BROOKE GLADSTONE: Over the last two weeks, we've considered the charge of liberal bias on NPR. We've raised some big questions, how the enterprise of journalism frames political debates, what the words “bias” and “liberal”’ and even “NPR” mean. We've heard from listener Sam Negus, a conservative evangelical Christian who said he rarely, if ever, hears his own views on public radio. And we called Tom Rosenstiel, who tracks and quantifies media content, to bring us some hard data on NPR’s coverage. We also heard from many, many listeners. We asked four of them, conservatives, to keep diaries of their listening and note moments when they heard bias. Now, keep in mind, they wrote to us. They're listeners, which means ultimately they mostly like what they hear. In every case, they said they were less irked by bias than by public radio’s seeming unwillingness to admit it. We talked to two of them. First, we return to Sam Negus, who wanted to test his impressions with a diary. He found bias in this Morning Edition story, contrasting the immigration policies of Arizona and Utah.

[CLIP]:

RENEE MONTAGNE: I'm Renee Montagne. Arizona’s tough immigration law has received extensive coverage, and there’s been a lot of talk about similar measures in other states, yet one of Arizona’s neighbors, also known for its conservative politics, has taken a very different approach. NPR’s National Political Correspondent Mara Liasson reports.

MARA LIASSON: If you were to choose a state that would allow illegal immigrants to come out of the shadows, work and drive without fear of deportation, you probably wouldn't pick Utah.

[END CLIP]

BROOKE GLADSTONE: Okay, is that a fair snippet?

SAM NEGUS: Yeah, I think that was pretty accurate of, of why I, I wrote that down. This is one of those partisan issues on which I'm more coming from the direction of, of the piece. It didn't offend me. I'm immigrant myself, so I don't have a problem with immigration. But I, I think there will be a lot of conservative voters, a lot of Republican voters in Arizona who would have heard that and been maybe offended or alienated and, and certainly would have perceived a bias towards one side or another.

BROOKE GLADSTONE: Was there anything inaccurate in what you heard in the piece, or is this, once again, a question of tone?

SAM NEGUS: I think it was a question of tone. I think, I think the piece is great. I thought it was just very exemplary of, of the way in which NPR’s news coverage can kind of drift towards one side of an issue or another. The piece doesn't consider, to the same degree, the feelings of US citizens in the State of Arizona who feel alienated or upset, for whatever reason, that there is this enormous population of undocumented migrant workers in their state.

BROOKE GLADSTONE: Mm-hmm.

SAM NEGUS: So I would think that if I were a kind of mainline Republican voter in the State of Arizona and I associated with that policy and approved of my representatives passing that policy, I would hear NPR telling me the policy that you like is bad and the one that we like is this.

BROOKE GLADSTONE: Mm-hmm.

SAM NEGUS: I think that’s what that piece was saying.

BROOKE GLADSTONE: Sam said that he found keeping a diary a useful exercise because it put his own perceptions to the test.

SAM NEGUS: I'm not changing my perception that NPR is kind of coming from the left. I do hear that. But it’s one thing to kind of hear a tone or, or a preference. It’s another thing to actually have a solid statement that you could put down in a court of law and say, Your Honor, this person said this. So when you do listen to those, those news shows, and I, I listened to a bunch of news shows this week, and I - I felt like, yeah, I can still kind of, kind of hear that, but how would I go about proving it? And, and if you can't really kind of prove it in that concrete way, then that’s the space in which you have to start thinking, okay, to what degree is m – is my sensitivity a factor in pinning this issue down?

BROOKE GLADSTONE: That was Sam Negus. Kevin Putt, a lifelong conservative, found bias in a question asked by All Things Considered co-host Michele Norris. She was interviewing Intel CEO Paul Otellini on a proposal that corporations that build factories in the US be given a five-year tax holiday. The question that Kevin Putt objected to was also noted by the conservative media criticism site, NewsBusters, which called it “a liberal question that demonstrated liberal skepticism.”

KEVIN PUTT: Her question, which I – isn't unreasonable, her question was, can the country afford that right now?

BROOKE GLADSTONE: Let's listen to that.

[CLIP]:

PAUL OTELLINI: Anyone that wants to build a, a new factory in this country, whether it’s a American firm or a foreign firm, why don't we give them a, a five-year tax holiday? It doesn't cost anything, right? You’re just deferring the tax revenues that you would ordinary get, but meanwhile you get a factory and you get jobs, and everybody wins.

MICHELE NORRIS: Can this country afford that right now?

PAUL OTELLINI: Well, it doesn’t cost anything.

[END CLIP]

BROOKE GLADSTONE: Let me tell you what I just heard.

KEVIN PUTT: Okay.

BROOKE GLADSTONE: As a reporter, I heard a reporter asking a very natural follow-up question that you would ask anybody.

KEVIN PUTT: Right. Well, to me it sounded like a presumption that foregoing any tax revenue right now is a bad policy idea. And that is associated, from my perspective, with a liberal point of view. Sometimes bias, you know, shows up very subtly in, in the buried premise of a narrative. That’s what I think was happening with Michele.

BROOKE GLADSTONE: And you thought it was a fair question?

KEVIN PUTT: Yeah, I thought it was a fair question that revealed a point of view.

BROOKE GLADSTONE: Do you think she would have been doing her job if she hadn't asked the question?

KEVIN PUTT: Oh, easily. She could have had that interview without asking the question, “Can the country afford that right now?”

BROOKE GLADSTONE: But you really feel that the discussion with Intel’s CEO would have been less biased if Michele Norris had not asked that question?

KEVIN PUTT: I mean, obviously, you didn't hear – you don't hear it that way, because you’re disagreeing with me. But my ear definitely heard it that way. And I go back to the spending/taxing debate that Congress is dealing with, and there are two sides to it. Are we spending too much? Are we taxing too little? And her question had to do with foregoing tax revenue.

BROOKE GLADSTONE: But you heard both sides.

KEVIN PUTT: Oh, so your point of view is that if he had the opportunity to respond to it, then that would be your concept of there not being any bias. It’s not that it’s bad reporting; I think it’s good reporting. I'm a fan, so it’s, it’s not like when I hear it, it turns me off. It just comes with the territory.

BROOKE GLADSTONE: That was Kevin Putt. Among our critics there were also a fair number of liberals who felt NPR actually leaned the other way. Next up, Steve Rendall, senior analyst at Fairness and Accuracy in Reporting, or FAIR, a liberal organization that monitors media bias. In a controversial study released in 2004, FAIR counted up the liberal and conservative sources cited in news reports on Morning Edition and All Things Considered.

STEVE RENDALL: And what we found was a very strong slant in favor of the GOP. Sixty-one percent of partisan guests who appeared on those two NPR shows in 2003 were Republicans.

BROOKE GLADSTONE: There was a Republican Congress, there was a Republican White House. I mean, doesn't that make sense?

STEVE RENDALL: You should see a few more Republicans on, but the number was 61 percent Republicans to 38 percent Democrats. And, we were repeating a study that we had done in 1993, when the Democrats had the White House and both houses of Congress. And in that study we found that there was the same bias, 57 percent Republicans at that time and 42 percent Democrats. So it didn't matter who was dominating Washington. Republicans had more guests.

BROOKE GLADSTONE: Now, I'm assuming that at least a third of our listeners, the third that identify as conservatives, and maybe a good number of the liberal listeners too, are thinking you’re a liberal research organization, and you make no bones about it. Why should we trust what you say?

STEVE RENDALL: Well, our studies are replicable. You could duplicate the studies. You can check the numbers. I think that, first of all, everybody comes from a point of view. Everybody is subjective. This is one of the things in journalism that you try to overcome. You accept that you have a point of view, you have a certain amount of ideological baggage, and you try to overcome that. And the way to overcome bias is by balance. We do come from a point of view that says that there are certain groups that are, that are slighted by our media, both public and corporate media. Now, does this mean that we would surmise that NPR is a conservative outlet? I wouldn't say that. But the thing is, we've had four decades of formal campaigning by the right, by groups like Accuracy in Media, the Media Research Center, the Heritage Foundation, to portray our media, corporate and public broadcasting, as being to the left of center. It’s paid off. And I think the fact that we're having this discussion here, the fact that there’s a debate in Congress, shows how much it’s paid off.

BROOKE GLADSTONE: And, not because there’s a kernel of truth in it.

STEVE RENDALL: Well, I would love to see the studies. I have looked at the studies, I have combed the literature, and I just haven't seen anything that really shows that to be true.

BROOKE GLADSTONE: That was Steve Rendall of Fairness and Accuracy in Reporting. And here’s another study. Tim Groseclose, a professor in the Economics and Political Science Department at UCLA, and Jeff Milyo, an economics professor at the University of Missouri, analyzed 20 mainstream news outlets, counting each time they cited a think tank or policy group in a news story. They gauged the political stripe of a think tank by how many times it was cited by a conservative or a liberal member of Congress. The Congress people themselves were rated based on their roll call votes. So, bear with me here. Let's say that based on her votes Michele Bachmann is deemed the most conservative member of Congress and she has a serious habit of citing the Christian Coalition, and so does The New York Times. The New York Times would thus be deemed to have a conservative bias equivalent to Bachmann’s conservatism. But that’s a hypothetical. The Times doesn't have a conservative bias. In fact, based on this criteria, 18 of the 20 media outlets studied were liberal, the most liberal, based on news coverage, the Wall Street Journal. And, what of NPR? Morning Edition was certainly rated as liberal but no more liberal than Time, Newsweek, U.S. News and World Report and slightly less liberal than The Washington Post and much less liberal than The New York Times. The PBS NewsHour with Jim Lehrer was closest to the center, as unbiased as any of the news outlets measured. We'll link to the Groseclose-Milyo study on our website. And, some more data for your delectation. Last week we spoke to Tom Rosenstiel of the Project for Excellence in Journalism, part of the PEW Research Center. The project researches media bias all the time and vowed to return this week with some data specifically related to NPR. One study assessed the coverage of 50 news outlets of President Obama’s first 100 days in office. Rosenstiel looked to see how NPR stacked up.

TOM ROSENSTIEL: Twenty-eight percent of the stories on NPR were positive about Obama versus thirty-seven percent in the media, generally. Fifty-two percent of NPR’s stories were neutral or mixed, versus forty percent in the media overall, and twenty-one percent of NPR’s stories about Obama were negative, versus twenty-three percent with the media, generally. So in this case, the bias of NPR, at least according to these findings, would be toward more neutrality than the media generally.

BROOKE GLADSTONE: We had a number of people write to us to say it isn't just what the NPR newscasters say. It’s the selection of stories that they focus on. In 2010, what you did is you took representative samples of NPR’s news coverage and you compared it to talk radio and headline radio. What about NPR’s supposed bias towards liberal issues, say, race and gender and gay issues?

TOM ROSENSTIEL: For 2010, one percent of the time studied on NPR was devoted to those issues. That was exactly the same percentage as we found on commercial headline services, the CBS and ABC headlines that you hear at the top of the hour. Talk radio, which is dominated by conservatives - and our sample is weighted according to audience samples, so it’s more heavily conservative – actually devoted twice as much time to those issues as NPR did in 2010.

BROOKE GLADSTONE: Let's return again then to NPR’s coverage of President Obama. Conservatives say that Obama gets a lot more press on NPR. How much coverage was there compared to talk radio?

TOM ROSENSTIEL: We have a category of stories that are focused specifically on the administration, and what we see here is that one percent of the coverage on NPR last year was focused squarely on Obama. That was the same, again, as on the headline services. But that’s about a tenth of what we get on talk radio. And this is typical. What we see in talk radio, and we see it in cable news, as well, is that wedge issues that relate to areas of controversy or agitation get more coverage.

BROOKE GLADSTONE: In every example we've gone through so far, NPR is either in synch with the mainstream media or more conservative. Am I right?

TOM ROSENSTIEL: Well, not exactly. You know, I think what you basically see in terms of story selection is that NPR is more focused on international affairs, they're more focused on sort of explanatory sorts of coverage. There’s lots of examples in the data here of spending more time on stories about things like Afghanistan and less time than some other parts of the media, particularly the more commentary or ideological components of the media, on divisive issues like health care reform or even the economy. There’s less coverage of the economy on NPR and more coverage of policy issues.

BROOKE GLADSTONE: Now, you referred last week to a consumption study. Could you share that?

TOM ROSENSTIEL: Yes. This is a part of a survey that we do with our colleagues at the PEW Research Center for the people in the press every two years, called the Media Consumption Survey. And we asked a number of things, including what it is that regular users of any outlet like about that particular outlet. We asked, do you go there for the latest headlines, for in-depth reporting, for views and opinions or for entertainment? And NPR stood out on that question, in that it was the only outlet where the number one reason that people went to that site was for a mix of all of those things. It was, in fact, the only outlet that was in double digits across [LAUGHS] all these categories. Another thing that we see that’s interesting about NPR is that it doesn't skew toward any particular age group. Eleven percent of -

BROOKE GLADSTONE: Hm.

TOM ROSENSTIEL: - Americans overall listen to NPR, are regular listeners to NPR. Eleven percent of those who are 18 to 29 listen to NPR. Twelve percent of those who are 30 to 49, twelve percent of those who are 50 to 64 and 10 percent of those 65 and older listen to NPR. This is very unusual. Now, NPR’s audience is a more liberal audience than conservative, but not dramatically so. It skews 2 to 1, Democrat versus Republican, but that is not a particularly unusual percentage. Sean Hannity skews 15 to 1 Republican versus Democrat. The New York Times skews 4 to 1 Democrat versus Republican. The only outlets that tend to reflect the population across the board, without any ideological tilt at all, are ones that are community based, so local TV news, people’s local newspaper.

BROOKE GLADSTONE: So help me out here, Tom. Stake your ground. Is NPR biased?

TOM ROSENSTIEL: [LAUGHS] I don't think that our data can answer that question. On story selection, NPR is more international in its focus, clearly. You are gonna hear somewhat more about policy. You are gonna hear somewhat less about political argument. Does that represent bias? Ultimately, I think these questions of bias may be in the eye of the beholder.

BROOKE GLADSTONE: Okay, and taking all of that into consideration, just one last question: Is NPR biased?

TOM ROSENSTIEL: [LAUGHS] I think every listener in the end has to determine that for themselves.

BROOKE GLADSTONE: Okay, but do you think NPR is biased?

TOM ROSENSTIEL: [LAUGHS] Brooke, I'm not going to answer that, that way.

BROOKE GLADSTONE: Why not?

TOM ROSENSTIEL: [LAUGHING]

BROOKE GLADSTONE: You've looked at the coverage of politics, you've looked at a variety of heavily freighted issues, and you say the data offers hints. Then you can't just walk away from your, your life’s work and say, I refuse to comment.

TOM ROSENSTIEL: Well, do I see proof of bias in the story selection? No. Do I see evidence in our tone studies that prove a liberal bias? No. Does this prove that NPR is not biased? I can't say that. We don't have tone for every topic. In the end, it’s very hard to establish, even if someone were to identify who’s on the air and what their political affiliations are. If you have a lot of people from one party on and the questioning is very tough, that goes one way. And if you have a lot of people on and the coverage are a bunch of softball questions, that goes another way. Bias, in the end, is often a matter of whether things are phrased in ways that I agree with or disagree with. In the end, you’re not gonna persuade anyone with data.

BROOKE GLADSTONE: Tom, thank you so much.

TOM ROSENSTIEL: [LAUGHS] My pleasure, Brooke.

BROOKE GLADSTONE: Tom Rosenstiel is the founder and director of the Project for Excellence in Journalism, a research organization that is part of the PEW Research Center – studies the news media.

[MUSIC UP AND UNDER] So, political bias doesn't show up in the data. And tone? Well, that’s still in the ear of the beholder. Darn!