Monday, December 19, 2005

Media Bias Study

There's a serious new statistical study of media bias out to chew on. I'm trying to gether up my own observations on that ever-hot topic, and I may eventually get them together in a post. But here's my take on the report.

The analysts took great pains to avoid bias or the appearance of it, going so far as to refuse outside funding and vetting their research staff to insure a balance of blue and red. The results of their labor confirm what a lot of us have felt: "Almost all major media outlets tilt left."

"I suspected that many media outlets would tilt to the left because surveys have shown that reporters tend to vote more Democrat than Republican," said Tim Groseclose, a UCLA political scientist and the study's lead author. "But I was surprised at just how pronounced the distinctions are."

"Overall, the major media outlets are quite moderate compared to members of Congress, but even so, there is a quantifiable and significant bias in that nearly all of them lean to the left," said co‑author Jeffrey Milyo, University of Missouri economist and public policy scholar.

The report is chock full of juicy tidbits and some surprises:

  • Of the 20 major media outlets studied, 18 scored left of center, with CBS' "Evening News," The New York Times and the Los Angeles Times ranking second, third and fourth most liberal behind the news pages of The Wall Street Journal.

  • Only Fox News' "Special Report With Brit Hume" and The Washington Times scored right of the average U.S. voter.

  • The most centrist outlet proved to be the "NewsHour With Jim Lehrer." CNN's "NewsNight With Aaron Brown" and ABC's "Good Morning America" were a close second and third.

  • "Special Report With Brit Hume" on Fox News, which often is cited by liberals as an egregious example of a right-wing outlet. While this news program proved to be right of center, the study found ABC's "World News Tonight" and NBC's "Nightly News" to be left of center. All three outlets were approximately equidistant from the center, the report found.

  • National Public Radio, often cited by conservatives as an egregious example of a liberal news outlet, according to the UCLA-University of Missouri study ranked only eighth most liberal of the 20 that the study examined. "By our estimate, NPR hardly differs from the average mainstream news outlet," Groseclose said.

  • The Wall Street Journal's news pages are liberal, even more liberal than The New York Times.

I have some objections to the methodology of this study. The challenge in any such survey is to quantify "liberal" and "conservative" into measurable commodities. The only place this has been done is in political institutions. U.S. Congressmen routinely are sliced and diced by special interest groups which rank them in point systems on the scale of "liberal-to-conservative." The researchers here chose one of the popular scorecards as a benchmark:

Americans for Democratic Action (ADA) tracks the percentage of times that each lawmaker votes on the liberal side of an issue. Based on these votes, the ADA assigns a numerical score to each lawmaker, where "100" is the most liberal and "0" is the most conservative. After adjustments to compensate for disproportionate representation that the Senate gives to low‑population states and the lack of representation for the District of Columbia, the average ADA score in Congress (50.1) was assumed to represent the political position of the average U.S. voter.

The next problem is, how do you turn the amorphous mass of news coverage into something that can be compared to the U.S. Congress?

Groseclose and Milyo tried to do it by going through 10 years of media transcripts and noting who was being quoted for context. They "tallied the number of times each media outlet referred to think tanks and policy groups, such as the left-leaning NAACP or the right-leaning Heritage Foundation." Then they went through the U.S. lawmakers' speeches and saw how often they cited the same organizations. "If a media outlet displayed a citation pattern similar to that of a lawmaker, then Groseclose and Milyo's method assigned both a similar ADA score."

It's possible to "cite" something in a negative as well as a positive sense. You can cite the Heritage Foundation in an article or broadcast and surround it by such other citations that the reader or viewer will come away thinking the Heritage Foundation is full of crap. But it's also possible to do this on the floor of Congress. So is it done more often in one place than the other? I don't know if the survey considers this.

The ADA scorecard sets out to measure "liberalism." The media survey seems to take the position that measuring liberalism is de facto measuring conservatism (the absence of liberalism) the way measuring heat is, by implication to measure cold (lower heat = more cold). But that is not quite true in this case. American politics tends to be bipolar and linear, but not entirely so. A Howard Dean compared to a John McCain compared to a Zell Miller compared to a Joe Lieberman compared to an Arlen Specter might break a simple up-and-down liberal-to-conservative model.

Based on the description of the methodology, the media study counts the number of times an institution is cited, but it doesn't appear to account for the amount of time or space given to each institution in the coverage.

Finally, NAACP is not the exact equivalent of a conservative think tank. The right's think tank culture grew out if its sense of exclusion from the universities. The think tanks were meant to be incubators of conservative ideas and supporters of conservative causes the way academe had become the incubator for the left. To make a more legitimate comparison, it seems to me, you'd have to count professors, too.

The NAACP is left-leaning, it is true. But its causes are not primarily identified with liberals to the exclusion of conservatives. Same with the ACLU. It's possible that some of the citations used to suggest "liberal" leaning don't really indicate that. It's also possible that a vast amount of liberal sentiment coming out of academic citations was entirely overlooked in this report. The effect of this would be actually to undercount the liberalism of the media, which is sort of scary, given the results the report did come up with.