writing bio research bagatelle home


On the Bias

Geoffrey Nunberg

Commentary broadcast on "Fresh Air," March 19, 2002
revised, 3/22/02

Bernard Goldberg is hardly the first person to claim that the media have a liberal bias, and his Bias is far from the best-written or best-argued book to try to make that point. Even so, it has climbed to the top of the New York Times bestseller list, maybe because Goldberg is himself a CBS insider with lots of tell-all tidbits to offer about the likes of Dan Rather and Bob Schieffer.

For the most part, Goldberg's book is a farrago of anecdotes, hearsay, and unsupported generalizations. But at one point he strays into territory that can actually be put to a test. That's when he claims that the media "pointedly identify conservative politicians as conservatives," but rarely use the word "liberal" to describe liberals. As Goldberg explains the difference: "In the world of the Jennings and Brokaws and Rathers, conservatives are out of the mainstream and have to be identified. Liberals, on the other hand, are the mainstream and don't have to be identified."

That basic premise is sound enough -- that the media mention things that they see as being out of the mainstream more often than they mention things that they see as in it. If a major company names a seven-foot-tall Hare Krishna from Tonga as its CEO, those attributes are more likely to show up in the story than if the new chief is a 5'10" Methodist from Ohio.

But does that difference show up in the way the media deal with liberals and conservatives, too? TV newscasts aren't easy to check, and Goldberg doesn't offer any research to back up his claim. But Goldberg and the other critics of media bias also make their charges about the language of the press, which is available online. So I went to a big online database and did a search on the articles from about 30 major newspapers, including The New York Times , the LA Times, the Washington Post, The Boston Globe , the Miami Herald, and the San Francisco Chronicle .

For purposes of comparison, I took the names of ten well-known politicans, five liberals and five conservatives. On the liberal side were Senators Boxer, Wellstone, Harkin, and Kennedy, and Representative Barney Frank. On the conservative side were Senators Lott and Helms, John Ashcroft, and Representatives Dick Armey and Tom Delay. Then I looked to see how often each of those names occurred within seven words of liberal or conservative , whichever was appropriate. Of course some of those hits involved extraneous noise, say when the word liberal just happens to find itself near Barbara Boxer's name with no real connection between the two. But when I checked a sample of the results by hand, it turned out that more than 85 percent of them did in fact involve the assignment of a political point of view, with phrases like "Paul Wellstone, the liberal senator," or "Senate conservatives like Jesse Helms." And with a sample of more than 100,000 references to the names on the list, the results were statistically sound.

In fact, I did find a big disparity in the way the press labels liberals and conservatives, but not in the direction that Goldberg claims. On the contrary: the average liberal legislator has a thirty percent greater likelyhood of being identified with a partisan label than the average conservative does. The press describes Barney Frank as a liberal two-and-a-half times as frequently as it describes Dick Armey as a conservative. It gives Barbara Boxer a partisan label almost twice as often as it gives one to Trent Lott. And while it isn't surprising that the press applies the label conservative to Jesse Helms more often than to any other Republican in the group, it describes Paul Wellstone as a liberal twenty percent more frequently than that.

At first I wondered whether I had inadvertantly included a bunch of conservative newspapers in my sample. So I did the same search in just three papers that are routinely accused of having a liberal bias, The New York Times , the Washington Post, and the Los Angeles Times . Interestingly, those papers tend to use labels of both sorts slightly less than the other papers do. But even there, the liberals get partisan labels thirty percent more often than conservatives do, the same proportion as in the press at large.

The tendency isn't limited to politicians, either. For example, Goldberg writes that "it's not unusual to identify certain actors, like Tom Selleck or Bruce Willis, as conservatives. But Barbra Streisand or Rob Reiner. . . are just Barbra Streisand and Rob Reiner." But Goldberg's dead wrong there, too. The press gives partisan labels to Streisand and Reiner almost five times as frequently as it does to Selleck and Willis. For that matter, Warren Beatty gets a partisan label twice as often as Arnold Schwarzenegger, and Norman Lear gets one more frequently than Charlton Heston does.

It's the same with other figures. Goldberg claims that Robert Bork is always called a conservative whereas Laurence Tribe is just identified as a Harvard law professor, but when you look at the data, it turns out that the two are labeled with almost exactly the same frequency. And the columnist Michael Kinsley gets a partisan label slightly more often than George Will does -- and more often than Jerry Falwell.

There are some exceptions here. Americans for Democratic Action gets a label slightly less frequently than the Heritage Foundation, though both are labeled very often. In fact, the ADA gets a label more often than the Young Americans for Freedom does, and almost three times as often as conservative groups like the Cato Institute or the National Association of Scholars. And the overall tendency is overwhelming: liberals are singled out for their views more often than conservatives are.

I found the results surprising, not because I assumed the press had a liberal bias, but because liberal has become such a problemmatic word that nobody seems to want to use it. Since the Reagan era, the right has gone after it as "the L-word," to the point where a lot of politicians are nervous about owning up to being liberals. And people on the genuine left have always been suspicious of the term, preferring to think of themselves as progressives. But nobody every talks about "the C-word," and people on the right are always happy to call themselves conservatives.

I'd have figured that all that would make the press, too, a bit reluctant to use the "liberal" label. But it turns out the newspapers label liberals much more readily than they do conservatives. Of course it's possible that things work differently on TV newscasts. But that's pretty unlikely, unless you're willing to assume that the language we hear on CBS and ABC has a much more liberal slant than what we get in the Washington Post and The New York Times, a position that not even critics like Goldberg have tried to argue. If there is a bias here, in fact, the data suggest that it goes the other way -- that the media consider liberals to be farther from the mainstream than conservatives are. Or maybe sometimes it's a case of the press bending over backwards to avoid the charge of bias. The one thing that's certain is that there's another bias operating here, as well -- the one that leads media critics to hear what they want to hear.

[1] Note: After this piece aired, I had a couple of emails from people who suggested that another choice of legislators might have made the results come out differently. That's true, but a look at the accompanying table will show that the tendency to label liberals more than conservatives was no less marked for other figures.


Results of Study









Copyright 2002 Geoffrey Nunberg All rights reserved.