Friday, April 24, 2009

I felt better about the poll before the pollster defended it

There's been much to-do in Colorado of late over recent PPP polls which show both Gov. Ritter's and Sen. Bennet's approval ratings to be a mere 41%. Now, PPP does robocalls, which are sometimes considered a tad methodologically iffy. The Bennet poll curiously calculated 75% name recognition for Sen. Bennet, which seems high considering he was just appointed a few months ago and had a name recognition of about 1% (I'm guessing) prior to that. Coloradopols, in particular, was saying that this poll just had to be wrong. PPP responded by saying:
Some blogger in Colorado whacks our poll today, saying there's no way 75% of voters in the state really have an opinion about Michael Bennet.
I think that's probably true, but it doesn't mean there's anything wrong with our poll.
Um, actually, that means that roughly everything is wrong with the poll. If we're accepting that 75% is way too high, then the sample is clearly unrepresentative of Coloradans, which calls the whole approve/disapprove thing into question.

No comments: