Who's not willing to vote for whom? (Rasmussen poll)

As always, let's start with the caveat that figures from public-opinion polling always require some speculative interpretation, and even then should always be taken with a grain of salt. Nevertheless, these results from a recent Rasmussen Reports poll are intriguing:

Seventy-one percent (71%) of the nation’s voters say they would be willing to vote for a woman for President. Seventeen percent (17%) say they would not and 11% are not sure.

Results are similar when asked about voting for an African-American—73% are willing to cast such a vote, 14% are not, and 13% are not sure.

I find it interesting that the declared levels of unwillingness + ambivalence were roughly the same for black and for female candidates (27% for an African-American and 28% for a woman). And these percentages seem to be only slightly different among respondents who describe themselves as politically "moderate" (though the proportions of "not sure" responses aren't given for them):

Among political moderates, 12% say they would not vote for a woman. Fifteen percent (15%) would not vote for an African-American.

Self-described Democrats are apparently less reluctant, though not all of them are willing to vote for a black or a woman, either.

Seven percent (7%) of Democrats say they would not vote for a woman. Nine percent (9%) would not vote for an African-American.

=> Two immediate impressions: On the one hand, I have no doubt that these figures represent a genuinely dramatic change in public attitudes over the past half-century. I feel sure that most respondents who said they would be willing to vote for a black and/or for a woman for President really meant it, and even the ones who didn't realized that expressing blanket prejudices against black or female candidates is no longer respectable--which counts for something by itself.

On the other hand, the fact that over a quarter of respondents are still willing to say that they might not, or definitely would not, vote for a candidate because of that candidate's race and/or gender is less heartening. That's especially true since I suspect that the number of voters who would have qualms about voting for a black or a female candidate is greater than the number who are willing to admit this to a pollster. That might be one way of interpreting the significance of some other findings from this poll:

While the overwhelming majority of voters say they could vote for a woman or an African-American, just 56% believe their family, friends, and co-workers would be willing to do the same. Among senior citizens, just 41% believe their peers would be willing to vote for a woman. Forty percent (40%) believe their peers would be willing to vote for an African-American.

What might these discrepancies mean? One possible implication is that a lot of respondents are significantly underestimating the open-mindedness of their friends, relatives, and co-workers. Another possible implication, though, is that some respondents who gave the 'correct' (i.e., non-bigoted) answer when asked about their own votes were using these other questions to signal that, in reality, they still feel more ambivalent than they're willing to admit.

(By the way, whether or not such attitudes are as widespread as this poll suggests, or even if they're more widespread, it's not obvious how they would affect the final outcome of the 2008 presidential election. For example, it's plausible that those voters who really are unwilling to vote for a woman and/or for an African-American under any circumstances would also be voters who are disproportionately unwilling to vote for any Democratic candidate. And this year all three of those categories might be outweighed by voters who, in the end, are not willing to vote for any Republican.)

I suppose we'll see what happens in November ... and then we can try to figure out what that means.

--Jeff Weintraub