When coverage outpaces interest – can statistics show us when the press is trying to lead readers to a story?

I’m working on some research on press coverage of Iran, wondering whether the heavy restrictions on international journalists reporting from within Iran led to a shift in journalistic practices, specifically a heavy reliance on citizen media. That’s basically Brian Stelter’s analysis in the New York Times – I’m curious whether this was a phenomenon specific to the Iranian situation, or whether this is a shift in journalistic practice.

In the process of researching, I came across a report from the Pew Research Center for People and the Press, ground zero for quantitative media analysis in the US, and the inspiration for a lot of the work we’re doing on Media Cloud. The June 24th report, an analysis of weekly figures of press coverage and reader/viewer interest, was titled “Iran’s Interesting… For a Foreign Story“. The authors pointed to the fact that 20% of 1000 Americans they’d surveyed that week listed the Iran protests as the story they were following most closely, a very high percentage for a non-US story.

What struck me was the graph that accompanied the piece, comparing news interest with news coverage. While 20% of readers listed Iran as the top story they were following, the Project for Excellence in Journalism News Coverage Index saw 28% of the newshole that week filled with Iran protest stories. (The PEJ Index (also a Pew project) is compiled by hand by a team of researchers who monitor a set of large and midsized newspapers, a selected set of blogs, radio and television broadcasts – while it doesn’t have the broad reach we’re trying to create with Media Cloud, handcoding means that there’s a high degree of reliability for story identification within the subset of stories they’re sampling.)

Okay, so 20% of people surveyed saw the Iran protests as the story they were following most closely, while 28% of the newshole was filled with Iran protest stories. So what?

Well, that’s pretty atypical, for two reasons. As the article suggests, it’s rare for international news to receive heavy attention from the American public. I reviewed Pew reports for this year and collected data that compared interest and coverage. Of the 152 stories I found that received the most reader attention (the top 5-6 stories readers reported the most interest in during the weeks Pew published this data during 2009), only 31 were primarily international stories. (I excluded stories about US politicans travelling overseas, like Obama’s trip to Russia and Italy.) Of the top 30 stories in the set, sorted by interest, only one was an international news story, while ten of the bottom thirty were international. This is consistent with Pew’s analysis of top interest stories for 2008, where the only international stories in the top 15 were Hurricanes Ike and Gustav (both of which struck the US, suggesting the coverage was primarily on US preparations and recovery) and the Beijing Olympics.

The second reason the Iran protest story was atypical is that Pew saw a higher percentage of coverage than interest in a story. That’s quite rare in this data set –
only 20 of the 152 tracked stories are represented more heavily in media coverage than in reader interest. (Another 12 had equal percentages of readers reporting greatest interest and percentage of media coverage.) It makes sense that most stories in this set – the five or six stories readers reported the greatest interest in for a given week – would have a higher interest than coverage percentage. Most news outlets report more than just the most popular stories – a good newspaper covers those popular stories, plus local and regional news, as well as stories that don’t generate a good deal of attention.

When a story occupies 28% of the newshole, but only 20% list it as the story they’re following most closely, it suggests that news media is pushing a story. Conversely, when reader interest vastly outpaces coverage, it suggests strong interest in a topic that the media might be downplaying (or simply not overhyping.) The top ten stories where interest greatly outpaces coverage include stories that might directly impact readers (three stories on sharp stock market drops, a story on a peanut recall, swine flu, unemployment) and celebrity stories like Michael Jackson’s death, the “octomom” story and the NCAA basketball tournament.

The stories where media coverage outpaces reported interest are harder to characterize. I expected to see a wealth of international stories in this set, supporting journalist friends who contend that they’re often pushing international stories despite lack of reader interest. But only five of the twenty in this set are international, almost exactly the percent of total international stories in the whole set. The stories where media attention appears to outpace interest are often focused on individuals in the US government beyond the US president and secretary of state – Senator Arlen Specter’s change in party affiliation, Justice Sonya Sotomayor’s confirmation to the Supreme Court, Rod Blagojevich’s resignation, Justice Souter’s retirement. Also in this set are stories about complex debates within government circles, like the debate about torture.

Some of my journalistic colleagues have been arguing that, by covering stories in detail, they can generate audience interest in complex stories. (The example that’s often thrown around is This American Life’s excellent “Giant Pool of Money“, which many anecdotally report increased their interest in financial crisis stories.) I see some evidence that this is true – in February and March, there’s only a small gap between interest and coverage on stories about the US economy, with two reports of more coverage than interest. By May, many readers are listing the economy as the story they’re following most closely, and the percent listing it as a top story greatly outpaces coverage. This might indicate that heavy reporting on the US economy helped generate interest. Or those factors may be uncorrelated. There aren’t many other stories that appear week after week in the Pew numbers that we can compare to, and virtually no international stories that appear in the set more than a couple of times.

I started looking at this data to see whether I could find evidence that reporters were pushing international news on an unwilling US audience.Does this data suggest that Americans are more cosmopolitan and interested in international news than I generally speculate? Probably not. Remember, the stories in this set are the ones that 1000 Americans, surveyed over the telephone, listed as the ones they’re following most closely in a given week. Extremely important international stories like the Honduran, Fijian and Malagasy coups/power struggles and the Indian elections didn’t make the list.

What’s most interesting to me are the international stories that generated high interest. They include stories about the Somali pirates (and the kidnapping of a US ship captain), the Iran protests, the ongoing Israel/Palestine conflict, the crash of an Air France flight near Brazil, and a number of North Korea-focused stories. At least half these stories aren’t reportable via conventional means – i.e., a journalist on the ground offering real-time, eyewitness reports. Somalia is too dangerous for American journalists to travel to, North Korea is closed to virtually all journalists and Iran has placed heavy restrictions on international journalists. Is it possible that we’re most interested in the news we’re not able to get?

I’d like to look more closely at Pew’s reported press coverage and interest statistics and see if a clearer pattern emerges over the course of years – that requires more data analysis than my eyes are really up to right now. I’d also love to do some cross-country comparisons – does anyone know of similar data sets (i.e., media coverage and reader/viewer interest) for other countries? It would be intriguing to see whether the US is particularly parochial, or whether we see a similar dynamic in other countries.