Stupid Media Study Tricks

Liberal media bias strikes most people as a subject that needs no elaboration. You either believe it's a real problem, or you don't. To conservatives, it's so obvious that it should be beyond debate. To liberals, what's obvious is that right-wingers are "working the refs" to force nervous journalists to bend over backwards toward a conservative bias.

But there are occasions when liberals try to argue that the sky is green, and the moon is made of cheese - that is, that our national media favor conservatives and Republicans. It seems absurd, but conservatives need to pay attention. Democrats recently protested that the cable news networks are favoring the Republicans by their live broadcasts during the day. They show briefings at the White House or the State Department or the Pentagon or presidential speeches, as if these are fawning press availabilities for the Bush White House. Watching Ari Fleischer get pummeled about Enron is not a GOP infomercial. If you want to see reporters eating out of someone's hand, tune into those Tom Daschle briefings. If this man ever got a tough question he'd probably faint from the shock.

Some bold liberals would even have you believe that former CBS stalwart Bernard Goldberg's bestselling expose "Bias" is only proving that the media's bias isn't liberal, but conservative.

In The American Prospect, a journal for lively liberal imaginations, linguist Geoffrey Nunberg claims that Goldberg is goofy to suggest that the media feel "conservatives are out of the mainstream and need to be identified. Liberals, on the other hand, are the mainstream and don't need to be identified."

To Nunberg, the book is simply "a farrago of anecdotes, hearsay, and unsupported generalizations." But it is he who has mastered this art. He claims that "however you cherry-pick the groups, there's no way to make the survey come out as Goldberg claims it should, where conservatives are systematically labeled more than liberals are." Clearly, Nunberg has ignored a decade of studies by the Media Research Center-hard numbers, no generalizations-showing that while conservative groups are often ideologically identified 33 percent or more of the time in news stories, liberal groups are labeled less than two percent of the time.

Here's a little list of some of the multi-year labeling percentages found in newspaper coverage. On the environment, newspaper stories on the Competitive Enterprise Institute included a conservative label in 28 percent of its 29 newspaper stories, compared to zero percent for Greenpeace (in 178 stories) and less than one percent for the Sierra Club (in 325 stories). In stories on retirement, they hunted for a lonely liberal label for AARP (zero percent) and Families USA (one percent).

On the social issues, the reporters really get in a labeling - or warning - mood, if you look at stories featuring conservative groups like Eagle Forum (75 percent), Concerned Women for America (71 percent), or the Family Research Council (63 percent). The exception is the National Right to Life Committee (6 percent). But try now their counterparts: NOW (2 percent), NARAL (5 percent), the Feminist Majority Foundation (2.6 percent), or Planned Parenthood (2.2 percent). If you're interested in the gay left, look at coverage of the Human Rights Campaign (1.3 percent), the National Gay and Lesbian Task Force (2.8 percent), or the Gay and Lesbian Alliance Against Defamation (zero labeling "defamation" there).

These studies were meticulous work. MRC analysts sat for weeks with enormous stacks of Nexis database printouts, reading page after page and checking each news story and label in context, weeding out all non-news examples, like opinion pieces. If "conservative" was used to describe a suit instead of a group, it didn't count.

By contrast, Nunberg did a "study" of ten legislators covered by 30 newspapers and found that "the average liberal legislator has a better than 30 percent greater likelihood of being given a political label than the average conservative does." CNN's Jeff Greenfield just put that number on the air, but didn't note Nunberg found liberal politicians were tagged in 4.8 percent of stories to conservatives' 3.6 percent. There's your 30 percent. Big, big deal.

Nunberg merely threw in his magic words - "Barbara Boxer" within seven words of "liberal" - and accepted the numbers his Dialog database spat out. He claims that out of thousands of stories in 30 newspapers, he "examined 100 citations by hand" - a tiny fraction of the mountain of content he's claiming. But don't worry, he says, those 100 citations showed only 12 percent of his citations turned out to be mistakes.

You wouldn't want to try this method out at a meat inspection, and it's not a very convincing media inspection technique, either. Please ignore the man behind the TV screen telling you he's really the wizard of conservative media bias.

More examples of the media's labeling disparity as documented by the MRC