Polls: "Party ID" is to 2012 GOPers as "Land Lines/Cellphone" was to 2004 Democrats.
There seems to be certain things that are pervasive in the right-wing-o-sphere at the moment, to allow good GOP voters to hand wave or happy-talk bad polling numbers away.
The most prominent is this idea that all these polls (that show Democrats and Obama ahead) are wrong because the proportion of self identified Democrats vs. Republicans is disproportionately high according to some apparently arbitrary standard.
Here's an example from today, freepland - thread about Fox News polls showing Obama up in VA/OH/FL:
http://www.freerepublic.com/focus/f-news/2933591/posts
Virtually every post is "what about party ID?" "Huh? Thought so! Bad poll!"
Here is the example of the "reasoning" - freep threads on this topic seem to often cite articles from breitbart on the "voter party ID breakdown" issue:
http://www.breitbart.com/Big-Government/2012/09/13/poll-obama-up-5-in-d-11-poll
Not to get into the merits too much but it's my understanding it's fallacious reasoning because there's no reason voter ID is a fixed property of the public and should be sampled with each poll and not "corrected". The correction is just doing more polls and sampling.
I'm just curious who is putting this out, I figure it has to be either Fox, Rush, or Drudge. Does anyone know?
The other big one is to cling to "Reagan was behind right up until election night, THEN look what happened!" which I understand to not exactly be true, not to mention that Romney is not exactly Reagan and Obama is not Carter.
I'm even seeing that old chestnut, the Bradley Effect, being trotted out, even though it didn't really materialize for Hillary or McCain in 2008.
I think cell phones WAS a confounding factor of polling, just not as big a one as I would have hoped in 2004, and the fact that pollsters are calling cell phones now is a strong indicator of that.
Oh well, hope springs eternal and all that.
= new reply since forum marked as read
