The New Hampshire results are in, and the big loser is the American pollster. The polls consistently predicted an Obama victory over Clinton by 8 points. Clinton won by 3. How do you explain all the major polls being wrong by 11 points?
I was just watching Hardball with Chris Matthews and the consensus opinion was that it was the Bradley effect. For those of you that don’t know, the Bradley effect is named for Tom Bradley, African-American former Mayor of Los Angeles. In the 1982 race for California governor, a large number of white voters who did not want to vote for Bradley because he was black, were embarrassed to tell pollsters for fear that they would be thought of as racist.
The round table discussion on Hardball focused on why the polls were wrong about Obama and how race might be involved. You’d think that people who make the Big Pundit Money would actually look at the numbers. Evidently they were too busy filing their expense reports, so how about we take a peek? Below are the RealClearPolitics.com final poll average, the actual vote and the difference between the two.
______________RCP Avg.______Votes______Poll Error
Obama________38.3%_________36.4%_____-1.9%
Clinton________30.0%_________39.0%______9.0%
Edwards_______18.3%_________16.9%_____-1.4%
Richardson______5.7%__________4.6%_____-1.1%
It doesn’t take Karl Rove to see that the polls were reasonably accurate for Obama, Edwards and Richardson. To a reasonable degree of accuracy, voters who told pollsters they were going to vote for the African-American, the Hispanic or the Rich White Male, were telling the truth. The error was in the polling for Hillary Clinton.
There was no Bradley effect. Lily white New Hampshire is no more racist than lily white Iowa was when it gave Barack Obama a huge victory. But I suppose that when your liberal angst compels you to believe that America is still a racist country and that nothing has changed in the past quarter century, looking objectively at the numbers may be a bit much to ask.
No comments:
Post a Comment