Thursday, September 27, 2012

UnSkewed Polls Shows Totally Skewed Nature of US Conservatives

The most recent buzz going on right now south of the border is the attempt by... lets call them right-leaning politicians, organizations, and zealots, far more polite terms than I usually think to call them, to "unskew" polling data put out by major polling organizations so far in the race for the White House between incumbent Barack Obama, and main challenger Mitt Romney.

The status of the race so far is pretty well trending towards Obama, according to Nate Silver at FiveThirtyEight.com, one of the best poll analyzers in the world, in my opinion at least. His projection system for the current presidential race is one that is based on sound, factual fundamentals, including polling averages, partisan identification, likely voter models, and other variables. If you wanted to do projections, this is one of the main guys to follow.

Then we come to Dean Chambers, owner and I assume the person doing all the "unskewing" of polls at UnSkewedPolls.com. This guy is a piece of work in his own right, and the links to Dick Morris and Rush Limbaugh at the bottom speak towards the amount of bias you're about to have blasted in your face.

I won't go into too much detail about Chambers' methodology, just the basics: Chambers saw a poll back during the summer where Obama was leading Romney, disagreed with it, and saw that the poll had more identified Democrats than Republicans, and thus concluded that this was the reason for the "skewed" nature of the poll. Chambers then went out and found that Rasmussen Reports, a polling company known for some slightly Republican-leaning numbers (though they're still considered as one of the more credible polling firms in the US), showed different (read: more favourable to Republicans) partisan identification. So what Chambers does now is filter all polls through those Rasmussen numbers, a process which even Scott Rasmussen has said it is entirely stupid:
Even the founder of Rasmussen Reports, whose surveys often show higher Republican numbers, cast doubt on Chambers' methods: Scott Rasmussen told BuzzFeed in an e-mail that "you cannot compare partisan weighting from one polling firm to another."
 
"Different firms ask about partisan affiliation in different ways," explained Rasmussen. "Some ask how you are registered. Some ask what you consider yourselves. Some push for leaners, others do not. Some ask it at the beginning of a survey which provides a more stable response while others ask it at the end."
 That is a pretty easy concept to grasp, or at least you'd think it would be. There is no unified data set that all pollsters share, each and every one have different ways of gathering that data and using it. The only data set everyone shares are the actual numbers we see on election day.

Yet here we are, and someone flies in the face of simple facts. I simply think Chambers and many other Republicans who are supporting this crackpot are in denial. Obama beating out Romney!? Never! Our gold-plated candidate with zero common touch and a tendency to not just stick his foot in his mouth but right through the back of his head, he would never be rejected by Americans for that socialist Muslim (Madonna is our source for that! Shes credible!) liberal pansy.

The cognitive dissonance that many American conservatives must have each day of their lives would probably kill me. The thing is, Chambers is probably just like me - some guy running a piddly little site on politics who has an interest in polling and the outcomes of elections, with the main differences being ideology and success (his site is bigger than mine, I will admit).

Those differences shouldn't account for the amount of stupidity UnSkewed Polls causes though. Never on my site have I called out a poll for being skewed simply because I don't like how it looks. I've called ones out for bumping against trends - I'm looking at you Ekos, Nanos, and, ugh, Campaign Research - but that's grounded in reality, and I do it no matter which way it goes, favourable or unfavourable to the Liberals.

Dean Chambers would probably disregard the law of gravity if it was pro-Obama in any way.

No comments:

Post a Comment