New study shows how newspaper inaccuracies transcend journalism cultures, national borders

In 1936, journalism professor and former magazine editor Mitchell Charnley published the first accuracy audit of American newspapers.

Charnley selected a sample of news articles, identified the sources quoted in them and mailed those sources a survey. The survey asked objective questions, such as whether their name was misspelled. It also asked subjective questions, such as whether their comments were taken out of context. The idea was that these sources had specific knowledge about the article, and could therefore provide valuable feedback on its accuracy.

More than 70 years since “Preliminary Notes on a Study of Newspaper Accuracy” came out, researchers have refined and expanded Charnley’s so-called accuracy check study.

The largest accuracy study of U.S. papers was published in 2007 and found one of the highest error rates on record — just over 59 percent of articles contained some type of error, according to sources. Charnley’s first study found a rate of roughly 50 percent. For a detailed and insightful look at accuracy research in the U.S., I recommend this post from the AP’s Jonathan Stray.

This form of study is the standard for measuring media accuracy, yet it is by and large an American phenomenon. Many countries have never done them, and studies of international media outlets are few and far between.

Italy and Switzerland were among those that hadn’t until Stephan Russ-Mohl, director of the European Journalism Observatory, initiated new research. He teamed up with lead researcher Colin Porlezza, and they recruited University of Oregon journalism professor Scott Maier — the leading newspaper accuracy researcher working today. He helped advise the study as a visiting research fellow.

The result is “News Accuracy in Switzerland and Italy,” a paper recently published in the journal “Journalism Practice.” The authors conducted accuracy surveys on five Swiss and five Italian newspapers, and surveyed respondents to evaluate how errors affect sources’ perceptions of credibility. They also compared the data they found to results from a previous Maier study of U.S newspapers.

The headline is the findings in Switzerland and the exploratory data in Italy are in many ways consistent with the level of accuracy, or inaccuracy, in the U.S. The response rate to their survey from sources in Italy, however, was too low for them to be able to describe their data for that country as anything other than “exploratory.”

The Swiss data are therefore far more important. Overall, they write, “The results provide evidence that newspaper inaccuracy — and its corrosive effect on media credibility — transcends national borders and journalism cultures.”

Below are highlights from their work, in addition to a Q&A I conducted with them to find out more.

From the Swiss findings:

  • In Switzerland, “News sources found factual inaccuracy — one or more objective ‘hard’ errors such as incorrect names or dates — in 60 percent of Swiss newspaper stories they reviewed, compared to 48 percent of US newspapers examined.”
  • In the U.S. and Switzerland, the two most common factual errors were the same: misquotations and inaccurate headlines. The least common factual errors were incorrect ages and incorrect addresses.
  • Swiss news sources deemed misspelled names the most egregious error. They also said that getting a location wrong was a severe error.
  • “By every measure, inaccuracy had a corrosive effect on media credibility,” the study’s authors write. “News sources, while surprisingly tolerant of errors, maintain high expectations that the news media will get the story right. Across nations, news accuracy really matters.”
  • Respondents were also asked to share why they think errors occur. Most U.S. and Swiss sources blamed a lack of understanding on the part of the journalist about the subject of their story. Also, “Swiss sources attributed inaccuracies to deadline pressure in greater proportion of errant stories than US sources.”

Highlights from findings in Italy:

  • Remembering that this is merely exploratory data, the study found sources “reported factual errors in 51.9 percent of Italian newspaper stories.”
  • The two most common factual errors reported were misquotes and inaccurate headlines. These two categories were prevalent in all three countries.
  • Also: “A significant relationship between error and credibility was found among the Italian newspapers studied, though the correlation was generally weaker than found with either the Swiss or the U.S. press.”

Below is the email Q&A I conducted with Maier and Porlezza. Among other things, I ask them about what they deem to be a “prisoner’s dilemma” for journalists when it comes to correcting errors. As they put it, journalists who are diligent about publicly correcting errors may seem less accurate than their counterparts who fail to admit to their mistakes and issue corrections. This underscores the fact that errors are bad, but corrections are good. It’s a distinction that often gets lost or overlooked.

Craig Silverman: Would you say this is the main takeaway of the paper? “Newspaper inaccuracy transcends national borders and journalism cultures”? If so, what does that mean for journalists/news organizations?

Scott Maier: It is indeed a key finding. Though this study obviously cannot speak for all media, it does put newspapers everywhere on notice: factual errors are likely common whenever breaking news is reported. That is not to say that journalists can shrug off inaccuracy as inevitable, but rather that the news media need to redouble efforts to get the story right.

There was an interesting element to the Swiss data. I think I’m going to call it the Swiss Conundrum. On the one hand, sources maintain a high level of trust in the press, yet they also proved pretty adept at identifying all manner of factual errors and rating them as statistically significant. What’s going on there?

Colin Porlezza: The Swiss respondents rated the errors with an average of 2.5, which means they rated them quite severe. However, the Swiss rating is less severe compared to the sources in Italy (2.7) or in the United States (2.8).

Yet, the willingness of the sources to act again as sources is significantly higher in Switzerland (56%) than in Italy (38%) or the U.S. (36%).

Taken together, both findings support the interpretation that the Swiss sources identify more minor errors (given the overall number of errors identified) than the Italians. On the other hand, they are aware of the “insignificance” of certain aspects of journalistic misconduct.

Thus, this seems to affect less the willingness to provide information as well as the credibility of the newspapers. This is confirmed by the high trust Swiss sources have in newspapers: On a 7-point scale from 1 (not credible) to 7 (very credible), respondents rated Swiss newspapers, despite the highest error rate, as relatively credible (5.5), while the U.S. sources rated the credibility of their newspapers 5.1, while Italians 5.2.

The high credibility attributed to newspapers, as well as the high willingness to respond within Swiss sources, can also be traced back to the importance of newspapers, particularly in less urban or in rural areas, where local and regional newspapers frequently have a dominant (monopolistic) position. Thus, such regional or local newspapers are often the only vehicle  to reach a bigger audience.

I like your characterization of how journalists face the proverbial “prisoner’s dilemma” when it comes to correcting their errors. Did you get a sense of the practice (or lack thereof) regarding corrections in Swiss and Italian papers?

Porlezza: At least in U.S. newspapers, there seems to be a consensus that errors need to be corrected, since such an act of transparency gives newsrooms more credibility. The famous quote from [Bill] Kovach and [Tom] Rosenstiel — news accuracy is “the foundation upon which everything else is built: context, interpretation, debate and all of public communication. If the foundation is faulty, everything else is flawed” – is still not very common in Europe, at least in the non-Anglo-Saxon part.

One of the main reasons … is the lack of a second editing layer within most European newsrooms. Until a couple of years ago, before the economic crises and several waves of layoffs, there were specialized people — professional proofreaders — who corrected the texts. Thus, journalists themselves never had a “close relationship” with errors-management.

Today, this is different: due to the layoffs, there are no more professional proofreaders so that journalists have to scrutinize their own articles. However, in most newsrooms established procedures are still missing. Journalists ask other journalists to read their articles, but usually this is done on an unsolicited basis. Moreover, usually there is no established quality discussion within newsrooms, as we found out in another research project called MediaAcT. The same applies for Italy, where quality management in newsrooms is even less existent.

However, today there is no longer any reason to put a blind eye on errors-management. One of the main reasons for eluding correction corners in newspapers was missing space. In a digital world, this is no longer valid.

Indeed, in Switzerland there are some initiatives that are very similar to what The Washington Post has done online. One of the biggest regional newspapers, the “Tages-Anzeiger” from Zurich, initiated a correction box for their online articles. For every article, readers can notify factual or formal errors with regard to spelling, punctuation, grammar, objectivity or technical aspects. Experience has shown that such innovations usually have to be adopted by leading media outlets, hoping that the herd instinct among legacy media works reliably.

Hence, the prisoner’s dilemma among journalists is very typical, particularly dealing with errors, since no one would like to be criticized within [the] newsroom. In a certain sense, this effect could be useful in order to avoid errors: if a newspaper publishes errors, journalists will try to avoid being mentioned in a correction corner, contributing in this way to reduce the error rate.

However, the most important prerequisite in newsrooms is to enforce an open dialogue about quality management with the conviction that in newsrooms, certain quality standards and procedures can be set up.

We have made it easy to comment on posts, however we require civility and encourage full names to that end (first initial, last name is OK). Please read our guidelines here before commenting.

  • http://twitter.com/jdthndr Jamie Thunder

    “In addition, the public should be invited to report errors directly to the newsroom by using innovative forms of collaborative action similar to crowdsourcing techniques that the Guardian applied to uncover the expenses scandal involving members of the UK Parliament” (p. 541)

    The Guardian did use crowdsourcing techniques for MPs’ expenses, but it’s surely a factual error to say it uncovered the expenses scandal – that was the Daily Telegraph…