« New at CJR: Media failures in Reid/Romney coverage | Main | New at CJR: In defense of convention coverage »

August 17, 2012



I enjoyed the Perspectives piece. Well done.


1. Brendan's paper seems to discuss none of the 5 questions set out by Henry Farrell.

2. Brendan's point, that GM's research has some questionable assumptions and possible statistical flaws, would be true of pretty much every political science study that I'm familiar with. E.g., consider the studies, of which Brendan approves, showing that various economic factors determine Presidential elections. These studies draw causitive conclusions from samples that are not random and which are much, much too small to justify causitive conclusions, according to normal scientific standards.

However, I agree with Brendan that these studies are better than the usual POOMA approach. Any effort to quantify political factors will necessarily need weaker assumptions than those used in the hard sciences.

The point is, with the election models, Brendan sees the glass as half full. With GM, he sees it as half empty.

3. Brendan himself sometimes states guesses as facts in a way that he criticizes in GM. E.g., in the portion quoted above, Brendan writes, "These differences [in political slant] are often significant and appear to be driven in large part by economic factors such as consumer demand and media competition." (emphasis added)

Brendan offers no proof that these differences are driven by economic factors. IMHO these differences exist despite economic factors. The rapid success of Fox News shows that there's a big market for conservative (or less liberal) news on TV. Yet, no other cable station has tried to challenge their large marketing niche.

Here's another example of a guess atated as a fact: "As a result[of media norms], previous studies of partisan bias in reporting on presidential elections have generally not found consistent results." (emphasis added)

Are media norms really the reason why many previous studies failed to find bias? Or, is because those previous studies were done by people whose liberism blinded them to media bias? (And, would Brendan fall into this group?)

The comments to this entry are closed.