Skip to content

Asymmetry in Political Bias

Tyler Cowen points to an article by Riccardo Puglisi, who writes:

Controlling for the activity of the incumbent president and the U.S. Congress across issues, I find that during a presidential campaign, The New York Times gives more emphasis to topics on which the Democratic party is perceived as more competent (civil rights, health care, labor and social welfare) when the incumbent president is a Republican. This is consistent with the hypothesis that The New York Times has a Democratic partisanship, with some “anti-incumbent” aspects . . . consistent with The New York Times departing from demand-driven news coverage.

I haven’t read the article in the question but the claim seems plausible to me. I’ve often thought there is an asymmetry in media bias, with Democratic reporters–a survey a few years ago found that twice as many journalists identify as Democrats than as Republicans–biasing their reporting by choosing which topics to focus on, and Republican news organizations (notably Fox News and other Murdoch organizations) biasing in the other direction by flat-out attacks.

I’ve never been clear on which sort of bias is more effective. On one hand, Fox can create a media buzz out of nothing at all; on the other hand, perhaps there’s something more insidious about objective news organizations indirectly creating bias by their choice of what to report.

But I’ve long thought that this asymmetry should inform how media bias is studied. It can’t be a simple matter of counting stories or references to experts and saying that Fox is more biased or the Washington Post is more biases or whatever. Some of the previous studies in this area are interesting but to me don’t get at either of the fundamental sorts of bias mentioned above. You have to look for bias in different ways to capture these multiple dimensions. Based on the abstract quoted above, Puglisi may be on to something, maybe this could be a useful start to getting to the big picture.


  1. Dubi says:

    Wait, if NYT gives more room to Democrat owned topics when a Republican is incumbent, doesn't that mean they give proportionately more room to Republican owned topics when a Democrat is incumbent? Where's the bias in that? If they ALWAYS gave Democrat-owned topics more coverage, that would be bias.

  2. Phil says:

    I can't read the article without filling out some forms that I don't want to fill out, but my initial reaction is more or less in line with Dubi's comment. If the NYT (or anyone else) always focuses on issues in which the incumbent party is perceived as deficient, that's not partisan bias, it's anti-incumbent bias or something.

    If the article finds that the NYT always focuses on perceived Democratic strengths, whether a Democrat or Republican is in office, _that_ might indicate partisan bias. But it might not. Health care has been a major issue for most of my adult life; it would hardly be "bias" if the NYT always talks a lot about health care in presidential campaigns. In fact, I find it very hard to make predictions about how subtle bias — as opposed to un-subtle bias such as printing falsehoods — would be manifested in the selection of news stories. Suppose a candidate makes health care a big issue of his presidential campaign. If a news outlet gives it a lot of emphasis — lots of big stories about the potential impact of the candidate's proposals, with in-depth looks at who will be helped or hurt by it — does that indicate bias for the candidate, or against the candidate? The answer might depend on whether the emphasis is on who will gain or on who will lose; it's very hard to define "unbiased" coverage in this context.

    Not having read the article, I can't say whether Puglisi addresses this point, which he very well might. Perhaps someone with access to the article can give us some insight.

  3. Bob Carpenter says:

    I've never undestood what people mean by "unbiased". Seriously. As David Byrne said succintly , "facts all come with points of view". There is no "objective" language, right down to the intonation and word choice level.

    Do we let everyone in the entire world weigh in with their point of view? Do we take surveys and balance number of column inches by number of believers?

    What makes no sense to me is that if you publish an article on topic X, you have to go out and find a point of view that's anti-X, so it looks "balanced". Of course, you can always find a crackpot to cite as an anti-X strawperson (the "person" was for you, Andrew).

    I find this in research, too. You have to make a half-hearted effort to compare whatever you do with some "opposing" point of view. My last conference submission was rejected for (among other things like being both too obvious and too hard to understand) for not providing a comparison with non-Bayesian techniques!

  4. But I've long thought that this asymmetry should inform how media bias is studied.

    You may not find it connected to claims of "bias" all that often, but this is a textbook example of agenda-setting, which is a major mass comm theory.

  5. Jonathan says:

    Who cares about bias? Isn't it all in expected mean squared error?

  6. J.J. Hayes says:

    "The New York Times gives more emphasis to topics on which the Democratic party is perceived as more competent (civil rights, health care, labor and social welfare) when the incumbent president is a Republican."

    The question for a news reader would be whether or not the facts are as reported. Do the facts as reported show that the Democrats are in fact more competent? If simply choosing a topic in which the Democrats are perceived as more competent constitutes a bias then it must be that the facts underlying the reporting support the perception. If the reported facts don't support the perception then the reporting isn't biased, unless its biased in the other direction.

    And what does "demand driven news coverage"? And how does one depart from it. Given the close split in this country aren't there a lot of people, and given that it is the NEW YORK Times, isn't it likely that there are a lot of readers who demand these topics be covered, and would read another paper if they were not?

  7. Steve Sailer says:

    The important thing is getting to decide what is respectable. All sorts of ideas that are, from the standpoint of Occam's Razor, superior to the conventional wisdom in accurate explanatory power are simply not respectable, and thus never get a hearing on either the NYT or Fox.

  8. Phil says:

    Bob, I agree with part (but only part) of what you say. I, too, am irritated by the American journalistic standard practice, which is not shared by all other countries, of always giving "both sides of the story" even when one side is right and the other side is wrong.

    However, your comment almost reads as if you don't agree that there is such a thing as an "unbiased" article. Actually I might agree with that in some limited; but there is definitely such a thing as a "biased" article, and some articles are more biased than others. Bias is hard to define, but like pornography I know it when I see it.

    I think bias is very hard to define and to quantify, but I agree that it exists.

  9. Bob Carpenter says:

    Right — I'm going all po-mo relativistic. I don't like the notion of bias, because it assumes there's an objective or unbiased way to say something.

    Language doesn't work like that. It's a fundamentally social activity, and every choice of what to say (and what not to say) and how to say it comes with social baggage. Lots of ways to say the same thing might be "true" (itself a dubious notion given the approximate nature of theories like Newton's laws and because of the way words in languages are defined analogically by example or approximately in terms of other words).

    Let's do a thought exercise. Let's say I'm writing for a newspaper and have just interviewed Andrew. Do I introduce him as "a person", "a man", "a father", "an Ivy-League academic", "a professional statistician", "in shirtsleeves and a tie" or not say anything at all? Do I continue with "Andrew", "Gelman", "Mr. Gelman", "Dr. Gelman" or "Prof. Gelman"?

  10. Phil says:

    Bob, I understand what you're saying. I just think it's a bit silly.

    I agree that it may be impossible to create a sentence that is truly "unbiased." But some are more biased than others.