight stories in print and online; in Wood’s story, he refers to several of them, sometimes identifying them by their headlines. In a separate, updated editor’s note, Barnes lists those stories in full. In an additional 64 stories, Barnes says, questionable quotes will be removed, and a note informing the reader of the changes will be attached, linking to Wood’s investigation for further explanation.
Both Wood’s investigation and Barnes’s editor’s note discuss growing doubts about Ward’s work, but neither explicitly say that he made up sources—a detail that, for readers, might elicit confusion. But Barnes says that was a deliberate choice. “You can’t prove someone is not out there,” she tells CJR. “We laid out the facts as we found them and phrased it very carefully for that reason, and let people come to their own conclusions.”
As local newsroom staffs and budgets have been diminished, workloads have increased and the importance of earned trust between reporter and editor has only grown. Like most daily newspapers, the Chronicle has no dedicated fact-checking department (though editors often double as checkers); Barnes says the total newsroom staff at the Chronicle is about 220 people, including approximately 40 line editors and copy editors. Line editors are generally responsible for a minimum of four reporters, though they can oversee as many as six; department heads oversee line editors. “The staff has shrunk a little bit, but not significantly,” she says. “Fifteen positions over five years would be my best guess. I do not believe that was a contributing factor in this case.”
Still, as part of Wood’s investigation, the newspaper performed what Barnes called a “spot check” of nine stories from three of its other reporters, as well as a reporter working at an out-of-state newspaper. All but one of the sources appearing in those articles were easily found. These kinds of safety checks had not been done in the past, Barnes says, but they might help prevent something this egregious from happening again. “Having learned from that, I might suggest standard spot checks for reporters just as a precaution,” she tells CJR—though she could not speak to any coming policy changes at the newspaper, citing her recent departure, and didn’t mention what might become policy at NPR. “I think having the researchers take an hour once a week and go through a random number of stories and do spot checks to see if they can find [the sources] might be a good precaution in the future.”