I can see why it was so easy to con the sheeple on the “conservative” side of U.S. politics. They don’t care what the truth is.

[shrug] It’s what I think of polling and pollsters. I have no doubt that at least some pollsters try to get it right. I remain unconvinced that they can succeed reliably. I also remain unconvinced that there is anything honest about political polling — by anyone, on any side, of any issue whatever. Some of them bend the results on purpose, others do it unintentionally — but one way or another, they all do it.

I don’t trust any political poll, and I don’t recommend anyone else do so either.

There are plenty which make serious but honest mistakes. Certainly, some polling orgs get clever with sampling, ordering, and wording to get their desired results. However, these dishonest manipulations are far more often designed to produce “newsworthy” results than a particular partisan bias. Subtle changes in wording which result in dramatic swings in polls which follow the existing media narratives are a classic example.

Several of the polling organizations are actually academic endeavors, where getting the methodology right is actually the central goal.

Nate Silver is popular (and respected) because he is extremely competent at statistical measurement. He has also quickly gained real expertise in the subtleties of polling and the use of surveys and polls to actually make real measurements.
It is notable that he started off in the world of predictive baseball statistics, where being right is all that counts.

PS: Nate had a post a while back on
How to Poll on the Public Option.
Beyond being interesting, I think it also illustrates Nate’s essential geekiness and concern for ‘doing it right’.

Has anyone ever been polled by this group?

That question came up at dinner last night, and the answer is best described here. Given the degree of ingenuity displayed (always assuming Nate is right) I wouldn’t rule out other demonstrations of brilliance. Still, the “thumb on the scales” hypothesis can’t be ruled out. Ockham’s Razor rather than Hanlon’s.

Either way, discovery will tell. Which is a good reason to expect that this will never actually go to court — too much other dirty laundry (e.g. the objectives requested by the sponsors) would come out.

Again: so what? What’s the difference between data “pulled out of their asses” and data slanted using a skewed party affiliation, or selective sampling, or loaded questions, or any of the other ways that pollsters have learned to use over the years? It’s all fraud, and all sides do it. If you get hot and bothered over this case and not over others, you’re being hypocritical.

D.C. Sessions: We have a call into our attorney on this…

*subdued cackle* – they “have a call into”? So their attourney doesn’t take their calls straight away? Maybe a receptionist makes a note, and later in the afternoon, when the serious work for the day is done, their attourney will look through the notes and decide whether to call them back – or whether to put in a quick nine holes on the local course?

I like the fact that Nate has asked for people to contact him if they were polled by this group and have some story. At first I thought this was dumb because the ‘story’ would be pretty useless. But then I realized his true motivation.

Has anyone ever been polled by this group?