It’s said that Tom Bradley lost a race for California governor in ‘82 even though polls showed him ahead because secret racists told pollsters they would vote for Bradley, a black man, and didn’t. This has been called the Bradley Effect. However, more recent examination of the data indicates this probably didn’t happen, that the problem was that the Bradley polling data was flawed. Moreover, races in 2006 with black and white candidates showed no Bradley Effect.

The overall accuracy of the polling data from the 2006 elections was cited, both by those who argue that the Bradley effect has diminished in American politics, and those who doubt its existence in the first place

Also, pollsters tend to focus on likely voters, those who have voted regularly in the past. Thus, they are not tracking what undoubtedly will be a ginormous turnout by black voters in November voting for Obama, some of whom might not have voted much recently.

2 Comments »