pageok
pageok
pageok
Interesting Pa. Exit Poll Datum:

9% of the voters in the Pennsylvania primary were African Americans who said race was not a factor in their vote. Of those 9%, 91% voted for Barack Obama, exactly the same as the overall African American percentage for Obama, including the 4+% who said race was a factor in their vote (the poll doesn't give a breakdown of the exact percentages for the latter group, but it can't be very different than 91-9).

By contrast, both men and women (especially women) who said that "gender" was a factor in their vote were significantly more likely to vote for Clinton. And whites who said race was a factor in their vote were significantly more likely to vote for Clinton.

Duffy Pratt (mail):
I'm not sure what your point is. One possible conclusion is that the blacks who said that race was not a factor were either lying or kidding themselves. Another possible conclusion is that 90-10 is the natural split in the vote for people who are not racially biased, and that shows exactly how racist the Pennsylvania whites are.
4.22.2008 11:10pm
DavidBernstein (mail):
I'm leaving conjecture up to others. The point is that it's interesting, not that I'm going to explain to you what it means.
4.22.2008 11:13pm
Duffy Pratt (mail):
Sorry David, but its only interesting if it means something. If you think its interesting, then that makes me believe you also think it means something. Leaving it up to others to conjecture is just hiding the ball.
4.22.2008 11:19pm
ithaqua (mail):
"And whites who said race was a factor in their vote were significantly more likely to vote for Clinton."

The party of Robert "KKK" Byrd strikes again.
4.22.2008 11:23pm
Anon e mouse:
DB -- Why is it interesting? It seems passive aggressive to say something is "interesting" without saying why you think it's interesting.
4.22.2008 11:24pm
Daniel Chapman (mail):
Lots of things can be interesting, yet completely meaningless. Even more things are interesting, yet completely incomprehensible. If he thinks it has some secret meaning he's not sharing, who the hell cares?

IMHO, the most interesting thing about this post is the fact that DB could post "the sky is blue" and someone would bitch about it...
4.22.2008 11:38pm
DavidBernstein (mail):
Okay, it's interesting both because the votes of individuals who believe themselves to not be considering race are so lopsided, and also because there is no meaningful disparity in the percentages between those who said race was a factor and those who said race wasn't a factor, unlike the disparity with white voters who considered race and both men and women who considered sex. I thought that much is implicit in the post. But what it "means", I don't know.
4.22.2008 11:39pm
byomtov (mail):
Why is it interesting? It seems passive aggressive to say something is "interesting" without saying why you think it's interesting.

It's interesting for the reason DB implicitly cites. Those who say race is not a factor break down the same as those who say it is. That's odd, hence interesting.

It is possible to find an observation interesting without having a theory to explain it. In fact, not leaping to invoke some theory or other is commendable.
4.22.2008 11:43pm
sef:
Ithaqua:

Yes, the party that people of color overwhelmingly support is the party of the Klan. ::tic::
4.22.2008 11:52pm
ithaqua (mail):
Sef:

Out of white Democrats who said that race was a factor in their decision, 75% voted for Clinton. If you can spin that in a way other than Dixiecrat ain't-gonna-vote-for-no-n*****ism, be my guest.

Why do you think the Democrat Party elite are so insistent that blacks need affirmative action, and college scholarships, and all the other accouterments that supposedly counterbalance America's 'inherently racist' society? It's because they look at white Democrats and project all that hatred and contempt and 'white guilt' onto genuine conservative Americans. They oppose the genuine equality offered by the Republican Party because they know granting white Democrats freedom of association would lead to de facto segregation and they think the rest of the country is just like them.
4.23.2008 12:10am
Peter B. Nordberg (mail) (www):
One thing these numbers might mean is that a voter's ethnic and cultural background can influence a voter's perceptions, priorities, and decisions in various complex ways, even if the candidate's ethnicity is not given conscious or unconscious weight in the voter's decision-making matrix.
4.23.2008 12:41am
neurodoc:
If he thinks it has some secret meaning he's not sharing, who the hell cares? IMHO, the most interesting thing about this post is the fact that DB could post "the sky is blue" and someone would bitch about it...
Perhaps you have noticed that whenever DB mentions Israel, the anti-zionists and antisemites swarm in response. So, to avoid that noisome crowd, he is now encrypting his posts. Those of us with decoder rings know exactly what he is talking about here.
4.23.2008 12:55am
A. Zarkov (mail):
If you look at the exit polling data you will find a fairly strong age effect for everyone. The older the voter, the more likely he's a HRC voter irrespective of race (approximately). Suppose blacks who say that age was not a factor in their vote tended to be young. That alone would decrease the their HRC vote. So we really can't tell if the blacks who say race was not a factor are lying. They might simply be the young ones.

There does appear to be a problem with the "age by race" table. It seems to me they reversed the black and white labels. The white percentages add up to about what you would expect for the black percentage.
4.23.2008 1:08am
John Thompson (mail):
Sef: what the heck is a "People of color" anyway?
4.23.2008 1:38am
CrazyTrain (mail):
So your point is? That black people lie? I don't get it.
4.23.2008 1:42am
eric (mail):
Having fired up my David Bernstein Zion Decoder Ring, I agree David, I agree.
4.23.2008 1:42am
eric (mail):
The point would be that the small percentage of blacks did lie or are deluding themselves and that the whites were open about their racism. At least that was undecoded meaning . . .
4.23.2008 1:44am
big dirigible (mail) (www):
I'm afraid that "race is a factor in their votes" doesn't really mean much.

Consider a hypothetical voter who has decided that he won't vote for a racist. This voter is hypothetically disgusted by the Wright revelations, and has decided that he would no sooner vote for Obama than he would vote for someone who had regularly attended Klan meetings for the past twenty years, reasoning that black racism is still racism, and he wants nothing to do with it.

Now to this voter, "racism" is definitely a factor in his vote. But can one have "racism" without "race"? When a pollster asks him if "race" is a factor in his vote, what should he say? If he says "yes, racism is a factor," are we justified in lumping him in with "racist" voters, even though he's about as "anti-racist" as a voter could be?
4.23.2008 1:53am
Grover Gardner (mail):

Out of white Democrats who said that race was a factor in their decision, 75% voted for Clinton. If you can spin that in a way other than Dixiecrat ain't-gonna-vote-for-no-n*****ism, be my guest.


Hold on there, bunny. That was 75% of the 13% who said race was a factor. A healthy majority of white Pennsylvania Democratic voters (66%) said race was NOT a factor. Over all, 80% of Democratic voters said race was not a factor. Gee, where did all that projection go?

Of course, we could look at Republican primary voters and see how race affected their choices, except...ur...um...
4.23.2008 2:06am
Grover Gardner (mail):

Those of us with decoder rings know exactly what he is talking about here.


Who needs a decoder ring?
4.23.2008 2:15am
Grover Gardner (mail):
Let's just reemphasize the point, shall we, ithaqua? A whopping eight percent of white PA Democrats appear to have voted for a white woman because of her race. Man, the place is just crawling with Dixiecrats, huh?
4.23.2008 2:29am
steve (mail):
The N is over 2200, so even if its 75% of 13%, thats still pretty suggestive. I think even more important though is that the difference is likely much larger among Republicans and Independents. Thats a lot of important (white) voters breaking against a Obama candidacy. If this is accurate, then this spells a lot of trouble for Obama in states like Minnesota, Colorado, New Mexico, Ohio, Pennsylvania, Michigan and Florida. He can't win a general if he loses any of those states and each of them have demographic similarities that can't be overcome by black/white elite voters alone. His recent snafu's and continuous missteps offending whites only needs to offend a fraction of whites to have a significant impact on his ability to build a winning coalition. These may not be Reagan Democrats of old, but they seem to be similar.. call them the "Guns and Bitter Democrats".

Props to Clinton for keeping the mistakes to a minimum under considerable stress. This campaign alone proves beyond any doubt in my mind that she has the moxy to scrap in the yard if need be. I'd vote for her if I didn't disagree with her on such fundamental ways. Obama on the other hand looks like an amateur who is riding a wave of elite-driven emotion, aided by an insufferably resistant media.
4.23.2008 2:33am
Brian G (mail) (www):
Good strategy for getting Obama votes: calling whites who don't support him racist.

Bottom line, as an urban cracker, raised in Philly, which ahs been owned by the Democrats since the 1930's, I won't vote for Obama because he is a Socialist, because he will throw Israel under the bus, and he is anti-2nd Amendment.

By the way, I saw that there were 21 murders in Obamatown, aka Chicago, this weekend. I'd like to hear Obama explain how that is possible given the gun laws there.
4.23.2008 2:43am
steve (mail):
Brian.. I don't think anyone is calling whites who don't support Obama racist. The data above shows that of the respondents who thought race was an important factor in their vote, 75% of them voted Hillary. They could be whites who are offended by Obama's handling of race issues. The data only allows us to speculate what the voters meant by answering that question in the affirmative, but it is suggestive that whites who did see race as important broke hard against Obama, not for him.
4.23.2008 2:49am
jum1801 (mail):
What conclusion may we draw from the fact that 91% of those African-Americans who said race was not a consideration voted for Obama? Well, that these voters' capacity for self-delusion is strong; and/or these voters appear to have indulged in the politics of racial identity. Either way, it's good for a chuckle.
4.23.2008 2:52am
Grover Gardner (mail):

These may not be Reagan Democrats of old, but they seem to be similar.. call them the "Guns and Bitter Democrats".


That's a heck of a lot fairer than calling the Dixiecrats.
4.23.2008 3:01am
Grover Gardner (mail):

I don't think anyone is calling whites who don't support Obama racist.


Am I crazy, or did someone not just do that?
4.23.2008 3:07am
Grover Gardner (mail):
And I don't mean you, steve.
4.23.2008 3:12am
steve (mail):
Grover.. sorry.. i missed the dixiecrat reference.. that is probably a low blow
4.23.2008 3:14am
Grover Gardner (mail):

By the way, I saw that there were 21 murders in Obamatown, aka Chicago, this weekend.


There were not 21 murders in Chicago this weekend. There were 32 shooting incidents and 6 deaths. It was a very violent weekend.


I'd like to hear Obama explain how that is possible given the gun laws there.


According to police, the usual culprits--an increase in gang activity and warmer weather.
4.23.2008 4:35am
Bob Wilson:
Nine in ten blacks voted for Obama. Six in ten whites voted for Clinton. Doesn´t it seems blacks are more racists than whites? As far as I know, the 9 in 10 blacks is what Obama wins generally. Is it really possible 90% of blacks think Obama´s policies are better than Clinton, or do they just want a black president no matter what?. Seems like the latter to me.
4.23.2008 6:39am
Lonetown (mail):
Let's see:
9 out of 10 blacks vote against Reagan.
9 out of 10 vote against Bush 1
9 out of 10 vote against Bush 2
9 out of 10 vote against Hillary.

I've got to say I think I know who the racists are.
4.23.2008 6:54am
neurodoc:
What conclusion may we draw from the fact that 91% of those African-Americans who said race was not a consideration voted for Obama? Well, that these voters' capacity for self-delusion is strong; and/or these voters appear to have indulged in the politics of racial identity. Either way, it's good for a chuckle.
If the candidate were Al Sharpton and 91% of the 9% who said race wasn't a factor in their vote had gone for Sharpton, we would know without a doubt it was all about race. But it wasn't Obama, and it could have been a Colin Powell, so it could be that they really believed their choice to be the best one for the Democratic Party to go with. Identity politics may be the most plausible explanation notwithstanding the "race not a factor" responses, but it need be that, nor self-delusion.
4.23.2008 8:32am
neurodoc:
Let's see: 9 out of 10 blacks vote against Reagan. 9 out of 10 vote against Bush 1. 9 out of 10 vote against Bush 2. 9 out of 10 vote against Hillary.

I've got to say I think I know who the racists are.
You could define "racism" that way, but to do so trivializes the notion of a real, pernicious phenomenon to nothingness.
4.23.2008 8:36am
neurodoc:
Can someone give me the shorthand rationale for why the Democratic Party elected to go about the business of choosing a candidate in the way they are doing it. When a candidate wins a hugely important state like Pennsylvania and gains so little in the fight for the party's nomination, it makes me wonder how this process serves the Democrats' self-interest, which presumably is to win in November. Speaking for an absolutely non-partisan political science sort of perspective, is the GOP doing it the smart/stupid way with their winner-take-all approach, or are the Democrats doing it the smart/stupid way with their "proportional" or whatever it is approach?

Perhaps one thing to be said in favor of the Democrats' approach is that they are collecting many times what the Republicans have in campaign contributions. Or maybe that has little to do with the fact that they have more of a horse race going on, and everything to do with the Democrats's pent up anger.
4.23.2008 8:50am
PersonFromPorlock:
neurodoc:

Can someone give me the shorthand rationale for why the Democratic Party elected to go about the business of choosing a candidate in the way they are doing it.

My pleasure: "It seemed like a good idea at the time."
4.23.2008 9:00am
JT234 (mail):
What a nice little data set to apply some basic statistics. Using numbers pulled from the CBS exit poll, we can create a model for the results. Start by noting that Dem voters in PA split 54:46 Clinton:Obama. Looking at different groups, each one adds a bias, or shift to the underlying voter pattern. For instance, those making over $100k shift 3 percentage points to Obama from Clinton (to 51:49 Clinton:Obama). Looking at gender and race groups, not surprisingly there is always a shift towards the candidate most like themselves, but the degree varies:

Females 3% to Clinton
Whites 8% to Clinton
Males 6% to Obama
Blacks 43% to Obama

Make of these numbers what you will. Obviously, blacks are (statistically) much more driven by race than any other group is driven by race or gender. I'm a bit surprised how little extra pull Clinton has among women.

Also interesting is the fact that by applying two shifts, one can predict, to within a percent or so, how, for instance, White Males will vote (8% to Clinton minus 6% to Obama is a net 2% to Clinton). This indicates that there isn't much interaction effect between race and gender.
4.23.2008 9:33am
T. Gracchus (mail):
Alright JT234 stop with the intelligent interventions. We are busy insulting one another and it is distracting to have thoughtful remarks interposed.
4.23.2008 9:42am
Ardsgaine (mail) (www):
Given that there is precious little difference in their policies, and given that there has never been a black US president, are we supposed to be surprised that black Democrats broke decidedly in Obama's favor? None of the "character issues" which have gotten so much play in the media were designed to move black voters, especially when set against the symbolism of having a black man elected president.

Democratic Party identity politics is racist, but, set against that background, the black vote does not strike me as especially racist.
4.23.2008 10:14am
Loren (mail):
By contrast, both men and women (especially women) who said that "gender" was a factor in their vote were significantly more likely to vote for Clinton. And whites who said race was a factor in their vote were significantly more likely to vote for Clinton.


Now this discrepancy I find interesting. Voters who said gender was a factor in their decision tended toward the novelty and newness of the female candidate. But white voters who said race was a factor leaned heavily in favor of the less novel and more traditional white candidate.

So for those who considered gender a factor, they voted for the candidate who would break the streak of 43 male Presidents. But for those who considered race, they voted to continue an identical streak of white Presidents.
4.23.2008 10:37am
Aultimer:

Bob Wilson:
Nine in ten blacks voted for Obama. Six in ten whites voted for Clinton. Doesn´t it seems blacks are more racists than whites?


On the last two PA Democratic primaries, ten in ten men voted for a man, and ten in ten white people voted for a white person. Is it fair to say you know who the sexists are?
4.23.2008 11:20am
JT234 (mail):
Working in opinion or market research, if you want to understand people's motivations, it is usually very good advice to focus less on what they say, and more on what they do. That goes double if the topic is one where some answers might be embarrassing or seen as socially undesirable; I think this is the case if you ask people whether race or gender is a factor in their vote.

Going back to how men, women, black and whites voted, you can say that gender is not much of an issue for any group (biases of 3% or 6% are not that big). Race is a bit more of an issue for whites (bias of 8%) but huge for blacks (bias of 43%).

Of course, this analysis only talks about groups as a whole, not individual motivations, but I think the result is pretty robust.

Personally, I'd prefer to think of this as a pro-black, rather than an anti-white bias, on the part of black voters. But neither the data nor the "is race a factor" question can say if this is true.
4.23.2008 11:29am
JD22:
When 92% of blacks are voting for the black candidate, I think it's obvious which group is voting along racial identity lines.

If there was ever a situation where the white proportion of the vote for Clinton was *higher* than the black proportion of the vote for Obama, then and only then could you claim that the white voters supporting Hillary are doing so out of racial identity politics.

So far, that hasn't happened.
4.23.2008 12:33pm
anonthu:
set against that background, the black vote does not strike me as especially racist.

Set against the statistical background (see JT234's post), I would say that the black vote is most definitely "racist", that is, race appeared to be a key factor in determining their voting behavior.
4.23.2008 12:51pm
Joe Bingham (mail):
I don't understand why anybody bothers to call white voters who cited race as a factor "racist" when only 9% of black voters said race wasn't a factor. That's incredible.

Personally, I think the 91% of black voters and 13% of white voters who didn't call race a non-factor are racists, and that it's despicable that race would factor into their decision here (not kidding). But one of those numbers is a lot more staggering than the other, and it's absolutely incredible that people would be singling out white Pennsylvania voters for being racist.
4.23.2008 12:54pm
Grover Gardner (mail):

I don't understand why anybody bothers to call white voters who cited race as a factor "racist" when only 9% of black voters said race wasn't a factor. That's incredible.


John, I think you're confused. Black voters comprised 14% of the Democratic vote in PA. Approximately two-thirds of those black voters (9% of the total Democratic vote, not 9% of black voters) said race wasn't a factor in their vote.
4.23.2008 2:03pm
Grover Gardner (mail):
Sorry--Joe.
4.23.2008 2:05pm
Displaced Midwesterner:

Let's see:
9 out of 10 blacks vote against Reagan.
9 out of 10 vote against Bush 1
9 out of 10 vote against Bush 2
9 out of 10 vote against Hillary.

I've got to say I think I know who the racists are.


I wasn't aware that Reagan's opponents, Bush 1's opponents, and Bush 2's opponents were black.
4.23.2008 2:54pm
jgshapiro (mail):

When a candidate wins a hugely important state like Pennsylvania and gains so little in the fight for the party's nomination, it makes me wonder how this process serves the Democrats' self-interest, which presumably is to win in November.

Maybe, but if you win by one vote, presumably that doesn't mean much in terms of who would win in November. So to give you all of the delegates and your opponent none may resolve the primary contest earlier, but it doesn't necessarily tell you who has broader appeal. At the same time, as the victory margin increases, you have a better idea of who can assemble a majority in the general. Maybe the better solution is to award increasingly more delegates as the margin increases, rather than use proportionality as the guide.

Also, I would dispute whether PA really tells much. Recall that in PA, independents can't vote. (I think the same was true for Ohio.) I'm not sure that Clinton has beaten Obama in any state where independents could vote in the primary, which leads me to question whether she can build a majority in the general against someone like McCain who clearly appeals to independents.
4.23.2008 4:01pm
LM (mail):
I don't see how a primary can take account of the different voting dynamics that determine a general election without replacing one-candidate-per-vote with a full blown questionnaire. Given that, though the current arcane system happens to favor my candidate, I wish Democratic primaries would go to a simple popular vote. For one thing, more than two or three people would understand it. More important, it would eliminate one source of the infighting and alienation we're inflicting on ourselves now. (Of course, being Democrats, for every means of bickering we eliminated, I'm sure we'd come up with two new ones.)
4.23.2008 4:54pm
pete (mail) (www):

I'm not sure that Clinton has beaten Obama in any state where independents could vote in the primary, which leads me to question whether she can build a majority in the general against someone like McCain who clearly appeals to independents.


Do you count Texas? Anyone can vote in whatever primary they want here and Clinton won the primary here. You just pick what party you are voting for when you walk up to the voting booth. I almost voted for Clinton, but ended up voting for McCain again.
4.23.2008 6:05pm
Gaius Marius:
Poor Barack Hussein Mohamad Obama (yes, he has an additional middle name that he chooses not to publicly display for obvious reasons). It must be publicly humiliating to be publicly emasculated by Hillary "Lady MacBeth" Clinton.
4.23.2008 6:28pm
SIG357:
"I almost voted for Clinton, but ended up voting for McCain again."



Same difference. You can't slip a razor blade between them.
4.23.2008 6:45pm
SIG357:
I would dispute whether PA really tells much. Recall that in PA, independents can't vote.



Well, how is that different from all the caucus states where Obama wracked up his advantage in delegates? Not only can independents not vote, it's not even representative of Democrats as a whole.
4.23.2008 6:48pm
Brian G (mail) (www):

There were not 21 murders in Chicago this weekend. There were 32 shooting incidents and 6 deaths. It was a very violent weekend.

I'd like to hear Obama explain how that is possible given the gun laws there.


According to police, the usual culprits--an increase in gang activity and warmer weather.


Ok, you told me a possible "why" but again, "how" did that happen under Chicago's gun laws? After all, we all know that if you have strict gun restrictions, there will be no gun crime. At least, that's what I'm told.
4.23.2008 8:12pm
EIDE_Interface (mail):

Brian G (mail) (www):


There were not 21 murders in Chicago this weekend. There were 32 shooting incidents and 6 deaths. It was a very violent weekend.

I'd like to hear Obama explain how that is possible given the gun laws there.


According to police, the usual culprits--an increase in gang activity and warmer weather.



Ok, you told me a possible "why" but again, "how" did that happen under Chicago's gun laws? After all, we all know that if you have strict gun restrictions, there will be no gun crime. At least, that's what I'm told.


Stop asking silly questions! We must never question the holy anti-gun gods.
4.23.2008 8:31pm
Grover Gardner (mail):
"After all, we all know that if you have strict gun restrictions, there will be no gun crime. At least, that's what I'm told."

No one told you that.

Atlanta has far more liberal gun laws, including CCW, and a much higher crime rate than Chicago. I'm not against gun ownership, and it doesn't bother me at all to be living in a state and city with very liberal gun laws. I won't have one in the house but that's just me.

But whining and saying stupid shit about gun laws instead of solving the other problems that cause violence in inner cities strikes me as particularly dumb.
4.23.2008 9:00pm
Joe Bingham (mail):
You're right, Grover, I was confused. Thanks.

And I'm relieved :-p At least until I see a full breakdown of the stats.
4.24.2008 1:36am
Grover Gardner (mail):
After I replied to your comment, Joe, I realized I made the same mistake myself somewhere upthread, in referring to percentages of white voters vs. percentages of overall voters. So mea culpa too. :-)
4.24.2008 7:01am