December 12, 2007
Want to see a journalist wince? Publish a sentence that begins this way: “In yesterday’s edition, it was inaccurately reported…”

Want to make a journalist squirm? Post these two words above an online story: “Correction appended.”


Corrections are journalism’s equivalent of Puritan-era stocks, clamping a wrongdoer’s feet, hands or head inside heavy wood frames, on humiliating display in the town square.

As painful as corrections are to journalists, the screw-ups they reflect do damage on a far greater scale to the news organizations they work for.

“Each misspelled word, bad apostrophe, garbled grammatical construction, weird cutline and mislabeled map erodes public confidence in a newspaper’s ability to get anything right,” a 1998 study commissioned by the American Society of Newspaper Editors concluded. “Even seemingly small errors feed public skepticism about a newspaper’s credibility.”

Enter Craig Silverman, a Canadian freelance journalist and author who decided that the best way to make the case for accuracy was to expose mistakes and efforts to — or not to — correct them. In 2004, Silverman created “Regret the Error,” a Web site focused “on media retractions, apologies, clarifications and trends regarding accuracy and honesty in the press.”

On his site, and in a new book, “Regret the Error: How Media Mistakes Pollute the Press and Imperil Free Speech,” Silverman presents an often hilarious collection of media mistakes and the corrections they engender.

My favorite from his 2007 list of Media Errors and Corrections is this one from The Sentinel-Review (Woodstock, Ontario):

In an article in Monday’s newspaper, there may have been a misperception about why a Woodstock man is going to Afghanistan on a voluntary mission. Kevin DeClark is going to Afghanistan to gain life experience to become a police officer when he returns, not to shoot guns and blow things up.
The Sentinel-Review apologizes for any embarrassment this may have caused.
He highlights other really funny mistakes like this one from the Dallas Morning News — “An Oct. 19 article on songwriter John Bucchino incorrectly stated that he doesn’t read. The sentence should have said he doesn’t read music.”

Yucks aside, Silverman is serious about mistakes. He sometimes takes a scholarly approach as well, tracing mistakes from “the 2 of December” that appeared in the oldest surviving English-language paper to the rise of external fact checkers. These include the bloggers that went after CBS and Dan Rather about the “60 Minutes story” on President Bush’s National Guard record that ended with Rather�s ouster. 

I recently interviewed Craig Silverman by email. Our exchange follows.

Scanlan: What was the genesis of “Regret the Error”?

Silverman: It was a combination of wanting to start a media blog and noticing that corrections in particular were a fascinating and un-mined part of journalism. I came up with the idea of tracking errors and corrections early in 2004 and was inspired to move ahead after I saw this July correction from the Lexington Herald-Leader: “It has come to the editor’s attention that the Herald-Leader neglected to cover the civil rights movement. We regret the omission.” That told me there would be a lot to say about corrections and errors. So I launched in October 2004.

Scanlan: How much time do you devote to the site? Why do you do it?

Silverman: “Accuracy is a key that can unlock innovative, investigative journalism.”Silverman: I spend between one and three hours per day. I started the site because I thought it would be a worthwhile addition to the media landscape, and a good way to raise my professional profile. I have to admit that there’s now a larger mission: to help journalists see that accuracy is an enabler of great journalism. Aside from being an essential nuts and bolts aspect of how we do our job, accuracy is a key that can unlock innovative, investigative journalism. It can help us forge a stronger bond with the public and with our sources. When we dedicate ourselves to preventing and correcting errors at the highest possible level, we open up the door to doing great journalism. If I can help spread the message about what accuracy can do, and why it is more important today than ever before, then I’ve done something good.

Scanlan: What kind of traffic does your site get?

Silverman: It’s usually about 1,000 people per day, but it spikes into the tens of thousands when I do my year-end roundup.

Scanlan: You don’t have any advertising on your site. Why is that?

Silverman: I made a conscious decision not to seek out or accept ads. The site is supposed to be a service to both the public and the press. I also don’t want to appear to be making money off of the mistakes of others. The site is more of a mission than a business, and not having ads is a way of communicating that.

Scanlan: How do items come to your attention?

Silverman: The majority are discovered by me via my database and Web searches, and by checking the online corrections pages of different organizations on a daily basis. I read probably about 150 to 200 corrections a day. But readers also play a role. Many people send me corrections or errors, and I’m always grateful for that. A lot of working journalists send in their organization’s corrections — or those of a competitor.

Scanlan: How significant is the problem of inaccuracies?

Silverman: There has been a lot of scholarly research aimed at discovering the level of error in U.S. newspapers. It began in 1936 and has continued since then, with people like Scott Maier and Philip Meyer doing a lot of the heavy lifting. Overall, the research suggests that between 40 and 60 percent of newspaper news stories have some type of error, be it factual or something of a more subjective nature. So that’s the frequency. But here’s the other part of the equation: Research from Maier published this year found that only 2 percent of factual errors were corrected. So we have a relatively high error rate, and that is compounded by an anemic correction rate. Errors are not being prevented, and they are not being corrected.

In a media environment where stories are often published in a paper, placed online and then loaded into various databases, the issue of uncorrected errors becomes even more urgent. The errors of today become the errors of tomorrow when they are accessed online or from a database at a later date. As much as we are creating the historical record, we’re also polluting it with errors. Errors can then be blogged, cited in research, used in press releases … they go farther, faster than ever before. In many cases, they exist forever. So we have a responsibility to do everything we can to prevent and correct them. It’s part of our job as journalists. Stories don’t end once they’re published; we are responsible for correcting and updating them.

The other piece, of course, is the effect that errors have on the public’s perception of the press. Put simply, errors erode credibility. The public notices mistakes, and they notice when we don’t correct them. A survey of newspaper readers by the ASNE found that over 60 percent of readers said they felt better when they saw corrections. They don’t expect perfection; they expect us to work hard to prevent errors and to correct any that occur. When we don’t do that, they punish us by tuning out.

Scanlan: Regret the Error has a global reach on errors. How did that come about?

Silverman: I initially planned to only focus on North America. But then I discovered the Guardian‘s wonderful corrections column and the astounding apologies of The Sun tabloid and I realized I should do my best to cover all English-language media. Readers are very helpful in this regard.

Scanlan: What are the most common errors?

Silverman: “Between 40 and 60 percent of newspaper news stories have some type of error … only 2 percent … were corrected.”Silverman: Common errors include misspelling names and titles, typos, incorrect calculations or incorrect numbers, and misquotes and misidentifications. The vast majority of factual errors are accidental. We know or can easily obtain the correct information, but something goes awry at some point in the reporting, writing, editing and production process.

Scanlan: What are the classes of errors you’ve identified?

Silverman: Errors fall into two main categories, as defined by Donald Norman in “The Design of Everyday Things” — slips and mistakes. Slips are what I previously described — errors that occur in spite of us knowing or having access to the correct information. Mistakes are the result of a conscious decision. The Chicago Daily Tribune declaring Dewey the winner in the 1948 presidential election (“Dewey Defeats Truman”) was a mistake. They made the wrong call.

Here’s the good news: Slips are highly preventable.

Scanlan: How can news organizations curb the problem?

Silverman: I wish I could say there was a magical accuracy machine that could be installed in newsrooms, but it’s not that simple.

The first thing to do is create an organizational culture that values error prevention and accepts that corrections are an important part of journalism. Get rid of the stigma of error that causes people to want to hide their mistakes and not learn from them. Then take that culture and make it real through training, effective technology and good processes.

It all starts with an attitude — a passion for accuracy and a recognition of the importance of corrections. I don’t doubt that within most newsrooms there are people who have devised their own great little way of checking and verifying information. Find a way to capture and share that information, and you’re already on your way.

Scanlan: What are ways that individual reporters can avoid errors from creeping into their stories?

Silverman: I think one of the easiest things to do is to use a checklist. I know many people roll their eyes when I say that, but remember that airplane pilots use checklists. They can work. Create a list of the most common facts in any story (names, titles, numbers, dates, etc.) and then also add a few items that you personally have had trouble with. Do you always misspell a certain word? Make yourself go through that checklist before finishing every story. I suggest getting it laminated so you can use a dry erase marker to check off each box as you go down the list. And leave space for a couple of write-in items that are specific to a particular story.

Also, remember that memory is unreliable. If it’s not on paper or in a definitive source, take the time to make a call or send an e-mail. Just a few extra minutes of time on the phone or e-mail can be the difference between a clean story and one with errors.

When in doubt, make the call. And when you think you’re done, run the checklist.

Scanlan: You’ve done some training in newsrooms. Can you describe how you try to help them with the error problem?

Silverman: I’m only just beginning to work on training, so I think I’m learning as much as I’m teaching, which is not exactly a bad thing. Right now I’m trying to go into newsrooms to present about error and correction rates, and explain the larger issues around mistakes: culture, process, technology, human error. I’m hoping to shift organizations away from the culture of shame and help them embrace the reality of error and the ethic of meaningful correction. I also try to inspire them to innovate when it comes to prevention and corrections. I show some of the great things different organizations are doing. So it’s more educational at this point. I hope to add more elements of hands-on training.

Scanlan: If you could change anything about news organizations’ response to errors, what would it be?

Silverman: “The vast majority of factual errors are accidental.”Silverman: There’s the larger issue of the stigma of error — that we think only bad reporters and editors make them. Then there’s the lack of data capture and knowledge sharing that ensures we aren’t learning from our mistakes. Finally, we need to be innovative about correcting errors.

Scanlan: If you were to design the perfect correction for today’s media world, what would it look like? How do we ensure our readers/viewers/listeners get the correct information?

Silverman: I probably sound crazy, but that strikes me as a wonderful, exciting challenge. Imagine the message we could send to the public if we began unveiling vastly more effective prevention and correction programs.

Scanlan: How well do newspapers deal with corrections? How about TV and radio outlets? Web sites?

Silverman: There are some bright spots, but for the most part I have to say that corrections are ignored and taken for granted. They are a core part of our contract with the public — that we will be accountable by publicly correcting our errors — and yet there is, for the most part, a lack of commitment to the meaningful act of correction.

That means creating a culture where reporters and editors are encouraged to own up to mistakes and get corrections issued. It means making it easy for the public to report errors, and thanking them for doing so. And it means presenting corrections in clear language and with a placement that ensures people will actually see them. A correction that is vague or hidden is a hollow act that only serves to absolve a news organization rather than actually correct the record and inform the public.

Newspapers are the most faithful media when it comes to corrections. But the effectiveness varies from one publication to the next. Some newspapers are serious about placing corrections online, while others don’t bother. Some have a toll-free number for the public to report errors. Others make readers jump through hoops to find the right person.

TV news organizations rarely issue on-air corrections. Some place corrections to on-air mistakes on a Web site, but what about all the viewers who don’t visit the Web site, let alone search out the corrections page? Same with Radio.

Online is the best environment for corrections that we’ve ever had, and there are some great innovations taking place. For example, not only does Slate place a correction within the article itself, they put an asterisk at the end of the paragraph in which the error occurred. Readers can click on the asterisk and a hyperlink takes them directly to the correction at the end of the article. Now readers know exactly where the error occurred. It offers context, and it’s great. They also have an RSS feed for corrections, an online corrections page, and Slate puts all corrections on the homepage at the end of each week.

Here’s the major shift that needs to happen with corrections in the online world: Rather than making readers search for them, we need to push corrections out to readers. Offer an RSS feed. Allow readers to enter their e-mail address on an article page so they can be alerted if that article is corrected. Add a “Corrected” tab to the homepage boxes that list the “Most Read” and “Most Blogged” articles. This would help people see if a story they read yesterday has been corrected today. Can we discover a way to automatically alert someone who has blogged a story that a correction has been issued? Can we create a way for members of Web sites to earn some form of reputational benefit from reporting a verified factual error?

I’m excited about what we can do with corrections. There are so many possibilities.

Scanlan: What have you discovered about how and why errors get published?

Silverman: There are two essential things journalists should know about errors:

1. Everyone, regardless of their level of training or experience (or lack thereof), will make them. Here’s what Ariel Hart, a fact checker at the Columbia Journalism Review, wrote: “Pound for pound, the most mistake-packed article I have ever checked was written by a Pulitzer Prize winner.” We are all fallible. Accept this and understand that:

2. Errors have a provenance. They should not be explained away with vague excuses like “I was sloppy” or “I messed up.” They occur because of a set of circumstances, and the same set of circumstances can produce an error again and again. We have to take a larger view of error and examine how our people, technologies, processes and sources play a part in causing them. Spell checkers cause errors. The way our brains process language causes errors. Handing a story from one editor to another causes errors.

If we take the time to examine exactly why errors occur, then we can begin to prevent them. In my book, I tell the story of how the U.S. military in WWII noticed an increase in incidents of “pilot error” that were causing crashes. But instead of simply accepting that some pilots were better than others, they focused on discovering the true root causes. After interviewing pilots and examining how they worked in and out of the cockpit, the military discovered that the way some controls were designed and placed inside the cockpit actually forced pilots to make mistakes. And so they changed the layout and design of the cockpit. The result? A reduction in so-called pilot error.

We have much to learn about the true root causes of press errors. A shift in how we perceive and examine them could reveal a lot, and put us on the right path.

Scanlan: What should a news organization do when it discovers an error?

Silverman: “I’m hoping to shift organizations away from the culture of shame and help them embrace the reality of error and the ethic of meaningful correction.”Silverman: Correct it in an effective manner (see above) and store the error in a database that captures where it occurred, how it occurred, and who discovered it. Then use that data as the basis for a prevention program. And if a reader or other outside party brought that error to the organization’s attention, they should be thanked for their time and effort. I also support the idea of contacting the “victim” of an error (be it a person or organization) to express regret and draw their attention to the correction.

Scanlan: Your site led to publication of a book. How did that come about?

Silverman: I didn’t foresee the site inspiring a book, but it became logical and perhaps even necessary. I had collected so many errors and corrections that I began to connect the dots and see patterns. Once I started looking into the history of media errors, accuracy research, fact checking and other topics, there was just so much wonderful material that a book became necessary. So now it exists and I hope people buy it and enjoy it. I really enjoyed writing it.

Scanlan: What’s your favorite correction?

Silverman: It’s really hard to pick. But I will offer up this one, which was published in the New York Daily News this past year:

A headline in Monday’s Daily News, “He regrets his role in ‘postal’ vid,” implied that Richard Marino, the subject of a YouTube video, was sorry for an incident in December at a Brooklyn post office. Marino, in fact, is not sorry. The News regrets the error.
It’s correcting and regretting the fact that someone doesn’t regret something! I love it.

Scanlan: And your favorite apology?

Silverman: This one from The Sun (UK) is tough to top:

On 13 February we published an article headed “Who bum it?” reporting that two Premiership footballers and a music industry figure had a “gay romp” in which a mobile phone was used as a “gay sex toy.” On 16 February we published a picture of Mr. Cole and his fiancee headed ‘Ashley’s got a good taste in rings.’

Some readers have understood that Mr. Cole was one of the two Premiership players involved in the gay sex and that Choice FM DJ, Masterstepz, was the music industry figure.

We are happy to make clear that Mr. Cole and Masterstepz were not involved in any such activities.
We apologise to them for any distress caused and we are paying them each a sum by way of damages. The Sun wishes Ashley all the best for next Saturday’s World Cup quarter-final.
Scanlan: How are you handling accuracy and corrections for the book?

Silverman: Great question! The book itself includes two accuracy innovations. The first is a “Statement of Accuracy” that follows the introduction. I describe the processes used in writing and editing the book, and I also state very clearly that I expect to have made errors. Because of this, I ask readers to set the book down and go to the book’s Web site and click on the “Corrections” page. That way, they can immediately see the latest corrections to the book. I already have some up there.

While on the site, readers can sign up to have new corrections automatically e-mailed to them, and they can also subscribe to my book corrections RSS feed. I don’t expect readers to check the corrections page on a regular basis, and so I’ve enabled them to receive corrections via either RSS or e-mail. I’m pushing my corrections out to readers.

The second innovation within the book is that there is an “Error Report Form” at the back. I also have the same form online. I ask readers to use either form to report any errors. I believe I’m the first author to include a Statement of Accuracy and an Error Report Form. I’m probably also the first to ask readers to put a book down and go read corrections before they get into the book.

I believe that all of these elements will, in fact, enhance readers’ enjoyment of the book, and that it helps ensure I will discover and correct a high number of errors. (I also offer credit to those who submit verified errors.)

It was essential for me to be frank about the reality of errors, and to offer a way for readers to easily read and submit corrections. If anyone has any suggestions for how I can improve this structure, I’d love to hear them. I hope other authors take what I have and build upon it.

Support high-integrity, independent journalism that serves democracy. Make a gift to Poynter today. The Poynter Institute is a nonpartisan, nonprofit organization, and your gift helps us make good journalism better.
Christopher “Chip” Scanlan (@chipscanlan) is a writer and writing coach who formerly directed the writing programs and the National Writer’s Workshops at Poynter where he…
Chip Scanlan

More News

Back to News