Bell Curve The Law Talking Guy Raised by Republicans U.S. West
Well, he's kind of had it in for me ever since I accidentally ran over his dog. Actually, replace "accidentally" with "repeatedly," and replace "dog" with "son."

Monday, January 21, 2008

The Abuela Factor

Some nationwide statistics from the US Presidential Election in 2004:

White voters (77%): Kerry - 41%, Bush - 58%
Black voters (11%): Kerry - 88%, Bush - 11%
Latino voters (8%): Kerry - 53%, Bush - 44%
From 2000 to 2004, Bush made a +9 point gain in the Latino vote, the largest gain in any racial group.

18-29 y.o. (17%): Kerry - 54%, Bush - 45%
30-59 y.o. (60%): Kerry - 47%, Bush - 52%
60+ y.o. (24%): Kerry - 46%, Bush - 54%
From 2000 to 2004, Bush made a +7 point gain in the 60+ vote, the largest gain in any age group.

Female (54%): Kerry - 51%, Bush - 48%
Male (46%): Kerry - 44%, Bush - 55%
From 2000 to 2004, Bush made a +5 point gain in the female vote, larger than his gain in the male vote.

In a comment to the Nevada post below, I posted exit polls from the Nevada caucus. In terms of party identification (which I will get to later), Obama held a +14 point edge with self-described Independents while Clinton held a +12 point edge with Democrats. But otherwise, Obama's support in Nevada showed very little ideological difference (liberal/moderate/conservative) from Clinton's voters. The most dramatic differences between the voting blocs were in terms of race, age, and gender.

Obama won a landslide from African-American voters and 18-29 year old voters--groups in which Democrats already do well. Clinton won a landslide from Latino voters, 60+ year old voters, and held +13 point advantage with women--all groups in which Republicans have made the greatest gains in the past election. They are also generally larger groups (note that Latinos will be closer to parity with African-Americans in the next election).

Of course, Nevada is Nevada, not the entire U.S. But the demographic differences were so pronounced that I have to assume they will carry over to some extent across the nation. And they are fully consistent with the New Hamsphire primary breakdown too. (I'll call Clinton's appeal to elderly Latino women the "abuela" factor.)

I think you can see where I am going with this. If racial/gender lines prove more important than party identification in the next election, it looks like Obama's demographic appeal is less useful to Democrats than Clinton's would be. To be crude, when it comes to playing the "Black Card" or the "Woman Card", you gotta remember that there are a lot more women and African-Americans already vote Democratic. Likewise the elderly, who apparently prefer Clinton's claim to experience, outnumber the young, who apparently prefer Obama's call for change.


USWest said...

I am interested in some of these age groupings. Business Week ran an article on the voting habits this election of the Millennials- those 18-29 year olds. They are all Obama all the time.

But I am not 29 anymore. And Gen X ers are the group squeezed between two huge generations and the next group poised to enter the halls of decision making. So I am curious what our group is doing. Anyone have anything on that, or are we just split between the 20 somethings and the 40 somethings?

I'd point out that 30-59 is a 20 year gap whereas 18-29 is half of that size. Do we adjust for these things when dealing with polling?

Dr. Strangelove said...

The statistics I referenced actually broke it out into 10 year gaps: 30-39 and 40-59. But the results were almost identical so I lumped them together above.

Raised By Republicans said...

The assumption that women who are not self identified Democrats will support Clinton in the same or even close to the same proportion as women who do is not a good one to make.

I'll be very interested to see what the independents did in 2000 and 2004. I'll bet you dollars to donuts that they broke for Bush.

And let's not forget another critical element here. The 2008 election is NOT a one sided decision. It matters who the Republicans pick. And who the Democrats pick will effect Republican turnout. I think it is safe to say that a Hillary Clinton nomination will cause a surge in Republican fundin support and turnout. If she also allienates independents the Democratic party could easily blow another election that they thought they had all sewn up.

And for what? The fantasy that we can get the 1990s back? The prospect of doing unto the Republicans like they did unto us for the last 7 years?

A vote for Hillary is a vote for 50%+1 politics. If it works (and I'm increasingly doubtful it will), we'd have a Democratic majority in all three branches of government tempted to cram over reaching pipe dream policies down the throats of a Republican party in temporary minority status.

That would run the risk of provoking a massive backlash against the Demcorats that would waste the opportunity to build a really solid center-left super majority in favor of moderate but fundamental change.

Why do you think that Democratic Senators from "Red" and "Swing" states are lining up behind Obama? Because they're scared to death of what Hillary will do to the turnout in their reelection bids.

The Law Talking Guy said...

Putting on my game theorist hat for a moment: doesn't the party that chooses its candidate later have the advantage, since it may adjust to the other party's choice?

Raised By Republicans said...

Yes. But parties aren't individuals. So it doesn't work as cleanly and strategically as you suggest.