Look at how F#$ked up USA/Today Gallup made their Likely voter sample for 18-29 year olds:

[Pirate Smile]

On Likely Voters and the Long Tail

Mark Blumenthal has a good article up at the National Journal about that controversial USA/Today Gallup poll released last week that showed John McCain with a 4-point lead in its likely voter sample. In turns out that only 10 percent of the likely voter sample consisted of voters between ages 18-29. By contrast, teens and twentysomethings represented somewhere between 16 and 18 percent of the electorate in 2004. And that number is likely to go up rather than down this time around, as youth turnout in the primaries increased by 52 percent as a share of the Democratic electorate.

Now, I appreciate that Gallup is willing to disclose so much about their methodology -- it certainly opens them up to more criticism. Having said that, winding up with a sample that understates the youth vote by perhaps 30-50 percent is pretty much prima facie evidence that something has gone awry. Indeed, I'm really not a fan of the Gallup likely voter model at all.

http://www.fivethirtyeight.com/2008/08/on-likely-voters-and-long-tail.html#comments

[Mz Pip]

1. I thought there was something

off with that poll since it didn't come close to any other poll.

Now the youth just have come out in November like they did in the primaries and we'll be celebrating on Jan 20! :bounce:

[Whisp]

4. hold on to your jumpies,... they just need the polls to Appear close

so they can steal another one.

yes, it can come to that AGAIN.

[KittyWampus]

16. check the days the poll it conducted. If it's Fri /Sat, youth voters not home.

[featherman]

2. "Likely Voters" methodology is always subjective. Need to check the tabs, as you did.

Unfortunately the media idiots just run with it like the dumbest of dumb clucks. Why are they pulling such hefty paychecks if they are so ignorant about what is so basic in what is supposed to be their field of expertise? Really.

[sandnsea]

3. They always manipulate the polls

Over-sample in the south, rural zip codes, age groups. They can pretend they don't, but it's obvious they do. The more information we supposedly have, the less anybody really knows.

[Gman]

5. Wasn't it the CEO/President (whatever) of Gallup that said in 2004

that he'd do whatever it takes to get Bush "reelected" (and I use the term loosely). Wasn't he the state chair in Ohio? And we all know the fiasco that was Ohio in 04. Gallup polls have little credibility.

[Fleshdancer]

6. I think that was the guy from Diabold. n/t

[Gman]

7. Yup, you're right

but I believe Gallup, or the Gallup family is heavily involved in GOP politics.

[zbdent]

8. when's the last time you heard of 18-29 yos with land lines?

boy ... skew much?

[KittyWampus]

17. and home on Fri and Sat when that last poll was conducted

[Phx_Dem]

9. Last week I heard Larry Sabato,

director of the Center for Politics at the University of Virgina talking about this on XM's POTUS 08 channel. He said this poll was completely bogus and every pollster and politico in DC was talking about it and saying the same thing.

He went on to say that out of every 20 polls there is always one that is completely off the mark, and this poll is that one.

The MSM truly is a disgrace. Apparently, everyone in DC is talking about this poll being total bullship, while the MSM morons are squeeling with delight and touting it as a sign of McBush pulling ahead. You can't tell me they hadn't heard all the policos saying this poll was garbage. They just didn't care. They want a horse race, and damnit they're gonna make one up if they have to.

L.O.S.E.R.S.

(We need a Loser smiley)

[Alexander]

10. K&R - They are fudging the polls to try to steal it...not going to work this time!!!

:thumbsup:

[phrigndumass]

11. Nate Silver is God :)

Make his website a daily read, it's the best way to learn what's going on behind the numbers.

http://www.fivethirtyeight.com

:thumbsup:

[Pirate Smile]

12. kick

[kristyt]

13. Fuckers...That's no accident

Even their shitty sample had Obama winning they just changed it to get headlines and make it race. Some USAToday subscriber should sue and do discovery. Stop this shit at Gallup and USAToday pronto.

[barack the house]

14. It would explain the gap between the huge lead in electoral college and the polling numbers.

[TexasObserver]

15. Gallup knew exactly what it was doing when it gamed the poll that way.

The mix of voters deemed an appropriate sample is the whole ball game. If you know how many of each group are polled, you can figure out easily how many of the overall group will be for a given candidate.

They oversampled old farts and under sampled younger voters. The result was entirely predictable, and that is how they created a poll where McCain does well.

Gallup is in bed with half the major corporations in the world, "partnering" projects with them.