Did 'the Bradley Effect' Hurt Obama in New Hampshire?

Well, that was a surprise.

Hillary Clinton came into last night trailing in some polls by double digits, and Barack Obama was expected to put it on her. But that ain’t what happened.

The folks over at Jack & Jill Politics, unabashedly in the Obama camp, said that the discrepancy between the polls and the end result was evidence of the Bradley Effect, named for black Los Angeles Mayor Tom Bradley (above, third from left), who lost to a white candidate in a gubernatorial race after leading in the polls. The idea is that white independent voters who are polled and say they would consider voting for a black candidate (or are undecided) in order to seem progressive but vote for a white candidate when it actually matters. Andrew Sullivan suggested the same thing.

Tonight is the first primary – not a caucus. People get to vote in a secret ballot – not in front of their largely liberal peers, as in Iowa. They may have told the pollsters one thing about voting for a black man, but in the privacy of the voting booth, something else happens. I don’t have any hard evidence for this, but the discrepancy in the polls is remarkable…The vast discrepancy between the last polls and the result puts it on the table. I hope it’s not true. But it could be.

A Stanford social psychologist named Jon Krosnick told Trailhead that evidence for the Bradley effect is largely anecdotal.

There is, however, a large body of research on the effect of the gender and race of an interviewer, both in person and over the phone, and Krosnick points out another scenario:

“People are startlingly good at detecting the race of a person over the phone,” Krosnick told me. An interviewer who is perceived to be black by the respondent can subconsciously influence an undecided voter in favor of a black candidate—something Krosnick describes as a “priming of positive images.” But the same could apply to on-the-fence voters who have some reservations about supporting a female candidate, but are subtly influenced by a capable female interviewer. (The same might hold true in the negative, but people are much more likely to hang up on incompetent interviewers before the interview is complete.)

“You can make up the story either way,” Krosnick says. But he was doubtful that this sort of effect was responsible on its own for the differential between Obama’s lead in the polls and Clinton’s victory tonight: “I just don’t see how you get the discrepancy.”

CNN says that it was much more simple than that: Hillary got women voters back. But, just as interestingly, that same story says that polls showed that her near-tears actually turned off female voters in New Hampshire.

Somebody was telling lies, either about their Obama support or about their opinion on the teary eyes.