Why do the national polls tell us one thing and the battleground state polls tell us another?

Many of us have been puzzled by the conflicting signals sent out by national polling (good news for Romney) and state polling (mostly good news for Obama). The disparity has led to speculation that Romney might well win the popular vote but lose in the Electoral College. But the normal odds against such an occurrence make us wonder whether something is amiss with either the national or the state polling.

Something does seem amiss with many of the polls of battleground states. In these polls, as we and many others have noted, the gap between Democratic and Republican participation in the samples used in these polls implies that Democratic and Republican turnout on election day will roughly mirror what happened in 2008. But it’s difficult to believe that this will be the case.

Some have defended the mix of Democrats, Republicans, and Independents in these polls on the theory that since 2008, a portion of right-leaning voters who once identified themselves as Republicans now identify themselves as independents. Thus, the theory goes, Mitt Romney isn’t really clobbering President Obama among true independents; he’s simply riding his comparative popularity among disillusioned Republicans. And, if true, the theory would explain why Republican representation in poll samples remains low.

However, as Jim Geraghty points out, the movement of disillusioned Republicans to Independent status cannot explain the level of Democratic representation in poll samples. And to the extent that such Democratic representation mirrors 2008 Democratic turnout, polls imply that there has been no disillusionment or diminution of enthusiasm among Democrats. Yet, it is implausible to believe that, during the four years of the Obama administration, the only disenchantment has been among Republicans.

Accordingly, Geraghty investigated whether, in fact, Democratic representation in the samples of polls in key states implies Democratic turnout at 2008 levels. That is, he examined, key state by key state, what percentage of the final vote identified themselves as Democrats in 2008, and asked how that compares to some recent poll samples. The states he investigated were Colorado, Florida, Iowa, Michigan, Minnesota, Nevada, New Hampshire, North Carolina, Ohio, Pennsylvania, and Virginia.

Sure enough, Geraghty found that in many cases, pollsters do implicityly project that Democrats will make up more of the electorate in 2012 than they did in 2008. This is particularly true of PPP, Survey USA, and Gravis Polls.

What about the national polls? Geraghty finds that none are presuming 2008-level turnout for Democrats, at least without throwing in Democrat-leaning independents.

Thus, he concludes that “it’s not all that surprising that Romney’s showing a lead in most national polls while trailing in a bunch of the key swing states.”

Recommend this Power Line article to your Facebook friends.

Responses