Alicia C. Shepard
Every day emails, phone calls and letters flood my office. They say basically the same thing about the presidential campaign but each comes from a partisan political perspective.
NPR is biased.
Too much coverage, one side says, is aimed at NPR's obvious goal of getting Sen. Barack Obama elected. "You cover Obama's campaign endlessly," wrote Javad Sarvestani of Los Angeles. "And when you do devote time to John McCain you seem to aim your coverage more at undermining his candidacy than reporting on it."
The other side says that NPR is purposely ignoring Obama and instead only doing stories that reflect well on McCain. "Recently, I have heard several interviews with people who support Senator McCain and I can't remember when, if ever, I have heard a positive interview or report on Senator Obama and the excitement his candidacy has generated," wrote James Summers. "I expect the commercial networks to obfuscate most of the time. I expect more from NPR."
Between June 9 and Oct. 19, 2008 my office received 282 emails specifically accusing NPR of favoring Obama and 252 emails accusing NPR of favoring McCain. Hundreds more insist the network is either too conservative or too liberal in general, and the writers allege specific bias with particular stories.
The volume and frustration from those who claim NPR is biased have troubled me since I became Ombudsman a year ago. Is there a way for me, as an independent voice for the public, to determine if NPR is biased? Instead of getting an answer, I got a lesson in how passionate NPR listeners can be.
The Ombudsman's office decided to study all stories and interviews about the presidential campaign airing on Morning Edition, All Things Considered (ATC) and the corresponding morning and evening weekend shows. We studied 429 campaign stories that aired between Aug. 1 and Sept. 30, a period that covered the two political conventions and the first weeks that followed.
We put each story into one of three categories: It focused primarily on Obama (or his running mate Joe Biden), or on McCain (or his running mate Sarah Palin), or about equally between the two. Our study focused on news coverage in the major shows and did not include commentaries, special convention coverage or newscasts. Click here for PDF with charts.
We found that, overall, NPR shows ran more stories about, and devoted more air time to, McCain and Palin than Obama and Biden.
During August and September, Morning Edition and its weekend equivalents aired 96 stories totaling 6 hours and 5 minutes about McCain-Palin. All Things Considered carried 72 stories in 4 hours and 47 minutes. Overall, these NPR shows focused on the McCain-Palin ticket in a total of 168 stories that covered 10 hours and 52 minutes.
On Obama and Biden, Morning Edition ran 74 stories in 5 hours and 1 minute and ATC ran 72 stories in 4 hours and 6 minutes, for a total of 146 stories that covered 9 hours and 7 minutes.
Of the 429 stories studied, 115 focused about equally on both candidates.
Overall, McCain and Palin got nearly two more hours of air time on NPR than Obama-Biden.
So, do these statistics mean that NPR is biased toward McCain? Or, as I suspect, did McCain's unexpected choice of a little-known running mate mean that NPR journalists spent a lot of time filling in the blanks in a way that was not needed for six-term senator Joe Biden?
"Obviously Palin became the new main focus of the campaign beginning the weekend before the Republican convention and dominating the news for several weeks thereafter," said Ron Elving, senior Washington editor who's in charge of NPR's political coverage.
"I think it's apparent that this coverage was more analytical than promotional, and that Palin was far more of a new subject than Joe Biden or by that stage of the campaign, Obama himself," he continued. "But for this period of time, taking into account all the show elements in addition to the reported pieces, it's not surprising that Republican ticket stories were more common than Democratic."
Elving noted that last winter Obama and Mike Huckabee, who won the Republican Iowa caucus, received more coverage than other candidates because they were new faces.
Palin certainly was a new face. Our study showed that 95 of the 168 stories that NPR did on the Republican ticket focused on Palin. She received 6 hours and 21 minutes of coverage during the two-month period examined — more time than was devoted to McCain.
It also needs to be said that counting the number of stories or minutes does not necessarily give a complete picture of how a broadcast network covers political candidates. For one thing, a 2-minute piece with riveting tape could pack more of a punch than an 8-minute piece that seems to drone on. Also, story length says nothing about content, and the overall impression it leaves listeners. A piece focusing on negative aspects of a candidate or his campaign obviously can give listeners one impression, while an upbeat piece has the opposite impact.
(A Pew Research Center campaign coverage study last week indicates that across the media McCain got a lot of coverage but 60 percent of it was negative, while Obama's coverage was 29 percent negative. It should be noted that Obama, and later McCain, received negative coverage as each began to drop in the polls.)
While our study is not a scientific one, it provides an illuminating snapshot of NPR's coverage during a two-month period.
But when it comes to core political beliefs, even scientific research may do little to persuade many complaining listeners that NPR isn't biased. It's just the nature of who we are as human beings and our passionate attachment to our core political beliefs.
Philo Wasburn, a Purdue University sociology professor who just co-wrote a book on media bias, knows this well. He told me that research going back to the 1960s shows how difficult, if not impossible, it is to change people's central core beliefs.
"When people are really committed to some ideological position, especially with politics, even if you present them with empirical evidence that supports the opposite of what they believe, they will reject it," said Wasburn. "Core beliefs are very, very resistant to change."
People also are often resistant to accepting statements or information that goes against core beliefs. While it may be inconceivable to Democrats that Palin is qualified to run the country, for example, it is undoubtedly inconceivable to her Republican supporters that Obama is qualified to be president.
Perceptions of bias typically arise when a listener disagrees with something said about a presidential candidate, explains Wasburn. So when listeners hear something they disagree with in, say a perfectly balanced piece about either Obama or McCain, they reflexively decide NPR is biased.
And they often complain to me. But I'm coming to believe that there is little I could say or any evidence I present that would alter either side's perceptions of bias.
"What they are really annoyed about is that something was said about McCain or Obama that they didn't like," said Wasburn. "The person calling in and annoyed with NPR most likely doesn't have scientific evidence. They are annoyed by something said or by some guest. It doesn't matter that their perception is not true. So soldier on. You are not going to change any minds."
Timothy Groseclose is a political science professor at the University of California, Los Angeles, who also studies media bias. He and another professor published a study in 2005 that concluded that 18 of the 20 major media outlets studied (including NPR) were left of center, as compared to the average U.S. voter. Only Fox News' Special Report with Brit Hume and The Washington Times scored to the right of the average U.S. voter. (Results are on P. 22 of PDF.)
"By our estimate, NPR hardly differs from the average mainstream news outlet," said Groseclose. "It had the same scores as Time, Newsweek and was slightly less liberal than the Washington Post and well to the right of the New York Times and CBS Evening News. One of the surprising findings is that NPR is not as left as everyone says it is."
Groseclose and co-author Jeffrey Milyo, a University of Missouri economist, published their results in the respected Quarterly Journal of Economics. They took no outside research money to avoid any perceptions of bias. They made sure half their researchers had supported a Democratic presidential candidate and the other half a Republican. Their study was scientific and their methods transparent, so the methods could be tested or duplicated by others.
NPR got a score of 66.3, with 50 being centrist and 100 being most liberal. The Wall Street Journal's news pages (not the well-known conservative editorial pages) got an 85.1 and The New York Times and CBS each got a 73.7.
Although this study shows that NPR is relatively less biased that some other major news organizations, I doubt it will sway those who are convinced that NPR is liberal and pro-Obama — or those who have the opposite impression that NPR is siding with McCain.
The results didn't sway liberal bloggers either.
"Just Google my name and you will find all kinds of left-wing blogs saying why my study used the worst method ever," said Groseclose. "Now, if I had found that most media outlets have a conservative bias, those same blogs would have said, 'Here's the best method ever and here's a study by a respected scholar.' They don't say that because they don't like the results."
As we near the end of an emotional presidential campaign, it is worthwhile for all of us to inspect our core political beliefs and ask if something we heard on NPR (or any other news media outlet, for that matter) is truly biased or just doesn't jibe with what we believe or what we want to hear.
"Hardly anyone's personal world view or value system is entirely neutral when it comes to questions of public life," said Elving. "But professional journalists take it upon themselves to present what they report with as much neutrality as possible. That is how we are trained."
And while that's an article of faith in the profession, it is another statement that people with strong views — in this case about the news media — may not want to hear or accept.