Thursday, October 23, 2008

Biased Polls: Party Affiliation (Part II)

DJ Drummond has another post on biased state polls. This one revisits an issue I have written about in the past--party affiliation.

Just to be clear, party affiliation isn't something that changes frequently during a persons life. One switch is unlikely. Two is very rare. Just think about your own personal party affiliation. Did you register with one party for one election, the other the next, independent for a third, and back to the original party for a fourth? I'm confident that the answer to that question is an emphatic no.

And I am not talking about voting for a candidate from the other party; that happens much more frequently. One can imagine a bitter Clinton supporter voting for McCain in protest of how Obama treated her during the primary. But this person is still a Democrat and will remain a Democrat in all likelihood.

With that in mind, look at some of the shifts in party affiliation Survey USA polling is using this year.
Pennsylvania: D+5 in 2006, SUSA using D+19, 15 point variance
Indiana: R+14 in 2006, SUSA using R+1, 13 point variance
Nevada: R+7 in 2006, SUSA using D+6, 13 point variance
Colorado: R+3 in 2006, SUSA using D+9, 12 point variance
Iowa: R+2 in 2006, SUSA using D+10, 12 point variance
Virginia: R+3 in 2006, SUSA using D+9, 12 point variance
Ohio: D+3 in 2006, SUSA using D+13, 10 point variance
Missouri: R+1 in 2006, SUSA using D+7, 8 point variance
North Carolina: R+1 in 2006, SUSA using D+5, 6 point variance
In every case, the variance is in one direction--towards the Democratic Party. And just look at some of the differences. Indiana was +14 Republican in 2006 and now suddenly it is only +1? Pennsylvania is +19 Democratic now? That stat is always a balance between the urban left of Philadelphia and Pittsburgh and the rural right of every other county. Ohio as +13 Democrat in what has always been a swing state. None of these are justifiable. Drummond comments:
I've looked at the publicly available records on historical election participation, 2008 new voter registrations, and the Census information on these states, but I can find no valid reason for such large and arbitrary changes in political affiliation weightings. I would therefore submit that the models being used for many of the state polls have design flaws, which threaten the credibility of their published results.
For the fifth time or so in the last few days, let me make this clear. Don't trust the polls. Dave at Hedgehog is panicking about an Ohio poll that shows Obama up by 14% in Ohio. Ludicrous. And when you look at the internals, now you know why.

No comments: