Composition of the Electorate Drives Poll Numbers

As I noted yesterday, the first burst of Big General Election Polls is upon us, and naturally enough, partisans have tended to focus on the data they find most pleasing, with Republicans celebrating the Gallup Tracking Poll that shows Romney up by two points, while Democrats prefer to tout a CNN/ORC survey showing Obama up by nine.

But what, precisely, explains the wide difference in the results from these two generally credible public opinion firms? Is is just statistical “noise” that should go away over time?

Ron Brownstein of National Journal has a pretty persuasive answer:

Four recent national polls, including three released in the past 24 hours, generally show the electorate dividing between President Obama and Mitt Romney along lines of class, gender and race familiar from the 2008 race.

The surveys-from ABC and the Washington Post; the Pew Research Center; CNN/ORC; and the first Gallup tracking poll, diverge in their overall results. The first three polls show Obama leading by seven, four and nine percentage points respectively; the first Gallup track placed Romney up by two percentage points.

But the Gallup track, which is conducted among registered voters, has a sample that looks much more like the electorate in 2010 than the voting population that is likely to turn out in 2012: only 22 percent of the Gallup survey was non-white, according to figures the organization provided to Emory University political scientist Alan Abramowitz. That was close to the non-white share of the vote in 2010 (23 percent), but in 2008, minorities comprised 26 percent of all voters, according to exit polls; the Obama campaign, and other analysts, project the minority share of the vote will increase to 28 percent in 2012. In its survey, Pew, for instance, puts the non-white share at 25 percent.

I know some people think minority voting will be down in 2012 because it’s not a “historic” election like 2008 or because minority voters are disappointed in Obama, but there are few objective signs of that among African-Americans, and the Hispanic share of the electorate is steadily growing. Moreover, turnout for any voter category is much more likely to resemble the previous presidential election than the previous midterm, and historically presidential electorates are much younger and less white than midterm electorates.

As we get closer to the election, of course, pollsters will deploy all sorts of devices to estimate who will actually vote, from likely voter “screens” aimed at separating serious from casual voters to “weights” assigned to this or that demographic based on more general turnout assumptions. Now, though, it’s a bit of a crapshoot. As Brownstein concludes:

Even with their modest variations, these four surveys paint a similar picture. Obama is largely holding the minority and college-educated white women who comprise two pillars of the modern Democratic base (along with young people.) But he is facing erosion among blue-collar white men and struggling to maintain even his modest 2008 support among the two swing quadrants in the white electorate: the college-plus white men and non-college white women.

For the moment, that division of allegiances is enough to provide Obama an overall advantage (he would lead slightly even in the Gallup track if the minority share of the vote was adjusted to its level in 2008).

For all the talk we are going to hear about undecided voters, it may be the demographic composition of the electorate that matters most if the contest is as close as looks likely right now.

Ed Kilgore

Ed Kilgore, a Monthly contributing editor, is a columnist for the Daily Intelligencer, New York magazine’s politics blog, and the managing editor for the Democratic Strategist.