Wednesday, November 07, 2012

Ace:


Remember: John Kerry Easily Won the Exit Polls In 2004, Leading to the "Seven Hour Presidency"

So the Exit Polls are being leaked now. Drudge headlines "Race tight." You'll be seeing more leaks, with real numbers.


But remember: These are just polls. Surveys. And they skew more Democratic than the actual polls do.

In 2004, Kerry was picking out new curtains for the Oval Office because the exit polls showed him destroying George Bush.

But then people started noting: In the Pennsylvania exits, there were 22% more women surveyed than men.
Could that be right?

No, it wasn't right. The exit polls had a massive Democratic skew for several reasons. The people conducting the polls were young college students, for example. Older voters avoided them (figuring they would look down on their vote) and they, in turn, avoided older voters (because older voters are square, man!).

Basically, exit polls have a very high level of self-selection. They are not random. The pollster decides who looks like a nice young Democratic-leaning voter who she might want to date, and chats him up about his vote. And the voters self-select, because they don't really feel like being judged for voting Republican by a 19 year old sociology major.

Also ignore the secondary findings from the exit polls -- stuff like "Are you better off than you were four years ago?" or "What's the top issue of this campaign?"

All of these results are skewed by a far-too-liberal sample.

Though I might hazard a guess that "Tight race" means "Romney winning" in actual votes.


I See Now... Drudge is saying the exit polls show Obama winning Ohio, Iowa, others.

Silly. These exit polls also said Kerry won Ohio. He did not win Ohio.

Completely Inaccurate: The skews are huge.
Four years ago, on Pollster.com, we gathered all of the official tabulations posted as polls closed and extrapolated the underlying estimates of the outcome for each state. When later compared against the final vote counts in each state, we found that the initial estimates had overstated Barack Obama's margins by an average of 4.7 percentage points.
It gets worse. That's just the average.. In some states, they overstated Obama's margin of victory by sixteen and a half points.

They also went the other way in some cases -- in a couple of McCain states, they overstated his support by 5.5 points.

I think the reason for this is that conservatives in blue states and in swing states just don't want to explain their vote to a sociology major.

 

No comments: