Politico notes that polls suggest the American public is a few hospital beds short of a care facility. How can people simultaneously love the public option, hate the public option, love Medicare, and hate government intervention? And why-oh-why do 8 percent of New Jersey residents seem to think that Barack Obama is the Antichrist?
Hey, these questions sound awfully familiar!
Regardless, the big question is whether these polls are in any way actually useful. And my answer is... it depends. As big-picture trend trackers, I think you can get a sense of swings in public mood, especially when you look at poll aggregates. If enough polls show a similar shift in public opinion, they're probably pointing to real trends. And in cases like health-care, where polls are all over the map, I think polls can suggest the considerable volatility of public opinion.
But as day-to-day debate movers, I think journalists put way too much stock in the minute-by-minute gyrations of the horse-race polls, and Beltway types get too wrapped up in the minutiae of organizational surveys. To a large degree, this is understandable: If politics is a game, polls are the scoreboard. But it's easy to forget that the scorekeepers—the public—aren't necessarily following rules that make any sense or are at all consistent from scorekeeper to scorekeeper. As pollster Celinda Lake tells the Politico, "Politicians need to understand that voters are very comfortable having mutually contradictory views."