To skew or not to skew, that's the question.

Yes, the polls are skewed.

No, the polling agencies didn't skew them.

Any poll with an R+3 result won't be reflective of the vote in November, or even of likely attitudes next week. In that sense, the poll IS skewed. In another sense, it isn't. In the sense that the poll is a typically tabulated snapshot of a particular period of time, the poll isn't necessarily skewed. Additionally, there's the timing of the poll. If a poll is taken during an atypical period of time in an election, the result will look skewed, ie it will be an outlier. And this is where some liberals have an argument. Polling experts will say it isn't the job of a pollster to avoid atypical snapshots. But then, most pollsters do running averages precisely to avoid that problem. So the poll defenders are contradicting themselves to some degree.

The problem here is one of semantics, and the eagerness of some liberals to beat up on other liberals for using a word that makes sense in everyday language, but which bad behaving liberals take to mean something else for their own purposes (normally it seems in order to to look smart or sensible to the mainstream).

Liberals beating up on other liberals for pointing this out are being obtuse, and creating division where there need not be any. Many of the liberals engaging in this bad behavior are community leaders on this website, and they should stop it for the sake of the progressive agenda. They need to put their egos aside and start treating their fellow liberals more respectfully.

Liberals who use the word "skewed" aren't all saying that there's some conspiracy to skew the polls for Republicans. Some may. But the majority of them are merely saying that in everyday language, the current poll results are being "skewed" by an atypical post-debate blip for Romney, and that some pollsters are timing their poll sets in such a way that they don't integrate all the data that could tether these results to reality. That doesn't mean they're conspiring, just that their results are being skewed from a broader reflection of reality by unfortunate standard practices. And yes, some liberals may have a good argument that some pollsters are behaving badly - even if it isn't the ones everyone's arguing about right now.

From Nate Silver:

If voters are feeling better about Mr. Romney after the debates, they might also be inclined to identify themselves to pollsters as Republicans.

It is probably also the case that Republicans won’t actually have a 5-point party identification advantage in the exit poll on Election Day. But it isn’t the pollster’s job to project what will happen on Nov. 6. (That’s my job, instead!)


First, is it really likely that Mr. Romney leads the race by 4 points right now? The consensus of the evidence, particularly the national tracking polls, would suggest otherwise.


The other valid line of inquiry concerns the timing of the poll. The Pew poll was conducted from Thursday through Sunday, although more of the interviews were conducted in the earlier part of that period.

Your Email has been sent.