« Why I'm skeptical of Nate Silver's pollster ratings | Main | Questioning the popularity of fact-checking »

June 23, 2010


The fact that Obama's results were close to the leading models may mean race was not a significant factor. Or, it may mean that several of the factors not in the models were significant, but canceled each other out.

ISTM that there were 3 potentially large factors that (I assume) were not reflected in the leading models:
1. Race
2. Obama's lack of experience
3. Biased media coverage

IMHO #3 cannot be separated from #1. Obama's media coverage was extraordinarily favorable because he was black.* Race hurt him because of racism, but it also helped him get more favorable coverage.

I would conclude that (Race + Media Bias) helped Obama by about the same amount that his lack of experience hurt him. I know of no way to estimate how small or large this amount might be. Nor can I estimate what the effect of race would have been had the media coverage been balanced.

*More precisely, Obama got extraordinarily favorable coverage because he's a black Democrat. Clarence Thomas has not gotten favorable media coverage.

For a recent example of media bias in favor of Obama, see Tom Maguire's evisceration of Reuters' curious perception of the reaction to Obama's "whose ass to kick" comment.

The comments to this entry are closed.