Original Sources vs Analysis

A little while ago, Brad Wardell wrote about a difference between Left-Wing and Right-Wing zealots. Basically, liberals tend to rely on third-party analysis of data, while conservatives tend to dig into the data themselves and come to their own conclusions. Each group will sometimes get into trouble sometimes because of this:

Bear with a generalization for a moment: people who are left of center politically tend not to be quite as analytical as the general population. You don’t find too many engineers, for example, arguing for liberal causes. And you don’t find too many artists arguing for conservative causes. There are exceptions of course but as a generalization, I think you’d agree this is true.

This creates an interesting set of diverse behaviors. Liberals tend to rely very heavily on third party data analysis. In other words, they don’t tend to look at the actual data and then form their own conclusion. Instead, they just repeat the analysis of someone else. …

But let’s not let the right wingers get off the hook either. If we’re going to offend people, let’s be equal opportunity offenders. There’s a reason that the term “Right wing kook” exists.

Conservatives, particularly very conservative people, are much more inclined to not trust the “liberal media establishment”. So often they’ll dig into data that is really out of their league to understand. So they’ll look at the data and come up with bizarre conclusions.

I tend to dislike generalizations like this, especially when they’re made about groups that are as vague and undefined as liberals and conservatives. However, this struck a chord with me as I could identify with the premise – I’ve debated a lot online, and I’ve seen this sort of thing in action. It turns out that I’m more in line with the conservative side of things, preferring to look at the data myself rather than relying on some analysis of the data (though I try to avoid referencing a “liberal media” and I don’t generally like conspiracies). Still, I’m hesitant to buy into such a broad generalization.

In any case, it is interesting to note that the recent 60 Minutes story that was thoroughly debunked* on the internet within 12 hours of airing provides us with a good example of Wardell’s theory in action.

CBS aired this story which questioned President Bush’s National Guard service on Wednesday night, citing newly discovered memos written by Lieutenant Colonel Jerry Killian (Bush’s commanding officer). The story was picked up by all the major media outlets, which ran the story the next day. Such stories were immediatately trumpeted by liberals in discussion forums and blogs everywhere. During that same period, some intrepid bloggers were piecing together evidence that the newly discovered memos were forged. It didn’t take long for the evidence to mount, and soon other independent threads of investigation were coming to the same conclusion and the media picked up this side of the story. Experts were consulted, family members spoke out, and even CBS sources came out against the documents. All along, some liberals denied the scandal, pointing to CBS’s continued insistence that the documents were genuine as proof (i.e. not actually answering the questions about the documents’ authenticity, but pointing to CBS’s third party analysis of the documents).

This is just a brief summary of the story, and others have much more comprehensive overviews*. What I found interesting, though, was that this is a textbook example of liberals relying on analysis, and conservatives digging into the data and coming to their own conclusions. I’m still not sure I subscribe to Wardell’s theory, but I found it interesting that this scandal clearly demonstrates his point.

Update: In the comments, Spencer notes that Wardell’s theory “seems like a conclusion one would make as a result of confirmation bias. This person ‘notices’ a slight trend in the way liberals and conservatives treat information, and then only notices the evidence that supports this trend, making it seem like a stronger and stronger hypothesis.” That makes a lot of sense to me, and though I still find the theory interesting, I don’t think it holds up in the end. There is something about the propensity to label people that is beginning to bother me more and more, and the division between liberals and conservatives is a prime example. On the one hand, it can be useful to think about such divisions, but it is also easy to get carried away.

* I’m linking to one of the original posts at Power Line because it is large and provides a good summary of the complaints regarding the documents. There have been many posts on many different weblogs that have continued the debate, ranging from the humorous to the serious. Here are some other blogs that might help you get a better grasp of the story:

• Power Line continues to be the leader in coverage of what is being called Ra^thergate.
• Donald Sensing’s Analysis goes at the issue from a different angle than pointing out typographical errors, instead focusing on the military content and practices that the memos got wrong.
• Glenn Reynolds has a number of good posts on the subject. (oh, and another good one)
• Belmont Club has some good analysis of the problem, and of CBS’s response.

This story was broken by the collective intelligence of the internet, with a bunch of different bloggers cooperating to build up a pile of evidence. In one sense it is amazing, in another, it’s annoying because there is no one place to link to that collects all of the commentary on this story. The original Power Line post is great because it collects so much of the analysis in one place, but since that post things have become more diffuse.