Speculation on root causes for the opinion about media bias

I commented in Are Washington Post and New York Times commenters becoming more conservative? on what appears to me to be an increasing frequency with which commenters on articles and opinions are expressing essentially conservative criticisms of the unstated biases and assumptions underpinning much reporting.

The catalyst was an original article by Jodi Kantor, A Brand New World In Which Men Ruled.

In writing up that post, a comment from Kantor opened up a new train of thought. It is a common criticism on the right that the Media is dominated by a progressive or left leaning worldview. While the significance of that progressive worldview in the industry might be disputed, its existence is largely accepted as true. Whether measured by party donations or political registrations or other indicative measures, participants in the mainstream media tend to skew left by a factor of 4:1 or greater.

The general assumption among conservatives is that progressives infest mainstream media intentionally. The suspicion is that those of a progressive intent see the media as a means to advance their worldview and that is why they enter the industry. Surely that is true for some, but for how many?

What if conservatives have the flow of causation backwards?

There are two steps to an alternate view, sparked by a comment by Kantor in her article, The Story Behind the Story which provides background to her reporting.

I also had to avoid the main pitfall of writing about gender: that readers are going to take one look at the article and think, “I’ve read this before” or “I know how this ends.”

At one level, that's just the age old dictum that "Man Bites Dog" is news whereas "Dog Bites Man" is not. But it introduces the idea that there may be a different flow of causation than conservatives believe.

I suspect conservatives see bias in the mainstream media arising from a left leaning/progressive world view amplified by two common errors/biases/fallacies, the Observer Expectancy Effect (a common error where a researcher expects a given result and therefore unconsciously manipulates an experiment or misinterprets data in order to find it) and Confirmation Bias (allowing your existing opinions, to influence the way that you interpret information.) That is perfectly logical and there is plenty of evidence to support it. But perhaps it isn't really the whole story. Perhaps there are other factors at work. Such as the ones implied by Kantor. What her comment suggests is that there is a systemic pressure to produce differentiated, novel reporting where the emphasis is on the unusual Man Bites Dog story which, being novel and unexpected, is preferred over the more factually representative story of Dog Bites Man.

Kantor's comment suggests there are a couple of different biases/errors/fallacies that might be in play, the Publication Bias (positive findings are more likely to be published than negative findings, which can skew the results that we see) and the Appeal to Novelty fallacy (argues without evidence that something new will be better than something older.)

If those two biases are indeed materially influential, they would lead to a similar outcome as what conservatives see as a bias against them. Conservatives, very generally, only want to see change which is justified on an empirical basis. They want to know why the change is warranted based on facts and figures and values. If there is no empirical case to be made for a change, they are generally comfortable leaving things the way they are.

Progressives and Liberals, with a similar broad brush, tend to be much more rooted in an aspirational ideal, a utopia. They tend to focus less on the improvement over the past (as conservatives often do) and instead often focus on the shortfall from the ideal.

Accepting these two broad caricatures, and then accepting that Publication Bias and Appeal to Novelty are both real media influences, then which side is likely to see more news that appeals to them, independent of the worldview of the journalists and editors? I suspect that publication bias and appeal to novelty will always generate news reporting that is more attuned to the progressive reader than to the conservative reader. Publication Bias and Appeal to Novelty will ensure, for example, that "20 Polling Stations Have Voting Problems" is the election story that gets written (appealing to a progressive perspective where the goal is 100% perfection) rather than the more conservative perspective which would generate the equally accurate headline "99% of Polling Stations Free of Problems."

I think there is one further journalistic commercial truism that influences the outcome. "If it bleeds it leads" - dramatic stories take precedence over mundane stories. But what happens in an environment where all forms of violence are declining whether volitional (such as murders and acts of terrorism) or accidental (earthquakes, droughts, etc.). If the commercial pressure is to produce unusual stories of violence and there are fewer violent stories available and they are more mundane in their details, then the pressure is to go further and further out on a limb. Trying to make the trivial seem important, the quotidian to seem exceptional (microaggressions), the routine to seem unexpected (choices have consequences) all lead to articles which I think are probably more appealing to the progressive mindset than to the conservative mindset.

Sure, the mainstream media might have an ideological predisposition for the exceptionally loud progressives such as Piers Morgan, Rachel Maddow, Amanda Marcotte and Keith Olbermann. But what if the commercial pressure for readable stories also produces a slant to the media content that is preferable to the progressive mind than the conservative. You end up with conservatives as seeing the mainstream media as essentially a mouthpiece of leftist ideology or as "Democratic operatives with a by-line" as one critic puts it. But this speculative analysis suggests that the imbalance is less a product of deliberate propaganda and more an outcome of commercial realities.