Gallup In The Tank?

wizbang ^ | Published: October 27, 2008 - 10:31 AM

[Perdogg]

Back in 2004, I jumped pretty hard on John Zogby. Zogby did two things which I considered, and still do, to be unacceptable conduct for a pollster. First, was that Zogby flat-out called the election for Kerry back in May of 2004, a prediction he hung onto through the rest of the campaign. The second reason was that Zogby started mixing results from his telephone polls with his online polls, which invalidates the results from both methods. I would also point out to the reader that in 2004 and 2005, I was unhappy with political affiliation weighting at the time, and had adjusted my own expectations by reversing the bias from polls. My point is that even four years ago I was challenging poll methodology when it deviated from NCPP guidelines, and even if Zogby is publishing prettier headlines now, that does not change my wariness from past experience. I will challenge any behavior at odds with valid practices.

[Perdogg]

ping

[snowrip]

Oh, say it ain’t so. /s

[tomkat]

Rhetorical question ?

[The_Victor]

Good analysis. Be sure and read the whole article.

[snarkytart]

It is good. I have hope we can and will win this.

[shhrubbery!]

I have never trusted Zogby. Ever. Including when his news was good for our side.

Rasmussen strikes me as an honest man, but even he seems to go with the flawed CW for adjusting raw numbers against party registration.

Which pollster predicted the GOP tsunami in '94? No one, as far as I remember.

Who predicted Mario Cuomo's defeat at the hands of unknown George Pataki? No one, as far as I remember.

What most pollsters are missing this year is the hottest intensity on the part of conservative voters, EVER.

People I know who couldn't get terribly riled up over the prospect of a Gore or Kerry presidency CAN'T WAIT to vote this year.

[ncalburt]

Gallup was in charge of the slanted Dem only questions peddling in the second debate by almost Dem VP candidate Tommy Brokaw !
The fact that Gallup refuses to give up their big fantasy massive youth vote and keep the 7 point Obama mix buried in there polls despite what there own poll about there is no huge youth in 2008 says it all. We should demand the end of the polling 30 days before an election and ban exit polling after the flat out bold faced lying involving PUSH polls this year ! Scottie Ras is just as slanted and bias as Gallup !

[GOPJ]

Looks like they're in the tank.

Baghdad Bob had an excuse - there was a boot on his neck. What's the MSM's excuse?

[JaneNC]

of course

[gov_bean_ counter]

Will these polling companies take responsibility for their inflated expectations when Obie looses and cities burn?

Sorry, silly me...

[greyfoxx39]

From http://www.redstate.com/diaries/swamp_yankee/2008/oct/27/we-must-spread-good-news-ourselves-first-fou/

Posted by: Swamp_Yankee

Monday, October 27, 2008 at 08:57AM CDT

Is McCain surging or are pollsters tightening up their models to salvage some credibility? Either way, the MSM is desperately trying to supress Republican enthusiasm and smug elitist Beltway conservatives are eating their baloney. We must spread the truth and generate energy ourselves.

Monday polling is historically bad for Republicans. Much of the data is based on Saturday or Sunday, when working famililes and people of faith are less inclined to answer the phone. Polls based on weekend data tend to favor Democrats. That being said, the first four polls of Monday show McCain has:

• shaven 3 points off Rasmussen - Obama +5
• shaven 5 points off Zogby - Obama +5
• shaven 1 point off GW/Battleground - Obama +3
• shaven 3 points off biased Kos pollster Research USA - Obama +8

Word is the McCain internals had him down four over the weekend. If McCain can get it within a couple of points, and he has over a week to do so, McCain can win. The majority of the battleground states are states that Hillary whipped Obama despite polling. If the polling/election trends for the general mimic the polling/election trends for Democratic Primary, McCain wins Ohio, New Hampshire and Pennsylvania.

Keep the faith. Keep fighting. Spread the good news.

[bobsatwork]

From Ann Coulter: Reviewing the polls printed in The New York Times and The Washington Post in the last month of every presidential election since 1976, I found the polls were never wrong in a friendly way to Republicans.

[GOPJ]

What Gallup has forgotten is the only product they sell is “trust” in their numbers. When that’s gone - they’re finished.

[irish guard]

My favorite on Zogby is that on the day of the election, he looks at the first round of exit polls before releasing his final poll on the election. Sure exit polls are flawed, but that hardly lends credence to his accuracy if he’s peeking.

[texmexis best]

Here is what I think is happening:

There are many pollsters who are having a great deal of difficulty with their old models of voter behavior. Many people are lying to the pollsters, the PUMAs, Republicans who are entertained by leading them one and a growing group of independents and conservatives who are scared to tell anyone that they are voting for McCain. The last group feels that they may be subject to retribution by Obama supporters if they tell their political leanings to a stranger.

They have no signs out, no car stickers, and they keep their mouths shut at work and in the neighborhood because they love their house and don’t want to see it graffite’d and they love their car and don’t want to see it keyed.

The pollsters as well as the party apparatus are well aware of what is going on. There are reports that the Obama camp feels that the Bradley Effect will be in the 6 to 7% range. If so, Obama is toast.

One of the oddest questions was added to one of the polls last week. One of the big pollsters asked whether their pollees would tell the truth to pollsters. 84% said that they would never lie, 16% said that they would:

1. 16% lying is enough to blow any polling into the 0% confidence level.
3. Of those 16%, how many would never tell the truth to a pollster? Probably most.
2. Of those 84%, how many were lying? 10%? 20%?, 30%?

I have not seen this type of question asked before and it makes me believe that the pollsters are aware that they are being gamed by a large percentage of the population.

The net effect of this, along with the huge spread in the polling results published, lead me to believe that the actual margin of error is most likely in the 6 to 10% range which renders these polls useless. The current confidence factor of 95% is probably 40 points lower.

I’ll make the call now. McCain/Palin by 4.8%. Just get out there and vote, take your friends with you, your family, everyone who is going to vote McCain/Palin. But get out there and vote. We are going to win and have been on track todo so for quite a while.

And I think that they are oorrect.

[texmexis best]

“And I think that they are correct” should be deleted. It refers to the pollsters view that they are being gamed.

[Gandalf_The_Gray]

3. Of those 16%, how many would never tell the truth to a pollster? Probably most.

And now this to make your head explode: If 16% told pollsters that they would lie to a pollster, were they lying to the pollster when they said it?

Let me know if you figure that one out!

Regards,
GtG

[kesg]

Hey, that guy stole MY thunder! He’s absolutely right, of course. It is as senseless to ignore party affiliation when weighting polls as it is to ignore age, gender, or race. He’s also right about what the early voting numbers say about the existence — or, more precisely, the lack thereof — of any “enthusiasm gap.”