Freakonomics Radio - 62. How Biased Is Your Media?

Episode Date: February 15, 2012

The left and the right blame each other for pretty much everything, including slanted media coverage. Can they both be right?

Glenn Beck, I'm an entrepreneur. I am a reluctant, believe it or not, commentator and dad. All right. Now to the theme of our conversation today. When I say media bias, you say what? Yes. From WNYC and APM American Public Media, this is Freakonomics Radio. Today, how biased is your media? and how do we know? Here's your host, Stephen Dubner. Let's say that Glenn Beck is right, that the news media is biased, whatever exactly that means.

Now, before you start foaming at the mouth, because I know that's what you're going to do, let me just say that you'll also hear from someone on the other side of the aisle, someone who sees things very differently. I don't know where to begin in describing how completely ridiculous I think that is. I mean, pick a spot and begin. Well, I mean, I don't. That's Andrew Rosenthal. I'm the editorial page editor of The New York Times, which means I'm in charge of the editorials. We'll hear more from Rosenthal later, but first, let's get back to the assertion that media bias is real and that it's a real problem.

How do you prove that? How do you measure something like media bias rather than just opine or bloviate about it? Here's Steve Levitt. He's my Freakonomics friend and co-author. So measuring media bias is a really difficult endeavor because unlike what economists usually study, which are numbers and quantities,

media bias is all expressed in words. And so in the last five or ten years, there have been some really tremendous advances in how we think about text as data. So how we take words and transform words into quantitative measures. And at the forefront of this endeavor have been people like Jeff Milo and Tim Grosklos and my colleagues Jesse Shapiro and Matt Jenskow. And so take the Grosklos and Milo work, for instance. What they did was they wanted to figure out how to compare media sources like the Wall Street Journal or the New York Times or

Fox News, how liberal or conservative those outlets were compared to, say, politicians. So there were 20 media outlets that I examined. Also, the original article had a co-author, Jeff Milo. That's Tim Grossklos. I'm a professor of political science and economics at UCLA. Grossklos wanted to see if he could answer questions like, how does the average article in the New York Times with its supposedly liberal slant compare to the average speech by a Democratic heavyweight like Harry Reid?

And similarly, how does coverage on Fox News with its supposedly conservative slant compare to the average speech by someone like Michelle Bachman? To make comparisons like this, Grosskloos had to start with what he knew, with what was easy to quantify, that is, the political leanings of politicians themselves. So to that end, he assigned each Congress member a score, which he called their political quotient or PQ. And this is just a way to say precisely how liberal someone is. So

100 is very liberal. It's about a Nancy Pelosi, zeroes, Michelle Bachman. These are all based on roll call votes in Congress. And in fact, I let the Americans for Democratic Action pick the roll call votes for me. This is a liberal interest group. And with each roll call vote, the ADA decides whether the yay or the nay alternative is liberal. Now, if you were listening carefully, you heard Grossklos say that he, quote, let the liberal ADA set the standard for each congressperson's political quotient, or PQ, which might make you wonder about Gross-Kloss's own PQ. So before moving forward, let's back up a bit. As part of this research project, Gross-Kloss wrote a quiz that anyone can take to assess his or her own PQ.

My hunch is that if you and Steve Levitt took my PQ, my best guess is you would turn out to be kind of left-leaning moderates. Yeah, Levitt put himself at about a 45. Is that right? Yeah. I would have predicted 55 or 60. But he didn't actually take your test. I think he guesstimated. And I'd probably put myself at – and again, I haven't taken it either yet.

I'd probably put myself at about 55 or 60. You're probably right on there. So together, we're right down the middle. And I say my political quotient is a 13, which means I'm very much on the conservative end, not quite Michelle Bachman, but kind of near John Boehner, Mitch McConnell. Give us some well-known politicians and what their PQs are, keeping in mind that 100 would be pure liberal and zero would be pure conservative, right? Right. Yeah. So like Nancy Pelosi, Barney Frank would be about 100. Barack Obama is not the most liberal on this scale.

Barack Obama is about an 88. Hillary Clinton was something like an 87.9. They were almost exactly tied on the scale. Joe Biden would be something like an 85, 84. Harry Reid, 80. Joe Lieberman was a 74. Now, I computed two PQs for Joe Lieberman, one when he was a Democrat, one when he was an independent. They were almost exactly the same. I think when he was a Democrat, it was like 74.7 as an independent, a 74. So he moved just the teeny, teeniest bit right after he switched from Democrat to independent. Okay. So Grossklos had given each politician a PQ or a political quotient with a hundred

representing a hardcore liberal and zero a hardcore conservative. Now, to make the connection between the politicians' leanings and the leanings of media outlets, he needed to take an intermediate step. This will be a little confusing at first, but bear with me. The intermediate step was to take more than 150 think tanks and interest groups and assign each of them a political quotient. The Center for American Progress. The Heritage Foundation. The Brookings Institution.

So now, having determined a political quotient for each of these think tanks, Grossklos could start to measure media bias. How? Well, he simply counted how many times the name of each of these think tanks and interest groups were cited in the 20 major media outlets he was studying. So let's say that a given newspaper cited a group like the liberal Citizens for Tax Justice much more often than it cited the conservative Americans for tax reform. That would drive up that newspaper's liberal score on what Gross Close called a slant quotient.

He used the same scale for the slant quotient as he used for the political quotient, which is that 100 is the most liberal and zero the most conservative. The New York Times got about a 74 of the 20, 18 of the 20 leaned left. Now, just about none of them were to the left of the average Democratic speech. So some of my conservative friends said the New York Times sounds about like a Joe Lieberman speech. Conservative friends said that's it? That's not very liberal. So in some ways, NPR was something like a 67, you know, so even to the right of a Joe Lieberman speech. So in some ways, my results, I think, say that the media are in some ways more centrist than lots of people have been saying. On the other side, I analyzed one Fox News show. This was special report when I analyzed it. Brit Hume was the anchor.

That slant quotient was a 40, which means it was 10 points right of center. And so it sounded about like Olympia Snow, Susan Collins speech. So, Gross Close's argument, based on his research, is that most news organizations empirically lean to the left, although not as dramatically as some critics might suspect. He ultimately wrote up his findings in a book called Left Turn, How Liberal Media Bias Distorts the American Mind. Now, how did he come to that conclusion, that the American mind is being distorted by media bias? Well, Grossklos combined his own findings and existing research to calculate that the average American voter has a natural PQ, or political quotient, of around 25 to 30, which is firmly in the conservative range. But as Gross-Klose sees it, the left-leaning media pulls some of those naturally conservative voters into the center,

which is why we generally vote about 50-50. Without media bias, Gross-Klose says, we'd be a much different country. So I suggest that we would be about like Texas or about like Kentucky. It might be even more conservative. If Tim Gross Close is right, and it's a big if for sure, I wanted to know this. How does that bias happen in a newsroom? Does journalism simply attract more liberals than conservatives? And think of all the choices a given writer, editor, producer will make in a given day. Which story to cover and which to not.

Which sources to cite and which to leave out. Which conclusions to draw and which to leave to the reader. Well, it's not about the culture of the newsrooms per se, because all sorts of people work in there, although I would say that the NPR newsroom tends to be much more like a liberal college fraternity type of environment. That's Juan Williams. He's a political analyst for Fox News. He used to work for NPR and Fox News at the same time. You may remember that in October of 2010, NPR fired Williams for some comments he made on the O'Reilly factor about Muslims. But when I get on a plane, I got to tell you, if I see people who are in Muslim garb,

and I think, you know, they're identifying themselves first and foremost as Muslims, I get worried. I get nervous. He was not expecting to get fired. I was in shock. I don't think I knew what to do. I went and ate some Chinese food. I did another TV show before I spoke to anybody about it. And that tells you how crazy I was, kind of lost in my mind over this thing. Naturally, Williams became a poster boy for media bias,

which was strange, or at least seemed strange to me, because he has one of the most nuanced views on media bias that you're likely to find. Well, the process, it can be amorphous, but here's the thing. It's a creative process. And I remember once sitting at a lunch with Catherine Graham, former publisher of The Washington Post, and we were talking about newspapers. And I don't know why, but I said to her, well, you know, 10 years ago, this would be a totally different newspaper. You know, I have a different group of people, different opinions, different strengths, different interests.

And everybody at the table said this is not a programmed, constant feed of news happening out there and the newspaper or the news program is a direct reflection of that news. To the contrary, it's a reflection of the people and the agendas of people in the room as they begin to tell their story. For example, I find that as baby boomers have moved through American society, the political society, the cultural society, the economic society, that they have, in essence, told their story and told it loudly. They are a dominant voice. So their interest, their point of fascination, in fact, their self-reverence has dominated news coverage for the last three decades. It's an interesting point Williams makes.

Putting together a news report is inherently a creative process and it'll reflect the people who do it to some degree. But is it really their decision? What about the owners and managers of these media properties? How is it that media firms choose their content, basically? Or how is it they choose to have conservative slant or liberal slant? That's Matthew Jensko. He's an economist at the University of Chicago. Along with a colleague named Jesse Shapiro, Jensko set out to ask this question. If newspapers are slanted, does the slant really come from the newsroom or maybe the corner office? really an economic decision that looks like any other firm choosing what flavor of ice cream to offer, what kind of shoes to sell? Or is it something different? Are those decisions driven by the personal tastes of the owners of media outlets or their political agendas? So Jensko and Shapiro wrote a computer program that would sort through millions of articles from hundreds of newspapers.

The program's job was to identify politically partisan phrases. And then, of course, the crux of the matter is how do you define what's a liberal phrase and what's a conservative phrase? And our idea, which builds on some work that Tim Groskos and Jeff Milo had done earlier, was to look at the speech of politicians, look at speeches by congresspeople. The nice thing about congresspeople is we know which ones are liberal and we know which ones are conservative. That's easy to measure. So we could ask, what are the phrases that liberals or democrats tend to say a lot relative to conservatives? And what are the phrases that conservatives tend to say a lot relative to conservatives and what are the phrases that conservatives tend to say a lot relative to liberals.

So they fed the entire text of the 2005 congressional record, which is a transcript of every congressional proceeding and debate into their computer program. And what did it spit out? I just have to remember with this table whether we – whether you should read down or across to get the top ten. I'll read the top ten down. So the top ten phrases used more often by Democrats are private accounts, trade agreements, American people, speaking out. Tax breaks, trade deficits, oil companies, credit card,

nuclear option, war in Iraq, and middle class. The two-word phrases on the other side used more often by Republicans are stem cell. Stem cell. Stem cell research. Natural gas. Affordable natural gas. It gets pretty cold in Omaha.

Death tax. Death tax. I'd vote for it. Illegal aliens. Illegal aliens. Illegal aliens into the United States. Class action. Class action. Class action reform is badly needed.

War on terror. Of our war on terror. You're winning the War on terror. Of our war on terror. You're winning the war on terror. In the war on terror. Embryonic stem. I am 100% opposed. Embryonic stem cells. To embryonic stem cell research.

To embryonic stem cells for research purposes. Tax relief. This isn't a tax relief bill. Tax relief. Tax relief. Illegal immigration. Do something about illegal immigration. And date the time.

What was that last one? Morning hour be deemed expired. The journal proceedings be approved to date. The time for the two leaders be reserved. So, I mean, I really like the phrase date the time because it reminds people that this is an automated method that we involve no judgment on our part and that these things are just spit out by the computer. Republicans had the congressional majority in 2005. And it's the majority that uses this kind of procedural language more often. That's why date the time makes the list.

Other frequent Republican phrases? Change hearts and minds. Border security. And believe it or not. Change hearts and minds. Border security. And believe it or not. Grand Ole Opry. 2005 was the 80th anniversary of the Grand Ole Opry. Some common Democratic phrases. The Arctic refuge.

Living in poverty. And Rosa Parks. This was shortly after Rosa Parks' death. So having categorized all this language along Democratic and Republican lines, Jensko and Shapiro looked at how often a given newspaper used each of these signature phrases. And from that, they were able to determine each newspaper's political slant. But it was the next step that really mattered, figuring out where a slant comes from. In other words, is it that reporters have a bias that gets into their stories? Or maybe newspaper owners demand a certain line of coverage?

They looked into these factors and more, including one very clever indicator, the voting patterns of the people who read a particular newspaper. They're finding the most important factor driving the slant of a given newspaper is the political leanings of the people who buy it. In other words, newspapers are giving the people the news that they want. Jason Kuznicki Yeah. I think the broad conclusion of our paper is that newspapers look just about like every other firm in the economy. What the people making the decisions in the newspapers are doing is trying to sell newspapers. It may be that the reporters have their own personal political views. It may be that the owners have their own personal political views. It may be that everybody involved would love to push their own personal political views a little bit more. But something we show in this paper is that if they did that, it would be really,

really costly. They would lose a lot of money. And I think at the end of the day, most of the time in most places, the people in control are not willing to give up lots and lots of money in order to change the content of their newspapers to satisfy their own personal views. Coming up, we ask Ann Coulter what she'd do if the New York Times came to her for help. I would dance a jig. And we go inside the Times to talk to the editor who gets this kind of email. Thank you for making me laugh this morning.

Another blithering idiot at the New York Times. Why don't you put on a tight skirt and just be a cheerleader for Obama? From WNYC and APM American Public Media, this is Freakonomics Radio. Here's your host, Stephen Dubner. So earlier we heard from Tim Grossklos, a UCLA professor, who says that the U.S. media is categorically empirically biased to the left. Now, if you think that's true, what do you do about it? Here's his idea. I would say maybe hire more conservatives. You know, Ann Coulter, the commentator and author, she's on Gross Close's side. I understand that Tim Gross Close has this quiz on his website.

You test your political quotient. You can kind of align yourself, match yourself to various politicians. Did you take that test? How did you do on it? Well, I would say I got an A-plus as a conservative. So let me ask you this, Anne. If, let's say, you know, Jill Abramson, editor of the New York Times, comes to you and says, you know what? I think you're right. I think all these years you've been right and we've been wrong, and we are starting over. We have a great infrastructure here.

We know how to report the news, but we want to render the news more down the middle. What are the top priorities that you would give, let's say, the New York Times? Hire 10 conservatives. It's as simple as that? It's just find a different set of people. I mean, just having a conservative around, it breaks the cocoon effect, I would imagine. A conservative working at the Times would be able to tell them, David Brooks isn't a conservative.

Could we have someone else representing us? And I think that's all it would take. I really, I mean, maybe there is invidious bias and they just secretly hate us or openly hate us. But I really think it's mostly a matter of not knowing any conservatives. Colter very much agrees with Tim Grossklos that a biased media pushes the American electorate further to the left than what it would naturally believe. I don't know where to begin in describing how completely ridiculous I think that is. I mean, pick a spot and begin. That's Andrew Rosenthal. I'm the editorial page editor of the New York Times, which means I'm in charge of the editorials, letters to the editor, the columnists who work for the New York Times, and the op-ed stuff, which is people who don't work for the New York Times

who we invite to write their opinions, and that's in the paper and online. So here's the thing. I used to work here, as you know, and when you work here, even if you're a fan of newspapers and journalism, which some people are, but not so many, you understand that there is a demarcation. There is a Maginot line between, or maybe a DMZ is a better phrase, between what you run, which is called Opinion here, and the news shop. So can you describe how it works here? Sure. Well, the purpose of it, quite simply, is to keep the expressed opinions of people who are journalists who express their opinions out of the news columns. It is to avoid the contamination of news with opinion. Not the other way around, obviously, because there is lots of news in opinion writing. And that is to maintain the independence of the news report.

The Times was historically one of the first independent newspapers. And by independent, I mean politically independent. Before the current publisher's family bought it in 18-whatever it was, the New York Times was a Republican newspaper Which people will find hilarious The guy, the first editor of the Times Actually was Abraham Lincoln's campaign manager For re-election And one of the founders of the Republican Party It is vitally important to remember in this context That the Republican Party of Abraham Lincoln

Had nothing whatever to do with the Republican Party Of John Boehner Was there much of a distinction Between editorial and news at the time? No. In fact, the editor of the newspaper ran the editorial page and the news pages. But there was an editorial page, but it just wasn't – Correct.

It was run by the same person who ran the news pages. Gotcha. And that was a common thing. And it's not entirely uncommon right now. I know people who are editorial page editors who report to the chief news executive of the organization that they're part of. And I think that that's the problem. And that's what we deal with here, is that at the New York Times, the most important thing is the news report just overwhelmingly.

It's our reason for existence. And that is run by Jill Abramson. Jill can't tell me what to editorialize about, doesn't know what I'm editorializing about, doesn't know what positions I'm taking, although a lot of the positions we take aren't big shockers if you've ever read the New York Times editorial page. And I don't give her advice on what news to cover, and neither one of us has any involvement in the other's personnel decisions. I don't go to their news meetings. The people from the newsroom

never come to editorial board meetings. We meet three times a week, the editorial board and the editorial writers. The people from the newsroom never come to editorial board meetings. We meet three times a week, the editorial board and the editorial writers. The people from the newsroom never attend those meetings, and they are not invited to be part of it. I view it as wildly inappropriate when people tell me what I should write on the editorial page, and it sometimes happens, and they all get the same response, which is, you know, my dear friend, or if they are my friend or my colleague, please don't ever ask me a question like this again. You know how this works. And then they stop immediately.

Everybody here knows what the deal is. And they don't lobby for positions. And I don't lobby for coverage. Let me ask you this. So here's some official language from, I guess, the Times website. The editorial department of the paper is completely separate from the news operations. And Mr. Rosenthal, it's you, answers directly to the publisher, Arthur Salzberger Jr. I'm just curious, do you ever wonder if that's a kind of, you know, quaint, old-fashioned demarcation that has maybe outlived its time?

In other words, that it's impossible now, the way that news travels, the way that news is consumed, for the reading public not to conflate the opinion report of a paper like yours with the news report? I think it's very hard for them not to. For people who grew up, for example, in Europe, it's almost impossible because they don't really have, well, they do, but there's a lot of European newspapers where they co-mingle. And there is something that has happened. Everything we've been talking about is about institutional separation and bureaucratic separation and about separating the stuff that my people write from the stuff that Jill's people write.

This does not guarantee the separation of opinion from news. news and what's happened over the last decade or so to a great to a great extent and constantly increasing extent and a lot of this is because of what's going on online and we can talk about that if you like but uh is is the seepage of opinion into news not from the editorial department into news but within the news departments of of, people – they are doing opinion. Including your own paper. Yes. So let me ask you this. Does it drive you crazy some days when you read the news report of your own paper, which is as good a news report as there is, and say, hey, that should be on my page, not on their page?

Do you see phrases? Do you see ideas? No, it's more like – and I think this is important and this when this comes to media bias i think that that by and large the news articles in the newspaper are very straightforward they are they they have no stake in the outcome that the writer does not believe that one way or another should be the way that the thing turned out certainly you do not have a situation where people are either twisting facts or leaving facts out to make something appear different than it really is in order to suit their ideology.

I think the language is pretty straightforward. I think there's more analysis in New York Times articles than in most – there used to be. Now everyone does it. What you see is actual – The implication being that analysis tends to hew a little closer to opinion than straight news. It hews closer to conclusion.

I mean if you've ever been a foreign correspondent like I was and I was a foreign editor for a while, foreign correspondents need to – they need to do reporting that reaches a conclusion. A foreign correspondent in Afghanistan who is not telling you which side is winning the war is not doing his job, and that's a conclusion. And it's not that I want this side to win the war. And also a little bit in political reporting, you have to reach conclusions about how candidates are doing. And if a candidate is losing an election, you have to say so. There is a kind of, I think, common analog.

I hope I'm not overstating it by saying that it's common, that Fox News is to the right what the New York Times is to the left. I'm guessing that you would see that as a false equivalency on a lot of levels. Tell me if I'm right. I think there's a word I want to use here, but even on public radio. We bleep so much on this show. Well, it begins with bull and ends in it, and you can figure out what comes in between. I think it's absolute pernicious nonsense. I think that there – I've been at this newspaper a long time.

I've been in a lot of newspapers. Fox News presents the news in a way that is deliberately skewed to promote political causes and the New York Times simply does not. We make mistakes. We don't achieve perfect balance. There is no such thing as perfect balance because there is such a thing as truth. Talk about you for a minute growing up

with New York Times, Inc. kind of in your bloodstream. So your dad was the editor of this paper for nearly 20 years, I believe, which were very important years for this newspaper in a lot of ways. They were formative years. Talk for a minute, though, about,

okay, so you are the man who runs the most important opinion section in newspapers. And you are the son of the man who ran the news report here for nearly 20 years. So just talk a minute about that household and what position The New York Times played in the orbit of that household. Well, it's the other way around. It was the household in the orbit of the New York Times. The New York Times, my dad was a man of his generation. He was born in 1922.

For him, the New York Times was the beginning and the middle and the end of his existence, and he was dedicated to it entirely. He believed, you know what it says on his gravestone, right? He kept the paper straight. I have a thing on my desk, which most people don't know what it is, but it's that reverse lead type. And he hired a reporter from the Philadelphia Inquirer, discovered after he hired her that she'd been sleeping with one of her sources at the Philadelphia Inquirer and fired her, even though it had nothing to do with the New York Times. And someone said to him, why do you care who she sleeps with? And he said rather memorably, and it's there on my, I don't care if you f*** an elephant just so long as you don't cover the circus.

And that was my dad. I really truly believed that you could do that to an elephant and he'd be okay with it, but no circus coverage. And, you know, but that is the, no, seriously, I mean, it sounds goofy. No, but that's the core. But it's profound. It's the core of journalistic principle. And by the way, journalistic principle applies to what I do too.

We don't lie. And we don't work for Barack Obama. We're not members of anybody's team. It's journalism. We're supposed to be writing our opinions about stuff as we truly see it. A great editorial is a strong position, firmly held, quickly and cleanly expressed based on actual reporting. You know what I learned today? I learned that media bias is probably an argument that will never go away.

We put out a podcast not long ago about how people choose to believe what we believe, about everything from the risk of global warming to whether we've ever been visited by UFOs. Now, it turns out that a lot of us unknowingly bend our beliefs to fit our political or social or family circles. We carry around all sorts of personal biases that we simply don't see. The best description I've ever heard of this comes from Danny Kahneman, the Nobel-winning psychologist and author of Thinking Fast and Slow. He calls it being blind to our blindness. Well, there's a lot of psychological research that points in that general direction.

We think other people are biased and we don't feel that we are biased. We feel that our opinions are right because our intuitions tend to land on one coherent interpretation of the world. And that looks almost like a sure thing. And so we tend to be overconfident in whatever we believe that's generally true. And, you know, we find it much easier to find the errors in other people. So we'll continue to point fingers at the other side while declaring ourselves blameless. That seems to be the nature of the human beast. But more troubling, at least to me, is that we seem to like it like that. One reason we love to argue about something like media

bias is that after all these millennia, we still thrive on tribalism. We still love to divide ourselves into us and them, left and right, liberal and conservative, whatever you want to call it. We find the tribe where we fit in and we rush over to join and rally around its flag and immediately start tossing grenades at the idiots who are flying the other flag. Now, personally, I find this instinct a little bizarre, this instinct to herd ourselves into one of two major groups. Why should we want to define ourselves as right or left? If I'm a hardcore environmentalist, that means I have to want higher taxes too? If I'm pro-death penalty, I can't also be pro-choice? We love to complain about partisanship in Washington and in the media,

but could it be that we get the partisanship we want, the partisanship we deserve? Here's my advice. If you want an enemy to root against, watch more sports. That's what sport is good for. It's a proxy for war, for violence, for tribalism. And then when the game's over, when you're ready for some nuance, come back to the real world. And only then, if you think you're calm enough to handle it, pick up a newspaper. Freakonomics Radio is produced by WNYC, APM, American Public Media, and Dubner Productions. This episode was produced by Susie Lechtenberg with help from Jacob Berman. Our staff includes Catherine Wells, Diana Nguyen, Beret Lam, and Chris Bannon. David Herman is our engineer, Jacob Bastian is our intern, and our executive producer is Colin Campbell. If you want more Freakonomics Radio, subscribe to our free podcast on iTunes

and go to Freakonomics.com, where you'll find lots of radio, a blog, the books, and more.