PDA

View Full Version : Hey, you know what often overstates Obama's support? (Conservative Effect)



megimoo
10-21-2008, 01:35 PM
The Obama Glee Club at the Politico took a 5 minute break from singing his praises today, to report on a curious fact: exit polls tend to, get this, OVERSTATE Obama's support.

snip

Anyway, I read the Politico so you don't have to (you lucky, lucky bastards) and here is an interesting nugget buried in their article:

In theory, exit polls should match election results. But for all the care that goes into conducting accurate exit polls, errant results arenít completely uncommon. Respected polling analyst Mark Blumenthal found that during the Democratic primaries this year, preliminary exit polls overestimated Obama's strength in 18 of 20 states, by an average error of 7 percentage points, based on leaked early results.

The reason? Obamaís supporters were younger, better educated and often more enthusiastic than Hillary Clinton's, meaning they were more likely to participate in exit polls.


See, Politico has finally figured out that in polling, there is this thing called "selection bias". There is also the related variable called "get the hell out of my face and leave me alone you nosy bastards." And exit pollsters, God love 'em, just don't seem to be able to figure out how to quantify this.

How bad is it? Check out 2004:

In the end, Kerryís vote was overstated in 26 states. The same was true for Bush in four states, according to a detailed post election analysis by Edison Media Research and Mitofsky International, who administer the exit poll for the NEP.

Funny how these errors almost always seem to overstate Democratic support all over the place, isn't it?

Now, this raises a question that Politico doesn't address. If an "enthusiasm gap" is responsible for skewing exit polls, why isn't an "enthusiasm gap" responsible for skewing pre-election day polling?

And if there were ever a year when one would be looking at polls and pondering "enthusiasm" doesn't it seem like that would be a metric that would be favoring Obama right now? After all, your own article says his voters have a history of getting all up in peoples face and "bragging" about their vote.

snip

Anyway, make of this what you will. An average error of 7 points for the One is amazingly high, though, given that usually the margin of error is around 3.5%.

Meanwhile, as Politico begins hedging its Obama bets in the wake of tightening polls, Peggy Noonan has fired her lawn boy for failing to cut the grass in the counter-clockwise swirls that gives her home-field advantage in her gated communities croquet tournament.

http://ace.mu.nu/archives/276120.php