A Mismeasure of Media Bias
Last week Media Matters published a critique of a paper titled "A Measure of Media Bias", by Tim Groseclose and Jeff Milyo (G-M for short). A number of people picked up this critique: Markos (Daily Kos - lost the link), Paul Rosenberg (MyDD), Gary Boatwright (Seeing the Forest), Brendan Nyhan, to name a few.
I just want to mention that I already published an even more detailed analysis and debunking of the paper early this year. Key excerpts were also published here at The Left Coaster as part of my series on How The Liberal Media Myth is Created.
Media Matters adds one aspect that I missed in my earlier debunking - that the authors also happen to be former members of conservative think-tanks (pardon the oxymoron). (I guess I should have checked that out but since I was trying to focus solely on the merits of the paper I had given them some benefit of doubt at the time).
The arguments I have read against the G-M paper this past week do address some of the important problems with the paper (which I discussed earlier). I want to add that my critique also addressed the fundamental problems with its methodology of using weighted-average ADA scores (not to mention its hopelessly nonsensical definition of "media bias") . Take a look if you are interested.