Pollster Mark Blumenthal has an interesting bit on how selection bias may be skewing the polls in Iowa:

The point here, in case it is not obvious: Non-response bias may have exaggerated the percentages of younger (under 45) caucus goers the 2004 Iowa entrance poll (something I wondered about a month or so ago). And since I’m assuming that age is strongly related to having attended a caucus in the past, the entrance poll estimate of the number of caucus newcomers in 2004 may be exaggerated as well.

Basically, older caucus-goers are less likely to talk with younger polltakers, which means that candidates like Dean or Obama that tend to have more support from younger voters look stronger in the polls than they actually are. The way in which the Iowa caucus system works heavily favors older voters who tend to be a part of the caucuses year after year. That’s why the results in Iowa may be vastly different from the way they’re portrayed—candidates like Obama don’t tend to have the same appeal to the typical Iowa caucus-goer as they do to the typical voter.

I’d be willing to guess that the story coming out of Iowa will be the resurgence of Hillary Clinton. Clinton’s made some major missteps in the last few months, but that’s made it easier for her to steal a page from her husband’s playbook and become the Comeback Kid—and with the latest CNN/WaPo poll putting her comfortably ahead in New Hampshire helps in painting a picture of a campaign on the upswing.

As for the Republican race, the effect of potential poll bias remains to be seen. It could be that Huckabee’s appeal to evangelicals matches with the target profile of Iowa caucus-goers. Or it could mean that Romney, McCain, or Thompson could get a boost from older voters. The race has been up in the air for weeks now, and the potential for change is so great that it’s simply impossible to make a worthwhile call.

Polling is never an exact science, which is why it only has limited utility in a campaign. A candidate like John Kerry who was in the single digits at this point in 2003 can suddenly sweep the nomination. A candidate like Howard Dean that appears unstoppable can end up losing big. Part of the fun in politics is in contests like this where anything could change. (Although it’s much less fun for those on the inside of a campaign.) The first rule of polling should be not to trust polling, as it’s as much of an art as a science. It’s quite possible that everything we’ve heard about the race in the last few weeks may be rendered moot when Iowans actually go into their caucuses and pick their candidates.