TNR BLOGS

December 01, 2008 | 2:02 PM
December 01, 2008 | 1:00 PM
December 01, 2008 | 12:42 PM

December 01, 2008 | 11:22 AM
December 01, 2008 | 11:10 AM
December 01, 2008 | 9:57 AM

July 26, 2008 | 2:24 PM
July 23, 2008 | 1:55 PM
July 17, 2008 | 3:56 PM

December 01, 2008 | 4:08 PM
December 01, 2008 | 1:36 PM
December 01, 2008 | 12:00 PM
COLUMNISTS
TODAY'S STORIES
16.01.2008
The Reverse Bradley Effect, Cont'd.

Mickey Kaus raises a good question in response to my post about the "other" reverse Bradley effect--the idea that some black voters would be reluctant to say they support Obama for fear of sounding like they're voting out of racial solidarity. Mickey asks (scroll down three screens or so):

But what if this black Bradley Effect operates in the other direction--black voters tell pollsters they are going to vote for Obama (because they feel that's expected of them) and then vote for Hillary or Edwards? In other words, they behave exactly like the white voters in the Standard Bradley Effect. That would take some of the sting out of the implicit charge of "racism" that always lurks underneath the Bradley Effect, no? ... Of the two possibilities, I'd guess the latter is more likely. Are African-Ameican voters really worried that they'll "sound like they're voting out of racial solidarity"?** I'd think fear of being considered a self-hater or Oreo (or practitioner of "middleclassness"!) looms larger in most black communities, unfortunately. But I don't know. ...

I'd say it hinges on the social dynamic during the polling interview. The fear of being considered a self-hater may loom larger in most black communities. But that only means you'd expect some respondents to lie and say they support Obama when the interviewer is black

On the other hand, I'd speculate that when African-Americans are in the presence of whites, the greater social fear is being considered a "race man" (or woman). Which means you'd expect some reluctance to express support for Obama when the interviewer is white. (For the record, I don't think many people--white or black--misrepresent their views to pollsters. Nor do I think many African Americans worry about being considered a race man or woman. We're talking about margins here, but margins can make a difference.)

Now, it's possible that there are more black poll interviewers than white interviewers. But, given how many fewer black people there are in the country/labor force than white people, I doubt it.

--Noam Scheiber  

Posted: Wednesday, January 16, 2008 2:55 PM with 9 comment(s)

Comments

You must be logged-in to comment.

Not a subscriber? Click here to get a digital or print and digital subscription to The New Republic!

virginiacentrist said:

I just don't buy this. People have had a whole generation of black politicians to get used to the idea...maybe the older folks respond to these social pressures, but younger voters are like Stephen Colbert - they don't see race.

January 16, 2008 3:22 PM

stgla said:

Survey researchers have been worried about bias due to socially acceptable responses for a long time, long before Tom Bradley's famous underperformance of his poll numbers.

There are time-tested ways of measuring social acceptability bias in opinion surveys.  One way is to vary the survey mode.  A crude version would be to compare the robocall poll results with the human phone interviews (ideally with an interviewer ethnicity code or assume "white-sounding" interviewers).  If the black candidate does better in the robo-survey (or self-administered mail, which nobody uses anymore for ballot preference surveys) then it's evidence of a Bradley Effect, or more accurately a race-related social acceptability bias.

I'm sure there are a dozen AAPOR papers on this topic.  (American Association of Public Opinion Research).

I suspect that the whole Bradley Effect is oveblown.  Talking about it too much (inducing a Bradley Effect Effect) is really unfair to black candidates, who probably prefer that we talk about policy and governance.  

Or perhaps polling companies should just use offshore (Indian) interviewers.

January 16, 2008 3:22 PM

teplukhin2you said:

What about people who say they'll vote for Obama when they really mean to tell others that they oppose him, in order to prevent people from assuming that they're going to vote for him, which they are, because he's black, when in fact they support him because he's the candidate who's "personally disorganized like me"? Or something like that.

My brain hurts.

January 16, 2008 3:28 PM

ratnerstar said:

Reverse double secret Bradley effect.

Seriously, this is getting absurd.  And by "is getting" I mean "long ago got."

January 16, 2008 4:01 PM

virginiacentrist said:

I just hate talking about it because it devalues Obama's electability. If Democratic primary voters believe that it applies, then they beleive it will apply ten-fold in the general election, where actual racists will weigh in.

January 16, 2008 4:05 PM

teplukhin2you said:

This kabuki reverse double secret stuff is to Mickey Kaus as pedophiles are to O'Reilly and lesbians are to Stern.

January 16, 2008 4:39 PM

TULLIUS said:

The King Effect

Reversing both the Bradley effect and the reverse Bradley effect, what if everyone in the secrecy of the polling booth just voted for who they think is the best person to be President based not on the color of their skin but the content of their character?  What should we call that?

How about the "King effect?"

January 16, 2008 5:19 PM

TULLIUS said:

The King Effect

Reversing both the Bradley effect and the reverse Bradley effect, what if everyone in the secrecy of the polling booth just voted for who they think is the best person to be President based not on the color of their skin but the content of their character?  What should we call that?

How about the "King effect?"

January 16, 2008 5:20 PM

The Stump said:

The Pew Research Center has an interesting piece up on the "Reverse Bradley Effect" I've

April 1, 2008 3:32 PM