If you have "no place to go," come here!

Claims of Bradley effect in NH 2008 bogus, says American Association for Public Opinion Research

You remember the "Bradley Effect," right? That's the hook that the Obama Movement used when they decided to smear the Clintons and their supporters as racists so they could win the next primary in SC. Well, as it turns out -- and I know this will surprise you -- those claims were bogus. CQ politics:

Remember all those polls that predicted Barack Obama would defeat Hillary Rodham Clinton in the New Hampshire Democratic presidential primary 15 months ago? A committee of polling experts on Monday issued a detailed report laying out some reasons why all the surveys were off.

One explanation is that the surveys of the likely New Hampshire Democratic electorate ended too early to detect a primary-eve shift in opinion toward Clinton, who won the contest by 3 percentage points. "There is compelling evidence that this was an important contributor to the error," said the report, which was compiled by the American Association for Public Opinion Research.

The report also said that Clinton supporters may have been underrepresented in voter samples. Clinton did well among voters with lower levels of formal education and income, who may have been tougher to reach by telephone than more upscale Obama voters.*

The report isn't swayed by an explanation that the polling errors can be attributed to the "Bradley effect," or the tendency of some white voters to tell pollsters that they prefer a black candidate but then don't vote for that candidate.

The full report is here.

Obviously, there'll nothing on this from the Obama apparat. It would be nice, however, if the "progressive" blogosphere engaged in a little self-criticism on this one. No grovelling needed; no apologies; a simple and suitably face-saving "mistakes were made" would suffice. Haw. Let me just go check The Obama 527 Formerly Known as Daily Kos on the front page and the diaries... Nothing. I'm shocked.

NOTE So, it was really all about class and not race, after all, wasn't it? Thanks for that, "creative" [cough] "class." Well done.

No votes yet


basement angel's picture
Submitted by basement angel on

Originally, the concern was that Obama supporters only used mobile phones and couldn't be polled accurately. Now, the concern is that Clinton voters are too down scale to own or answer a landline? What the hell that?

I think the answer is very simple. Pollsters went into it with a pro-Obama bias, and missed the real action because of it. They stopped polling too soon because they bought the spin. I bet the truth of the matter is that Clinton voters were underrepresented all along and not because they are too downscale to have phones.

I think I'll go read some 18th century French history and cheer myself up.

gqmartinez's picture
Submitted by gqmartinez on

I was polled several times and only have a cell phone. Several people I know who also have only cell phones were also polled. Cell phone users are definitely polled.