DEMOGRAPHY IS DESTINY.
Patrick Ruffini has a very smart overview of how flawed the polling has been in the Democratic primary. Overall, the polls have underestimated the winners support by about 5.86 percent, and had an average error of almost seven points. In certain contests, this had huge impacts on the race, because the polling set the expectations. Clinton's win in New Hampshire was so contrary to the polls that the American Society of Public Opinion Research formed a committee to evaluate the discrepancy. In South Carolina, polls showed Obama up by 11 points, he won by almost 30 percent. In Wisconsin, polls showed Obama ahead by 4 percent. He won by 17.
But these weren't random swings or unpredictable outcomes. Rather, they all fit a precise pattern, and could indeed be predicted with fairly simple formulas. Ruffini writes, "Now that we are approaching a perfect 50-state electoral breakdown of Clinton-Obama -- marred only by the presence of small-state caucuses -- we see that the results are totally internally coherent based on demographics. In fact, I'm prepared to argue that we would have gotten almost the same margin in each state in a national primary held February 5th. Because the candidates themselves polarized the electorate demographically, momentum and events (and polls) made little difference." Indeed, the absence of a momentum effect, a la Kerry, has been starling. Ohio didn't care what West Virginia did, New Hampshire had no interest in Iowa, and North Carolina was unmoved by Pennsylvania. In this race, demography really has been destiny, and that doesn't strike me as an encouraging conclusion.
Feeds: 


COMMENTS (27)
Doesn't this just mean that people voted demographics (ie, their own lives) instead of momentum (ie, media narrative)?
Posted by: Joel | June 2, 2008 3:04 PM
While I agree that it seems disturbing that a lot of what seems to have driven the Dem primary vote is identity -- and that's not just are you black, or a middle-aged and up woman, but also cultural and social kinds of identification -- I wouldn't overinterpret it either. Remember that there wasn't much difference between these candidates on policy or ideology -- in fact you could barely force a playing card between them. So people didn't have much else to go on. The general election, with a strong ideological and policy contrast between the candidates, is a different story entirely.
Posted by: cervantes | June 2, 2008 3:04 PM
It seems like that means that the media being a bunch of big doodie-heads (an argument I'm willing to grant on the merits with the maximum prejudice) doesn't really have much to do with it either.
Posted by: dbt | June 2, 2008 3:14 PM
I apologize that I can't dig up cites on this stuff because I'm at work, but doesn't this contradict what we were hearing about the shifts in the black vote away from Clinton early on? That seems to really heavily contradict this view that it was all just demographics.
I'm not contesting the idea that the party has kind of coalesced around demographic blocs. But I think Chris Matthews' mockery of Clinton, the "did she cry or not" sleaziness in NH, Bill's comments in South Carolina, and Geraldine Ferraro had a lot to do with forcing people into those demographic blocs.
Posted by: NS | June 2, 2008 3:16 PM
Hell, I pointed that out back in late February.
However, what the demography reveals is that except for blacks, Obama only won the liberal vote (roughly 30-40% of Democrats overall). It's just that the liberal vote was very useful in the primaries.
Posted by: Cal | June 2, 2008 3:19 PM
What Cervantes said.
The two remaining candidates are basically identical. What else is there to vote on, apart from personality and demographics (and the former is just an indirect way of reflecting the latter)?
Posted by: DHS | June 2, 2008 3:21 PM
I don't see what's wrong with this at all. None of these demo groups voted 100% for a candidate, so clearly there is still some free will.
It's just of the groups that determine a person's political opinions, it's not bs media stories or their particular "state" or "region". Frankly, demographic is a step up from either of those.
This is only shocking because we haven't had an election like this before. But I imagine doing regressions of past presidential elections would show that say the "male middle income black vote" in Ohio voted in the same proportion as in California.
Posted by: Shock Mouse | June 2, 2008 3:25 PM
Ohio didn't care what West Virginia did
Ohio was before W. Virginia
Posted by: Anonymous | June 2, 2008 3:33 PM
Which came first, the demographics or the campaigns?
It seems to me that campaigns these days pick their demographic group off the shelf from the ones they think will be left available to them, recognizing that people do tend strongly to vote based on their demographics. It's a form of special interest politics, and not very attractive. I was a strong Edwards supporter but deep down I'd have to admit that he probably looked at the landscape of running against Hillary and picked the best available niche of a populist, anti-establishment candidate and ran with it. This campaign started out being different, but soon feel into the same old predictable rut. The fascinating thing to me is to see how adroitly Clinton moved into the space left open by Edwards when he left.
Posted by: Green VTster | June 2, 2008 3:38 PM
I'm not sure what's discouraging.
Educated people voted for Obama in overwhelming numbers.
Folks who quit at high school or lower voted for Clinton.
Teach America to read and they will stop voting for lying, sociopathic monsters.
The amount of smoke blown up cracker America's ass around every election is disgusting. These aren't "real" Americans. They're the dregs that we need to wipe out by educating their children.
Appalachia is not Obama's problem. It is America's problem. They're like this country's mentally retarded cousin.
Posted by: CrackerCentral | June 2, 2008 3:42 PM
doesn't this contradict what we were hearing about the shifts in the black vote away from Clinton early on?
Maybe not -- I think we were hearing that Clinton was polling well among black people until Iowa (and maybe second place in New Hampshire) showed that Obama was for real. That would mean that the only primary before the shift was New Hampshire, and since New Hampshire is only 1.1% black it just doesn't matter whether the vote had shifted or not.
Posted by: Matt Weiner | June 2, 2008 4:17 PM
That being said, it's not like the demographic issues are going to disappear once the primaries are over. Ezra's link to the article about the likely GOP strategy illustrates the point. The GOP knows that the only thing propping it up from complete collapse are its brand and its appeal to identity politics.
McCain will do his level best to try to make the apparent fractures in the Democratic base into actual fractures. Obama and the Party will have their work cut out for them to make sure that doesn't happen by running the race on our terms, not on the GOP's.
Obama's organization is probably in the best position to make sure that's the case. It does a tremendous job of making up for any disadvantages that crop up in the media war. In Indiana, he closed the gap to a 2% margin. In PA, he cut Clinton's lead in half. Both of those results came despite weeks of media coverage that pushed identity-based scandals to the forefront of the campaign.
As voting patterns in Appalachia indicate, some folks just won't vote for a candidate no matter what you do. But fortunately, most of the American electorate isn't like that.
Americans barely respond to media stories unless they confirm pre-existing biases. They don't trust what politicians say.
Who can convince them? Their neighbors. Obama has lots of neigbors working to convince their neighbors to vote for him. It's an intrinsic advantage that we should not discount.
Posted by: Unapologetic Andrew | June 2, 2008 4:20 PM
I don't understand. If it's all demographics, then why did Obama's poll numbers in so many states increase significantly in the weeks before the primaries when he campaigned in those state? Is that supposed to be some sort of weird polling fluke not measuring actual changes in people's candidate preference?
Posted by: KCinDC | June 2, 2008 4:28 PM
Educated people voted for Obama in overwhelming numbers.
Wrong. They voted for Obama about 56-44, which is a solid majority but not overwhelming. Clinton did get overwhelming support among high school graduates, well over 60%. She has picked up support among college graduates since March 4th, and has actually won voters over $100K since then as well (Obama won the category before that).
There was no shift in the black vote. They always voted overwhelmingly for Obama.
Again, the demographics aren't really the story except that they illustrate Obama's two strong areas: blacks and liberals.
If you knew the percentage of liberals in a state and blocked out whites and hispanics based on that percentage, then assume 90% of the black vote goes to Obama, you'd get a pretty good prediction of most primaries. The few exceptions were earlier primaries when Republicans hadn't figured out how unpleasant Obama's associates were and were taking an opportunity to vote against Clinton. But if you narrow it down to white Democrats, it's all very predictable.
Posted by: Cal | June 2, 2008 4:40 PM
18 month primary season and it wasn't about issues at all. And these people want to run the country??
Obama's only mandate is he's black..he just needs to be black, but not too black.
Posted by: Anonymous | June 2, 2008 4:40 PM
the demographics changed a bit over the course of the campaign.
In early states Obama did better in rural areas and with upper income/education voters. He slipped in both categories by the end but increased among the poor (under 15K). Clinton did progressively worse with black voters but increased her share of white women.
I think in the end it was just a routine contest between the wine track (plus African-Americans( and the beer track (supplemented by more women) but because of the candidates involved and the length of the contest it appeared to be some titanic battle.
Posted by: fairly average | June 2, 2008 4:59 PM
In this race, demography really has been destiny, and that doesn't strike me as an encouraging conclusion.
This sounds like "damned if you do, damned if you don't" to me. In every primary for the last 20 years, momentum has been destiny. That's bad. Among other things, it makes the mainstream media too much of a player. It gives grossly disproportionate power to the residents of a handful of states, which would be bad even if those states were representative of the country, which they aren't. It also indirectly reduces turnout and participation, at least in my opinion, because if people have primaries that matter then they're more likely to vote in those primaries and in the general. This time, momentum hasn't been destiny... and that's bad too. Huh?
Also, "demography really has been destiny" doesn't mean what it sounds like. Demographics tend to correlate with political views. You have to parse things very closely and draw some kinda arbitrary demographic lines to say that demographics have been destiny (Latinos and older voters supporting one candidate, black people supporting one candidate but so did overwhelmingly white Vermont and Iowa), but if you draw liberal versus conservative lines — maybe socially liberal vs. socially conservative, to be a little more careful — it looks pretty ideological after all.
And when I follow the link, I find it's to a professional Republican strategist. Three of the four commenters seem to agree that Democrats are racist (or in thrall to identity politics, whatever that means). What a surprise.
Posted by: Cyrus | June 2, 2008 5:10 PM
To be fair, I guess I should say that the authors of thenextright.com seem better than most conservatives and/or Republicans out there. Three out of four saying Democrats are racist is probably fewer than at most Republican blogs. And I don't read many right-wing blogs except as they're linked from left-wing blogs, which probably gives me a pretty jaundiced eye in the first place.
Posted by: Cyrus | June 2, 2008 5:19 PM
The best McCain can hope for is a low-turnout election decided by the dedicated religious organizations. But I doubt he'll get it, the country is too awake and pissed off about what is going on (most of which McCain has publicly stated he supports, and wants to continue).
Posted by: Chris | June 2, 2008 7:58 PM
Really old white men are Democrats, but very hesitant to vote for a black man. Sad to say but Obama as Muslim has traction among the elderly. I sat in a waiting room at the VA with my father and heard the vets discussing the 'fact' that Obama took the oath of office on the Koran.
Obama will not be able to reach the hard core Fox viewer. However, I think the age issue will influence a significant number of the elderly in Obama's favor. Obama is tall and young and McCain is old, small and mispeaks a lot. Young people tend to forget that the elderly are acutely aware of the infirmities of age.
Posted by: Nat | June 2, 2008 11:04 PM
There was no shift in the black vote. They always voted overwhelmingly for Obama.
Literally true, but not 'accurate.'
The early (i.e. Nov Dec 07) polling showed something like a 50-30 preference for Clinton over Obama in SC among african americans
And even as for literal truth: when the actual voting started the AA vote has gone from something like 70-30 in the earlier contests to 90-10 in the latter ones.
Posted by: Kolohe | June 2, 2008 11:24 PM
He will try to win Latinos and Hispanics to his side, and there's some indication he might have success with that. And he will try to drive down turnout among working-class Caucasian Democrats. Will he be successful? Who knows. But he'll try it.
It's up to Obama's campaign to make sure that the race isn't run on narratives that make such a scenario more likely. It's up to Democrats who give a fuck about seeing a Democrat in the White House to do the grassroots work to make sure it doesn't happen.
Posted by: Unapologetic Andrew | June 3, 2008 1:17 AM
"The early (i.e. Nov Dec 07) polling showed something like a 50-30 preference for Clinton over Obama in SC among african americans" -Kolohe
I think the missing factor here is name recognition. It's not like Obama had been a huge figure in the African American community for years, he's fairly new no the scene. On the other hand the Clinton name belongs to the last democrat in the white house, and times were good. I think the name recognition factor alone explains her large support among that demographic prior to any campaigning being done in their states.
I think a lot of people forget about the name recognition factor. After all we had been paying attention to the race since early 2007, so of course we knew all the candidates and their positions. The fact is most Americans just aren't that plugged in, and I wouldn't be surprised if a lot of people getting polling calls in later 2007 only recognized one or two of the names. I bet you can guess which candidate was the one most people instantly recognized.
Which leads me to another point. I don't think the results would have been the same if there was just one national primary day on Feb. 5th. I think Senator Clinton is almost sure to have one the nomination if that had been the case on the strength of name recognition, and the presence of Edwards in the race. In my view the dynamic of the race prior to Iowa was that Clinton had built a strong coalition that was around 45-50% of the democratic electorate. The flip side of that coin was that was also not far from the ceiling of her support as well. Given her built in name recognition, high profile medial coverage as the front runner, and strategy of inevitability, most of the democratic electorate had already kind of decided if they wanted her to be the nominee. This suggests there was a descent portion of the electorate that was looking for a different candidate. I think it was clear that this second group was going to coalesce around someone, but before Iowa the questions were who, and how fast. If it didn't happen before Super Tuesday then Clinton was very like to have an unassailable lead. If it happened early enough then the not Clinton candidate would have a chance. If there had been a national primary there never would have been that consolidation, and Clinton would have cruised to the nomination.
The common wisdom is that the mistakes of the Clinton campaign was two fold, they ignored the caucus, and they didn't have a plan for after Super Tuesday. The more interesting question is why. I believe it was because they did not expect the race to so quickly devolve in to a two person race. Their expectation was that they would win in Iowa, and Obama and Edwards would be fighting it out for second place and the right to challenge her. I believe they assumed that portion of the race wouldn't be clear until right around super Tuesday, at which point it would be to late for the not Clinton candidate to put up much of a fight given the huge number of races happening on one day. What actually happened was that the not Clinton candidate became clear in Iowa when Obama actually won. It also meant that her challenger would be the candidate in the best position financially and organizationally to challenge her in the rest of the contest. Most importantly he also brought his own large demographic base with him.
Media coverage after Iowa which basically set the rest of the contests up as a two person race allowed Obama to start overcoming the name recognition advantage Clinton had. After Super Tuesday, with time to campaign in each state the race basically played out along demographic lines.
Probably the best place to see the effects of name recognition in the race would be to compare early voting vs. day of voting in the Super Tuesday states. For example Obama did much better in day of voting in California compared to the early voting. If Super Tuesday had been a week or two later, and if there hadn't been early voting Obama appears likely to have come close to winning the state.
Posted by: Jason K | June 3, 2008 3:35 AM
Demography normally is destiny in presidential elections. The electorate almost always splits along demographic and regional lines.
Over long periods, however, the bases of demographic splits change. From 1930-60, white racists and blacks voted Democratic; FDR's coalition, based on economic equity, trumped racial differences. By 1970, racial resentment began to outweigh economic considerations, leading to a generation-long split between blacks and working-class whites (aka "Reagan Democrats").
Also since 1970, gender- and religion-based based partisan preferences have become more pronounced. And since about 1990, we've seen a gradual switch from Republican to Democratic preference among more highly educated voters.
New registrants and higher voter turnout in this year's primaries have confounded pollsters, because many of the new voters are hard to sample. Younger voters, for example, are less likely to have landline phones or longtime addresses, making them invisible in traditional telephone polling.
The demographic wild cards this year will be: (1) Whether educated younger voters and minorities turn out in higher numbers than in the past; (2) whether Obama enjoys the same advantage among women and Latinos that Democrats have had; (3) whether over-65s skew heavily toward McCain; (4) whether economic status/security becomes as decisive as it was pre-1960.
If Obama has a racial disadvantage, it will be reflected in the votes of Latinos and over-65s, not working-class whites. WCWs who wouldn't support a black candidate have voted against Democrats for decades.
Posted by: allbetsareoff | June 3, 2008 8:43 AM
I pretty much agree with Jason's post, but I'm not convinced about California. I think the polls are just wrong about California, and showed it in predicting an Obama win in the primary. (I wouldn't trust them too much for the general, either).
I agree that almost certainly Clinton was receiving black support in early polls because the black respondents didn't know who Barack Obama was, much less his race.
I'm also glad to see that someone else is arguing that the big problem for Clinton was Edwards dropping out before Super Tuesday. Had he stayed in the election through mid-February, I think Clinton would be the nominee.
Posted by: Cal | June 3, 2008 8:47 AM
Obama's only mandate is he's black..he just needs to be black, but not too black.
Sadly, that has turned out to be the case. Over 90% of the blacks have thrown the Clintons under the bus...the Clintons, that by any measure have carried the torch of black interests for years...under the bus for an unknown relatively newcomer with little experience....BUT HE'S BLACK.
This doesn't seem to bother the Democrats. However, if whites were to make their choice based on people that looked like they do.......
Posted by: Anonymous | June 3, 2008 11:15 AM
Sounds like Obama wasn't surprised by this at all...from Thomas Shaller in today's Salon:
"I think Obama can win and I think the demographics are not stacked against him in such a way that he cannot win, but I actually think one of the strange ironies about the Obama general election candidacy is what we've learned from his primary candidacy, which is that demography is extraordinarily probabilistic. Not deterministic, but the fact that [the Obama campaign] wrote a memo in November and they got every state right except for Maine and Indiana, I think, speaks to the fact that it's hard to move people. So, the irony of his argument for general electability is that his campaign, by their own internal analysis, believes that people are predictable based on their demographic background. And his rhetorical argument is, I can move people from where they might be initially politically because I'm this transformative candidate. And I think there's a central paradox to Obama's candidacy, there's a couple of them, but if there's one from a demographic standpoint, I think it's that."
Posted by: Asinistra | June 3, 2008 3:11 PM