Gallup has a 16 point party ID swing in 7 days

[○∙◄☻¥tπ[╪AV┼cVê└]

It was Republicans +12, now it's Democrats +4
Gallup=junk

Likely Voter Samples
Poll of September 13-15
Reflected Bush Winning by 55%-42%

GOP: 40%
Dem: 33%
Ind: 28%

Poll of September 24-26
Reflected Bush Winning by 52%-44%

GOP: 43%
Dem: 31%
Ind: 25%

Poll of October 1-3
Reflected Dead Heat 49%-49%
772 Likely Voters

GOP: 35%
Dem: 39%
Ind: 26%

[The Vorlon]

It was Republicans +12, now it's Democrats +4
Gallup=junk

Likely Voter Samples
Poll of September 13-15
Reflected Bush Winning by 55%-42%

GOP: 40%
Dem: 33%
Ind: 28%

Poll of September 24-26
Reflected Bush Winning by 52%-44%

GOP: 43%
Dem: 31%
Ind: 25%

Poll of October 1-3
Reflected Dead Heat 49%-49%
772 Likely Voters

GOP: 35%
Dem: 39%
Ind: 26%

Please provide a source for the Gallup party IDs. I do not believe they are correct actually.

EDIT - Ok - these are at the LV level... re-calculating

I was working at the RV level... ;(

[AuH2O]

Me neither.

But regardless, a true random sample is theoretically the best method. The only problem is sample size, which because of economic reasons is not as big as it ideally would be.

[Fmr. Gov. NickG]

The huge party ID swings present in Gallup's polling this year go well beyond the expected swings due to sample size....into the range of 4-5 standard deviations on more than one occasion.  There is something seriously wrong with their methodology.

[○∙◄☻¥tπ[╪AV┼cVê└]

It was Republicans +12, now it's Democrats +4
Gallup=junk

Likely Voter Samples
Poll of September 13-15
Reflected Bush Winning by 55%-42%

GOP: 40%
Dem: 33%
Ind: 28%

Poll of September 24-26
Reflected Bush Winning by 52%-44%

GOP: 43%
Dem: 31%
Ind: 25%

Poll of October 1-3
Reflected Dead Heat 49%-49%
772 Likely Voters

GOP: 35%
Dem: 39%
Ind: 26%

Please provide a source for the Gallup party IDs. I do not believe they are correct actually.

Here's the older ones.

http://www.usatoday.com/news/politicselections/nation/president/2004-09-29-gop-poll.htm

[freedomburns]

That Gallup Poll that was fed to CNN and USA Today last month, as well as the rest of the media was seriously flawed in that the likely voter sample they used assumed a 12% advantage for the GOP (43%-31%) that doesn't exist now and has never existed in the last three elections.

The Democratic advantage from the 2000 exit polls was 39% to 35% over the GOP. Yet this flawed poll showed a narrowing Bush lead from their similarly flawed poll of two weeks ago. So if a poll with an insupportable GOP bias of 12% in its likely voter sample, shows an 8% Bush lead amongst likely voters when a poll they used two weeks ago with a 7% GOP bias showed a 13% Bush lead with likely voters, then how can anyone not conclude that Kerry is doing much better than Gallup would believe?

Gallup smokes crack with Rush Limbaugh (because he knows where to get the good drugs).

[A18]

Gallup is supposedly supposed to show trends, which means Gallup doesn't claim that's what the real swing was. Just for their polling purposes.

Yes, there was a 16 point swing going by Gallup's super-accurate methodology.

[mddem2004]

Ah......its what I've said for a few weeks now......You get the party ID close to what the 2000 and 1996 levels were
(ie 39% Dem 35% Rep 26% Ind) and you will see the race is and ALWAYS was much closer than Gallup said it was...

And frankly, with the reports we are now seeing of RECORD BREAKING LEVELS of new registrations, mostly in urban areas, the LV model may not be quite as accurate as many have come to expect.

[Shira]

It was Republicans +12, now it's Democrats +4
Gallup=junk

Likely Voter Samples
Poll of September 13-15
Reflected Bush Winning by 55%-42%

GOP: 40%
Dem: 33%
Ind: 28%

Poll of September 24-26
Reflected Bush Winning by 52%-44%

GOP: 43%
Dem: 31%
Ind: 25%

Poll of October 1-3
Reflected Dead Heat 49%-49%
772 Likely Voters

GOP: 35%
Dem: 39%
Ind: 26%

Please provide a source for the Gallup party IDs. I do not believe they are correct actually.

EDIT - Ok - these are at the LV level... re-calculating

I was working at the RV level... ;(

I don't understand the problem you see with Gallup.
The latest before the debate was Kerry = -8% and now it is tied. This means that Kerry went up and Bush went down by approximately 4%. That's exactly what Newsweek's and CBS's polls are showing.

[A18]

Then I guess Bush will do even better, seeing as he's up more with RV.

Newsweek is a trash poll. CBS, okay, but look at ABC.

[A18]

According to Gallup there was an 8% shift. And Gallup uses methodology to detect energy among Reps/Dems.

ABC has Bush up 5, as does Pew.

[Shira]

"Gallup has a 16 point party ID swing in 7 days"
According to Gallup there was a 4% and not 16% shift

[A18]

No, it was an 8%.

He was talking about party ID, not the horse race numbers.

[J. J.]

No, it was an 8%.

He was talking about party ID, not the horse race numbers.

How did they calculate the Party ID numbers?  Were they self identified?

[ATFFL]

No, it was an 8%.

He was talking about party ID, not the horse race numbers.

How did they calculate the Party ID numbers?  Were they self identified?

Gallup asks for your party ID.  If you say independent or none they ask which party you favor.  If you say independent or none they ask for a preference.  Then they ask which way you lean.  Then they might ask one more, but I think that is it.

In otherwords they heavily push for a party ID.

[Friar]

I have no problem with polls showing party ID close to the turnout from the last presidential election.

So, I don't see why republicans complain from this poll? They didn't complain when it was showing Bush up 12 with oversampling republican voters...

Right now the race is anywhere from Bush +2 to Kerry +2.

I didn't expect such a big shift just from 1 debate. What I expected is a small shift in the momentum and by election day Kerry leading by 1.

Make no mistake about it. The country is polarized.

The party that turns out its voters on Nov2 wins.

[Vasall des Midas]

No, it was an 8%.

He was talking about party ID, not the horse race numbers.

How did they calculate the Party ID numbers?  Were they self identified?

Gallup asks for your party ID.  If you say independent or none they ask which party you favor.  If you say independent or none they ask for a preference.  Then they ask which way you lean.  Then they might ask one more, but I think that is it.

In otherwords they heavily push for a party ID.

Which would explain why their party ID figures can swing around a lot - undecided independents who currently slightly favor candidate x getting identified with his party.

[Democratic Hawk]

As far as party ID goes, I recall a survey of 10,000 voters (which I read about on CNN)  in which respondents were 43.5% Republican and 43.3% Democrat.

As to whether this is a true reflection of party ID, I don't know.

Dave