On the eve of the final pair of Democratic primaries, a new poll of South Dakota Democrats gives Hillary Clinton a 26-point lead over Barack Obama – a surprise, at least to me, in a state that had the early makings of an Obama stronghold.
The survey, from the independent New Hampshire-based American Research Group, shows Clinton pounding Obama by a 60-34 percent spread. The only previous poll in the state, from about a month ago, had Obama up by 12. Because every other state around South Dakota previously went overwhelmingly for Obama, there had been little expectation in the national press that Clinton would win there.
Not surprisingly, the ARG poll is being greeted with some skepticism. But in a phone conversation a few minutes ago, Dick Bennett, ARG’s president, told me that the assumption that South Dakota is Obama country has been flawed from the start.
"The numbers we got are consistent with what we’ve seen in other states," Bennett said. "The groups that are usually supporting her are supporting her. The people who live there look more like Clinton voters than Obama voters."
Bennett pointed out that South Dakota is a closed primary, meaning that only registered Democrats can participate: "So the turnout is going to be lower. The bonus votes that Barack Obama has been getting among independents? They’re not going to be able to turn out. When you look at the other states that are like this, Hillary Clinton has done very well."
Bennett also noted South Dakota’s large Native American population, its high median age and low number of college graduates, factors that also work in Clinton’s favor. And then there’s the gender issue.
"I talked to our supervisors [at the polling center] on Saturday and asked them what was happening, and they said that it’s women – they’re just not ready to let Clinton go," he said.
Bennett said he’s surprised that anyone would be surprised by his findings. When I pointed out that every states that borders South Dakota – North Dakota, Minnesota, Iowa, Nebraska, and Wyoming – has sided with Obama (often by overwhelming numbers) he replied: "Those weren’t primaries. Those were caucuses."
The results from Nebraska’s non-binding May 13 primary suggest there is something to this line of thinking. In that primary, with turnout three times higher than in the state’s February caucuses, Obama won by just two points. In the caucuses, his margin was 36 points. Still, he did win. Could he really be trailing by 26 points in South Dakota?
"Could it be the sample size?" Bennett asked. "I guess. But we broke the sample up over two days, and we didn’t see any great shifts."
If the numbers do prove true, Bennett said it’s not a sign of any broad shift toward Clinton and away from Obama among Democrats, since they are both performing as well (or poorly) as they usually do among various demographic groups. But it would mean that national political observers (myself included) have badly misread the demographics of South Dakota.
A big South Dakota win for Clinton could, depending on the outcome in Montana, also shift the popular-vote math in her favor, allowing her to claim a victory without counting Michigan (although she would still have to exclude caucus states like Iowa, in which no official popular vote tally was maintained but where more Democrats clearly preferred Obama).
Follow Steve Kornacki via RSS.
