Today I Like Gallup Better Than Zogby

[DemocratSinceBirth]

www.pollingreport.com


Actually, Zogby's numbers are more static while Gallup's numbers are more fluid....


After the debate debacle less people are admitting to be Republicans and since Gallup doesn't "weight" their polls for party affiliation they are showing a much bigger bounce for Kerry....

[Jersey Devil]

1. Gallup does indeed "weight" its polls

In fact that is the controversy surrounding Gallup since it weighs republican votes in higher percentages than have ever been made in previous elections. With the correct proportions of Dems and Republicans Kerry would be far ahead. Zogby weighs them too though I do not know his methodology.

[DemocratSinceBirth]

4. Nope...

Gallup doesn't weight their polls for party affiliation as they believe party affiliation is not a constant like race, religion, education, income, et cetera...


They make enough calls to get 1,000 likely voters and the percentages of Republicans, Democrats, and Independents are what is reflected in their polls....

Zogby weights for part I D . His model assumes the same percentages of Dems, Reps, and Indys will be voting as they have done in the recent past...

That's why the Gallup poll is so volatile...

[Jersey Devil]

6. Thanks for the explanantion, I misunderstood

I thought Gallup was adding arbitrary weighting for party affiliation but if all they are doing is taking the percentages as they come in that would indeed account for it being extremely volatile.

[DemocratSinceBirth]

7. Which In This Case Makes Sense...

Kerry jumped eight points in the Gallup Poll since the debate....

It seems less people are admitting to being Republicans....

[AndyTiedye]

11. People Don't Switch Party That Often!

[On the Road]

12. This is the Legitimate Argument for NOT Weighting

by party affiliation -- because it's fluid. I've heard Zogby say the same thing on CSPAN radio. People often change their party ID (or label themselves independent) based on how they feel at the moment. And because Bush was doing better before the debates, some of that Rep/Dem imbalance may have been real.

At the same time, the Gallup people HAVE to know that a +8% margin in party ID does not bear the slightest relation to the group that will show up to vote on election day.

Maybe there's a better way. If you could get actual party registration, for example, that would eliminate the response variable.

[KharmaTrain]

2. Steady As She Goes...Zogby Is Still The One

John Zogby is still sticking with his models and the data still hasn't factored in several days of spin that keeps making Bunnypants look worse.

I've noticed Zogby moves slower than most polls, and I'm still waiting on the results of the online poll I took over the weekend :grin:

If you want numbers really to watch, look at the newspaper polls in the various swing states...they're a good barometer that Zogby will pick up and Gallup doesn't...and a major reason Zogby is far more reliable.

[mvd]

3. I'm also sticking with supporting Zogby

I think Gallup might just be coming down to earth. The fact that 8% are undecided in the Zogby poll makes me think that if he pushed leaners, Kerry would likely be in the lead.

[KharmaTrain]

9. We'll Need To Do A Poll On The Pollsters...LOL

The "story" of this election will probably be how goofy the polls are. Gallup has to be the most obvious and I think they're trying to save some face here.

One very boring, but excellent thing about the Zobgy poll I take is how he always asks the same major demographic questions...and throws in some off the wall ones now and then to measure the sample from a different variable. Zogby really gets to know who he's polling.

Gallup's been caught with their polls up their asses. As Tweety said the other day "there's no way 11% of a polarized nation can swing from one side of the street to the other overnight". Cite what you want...poor sampling techinques, avoiding un-registered...but now registered voters, cellphones, voice mail, pranksters who give bogus answer (now I'd never think of doing this :evilgrin:)and a overweighing Repugnicans is making them backtrack.

I hope we see a blame game with these bloodsuckers as they try to explain how they were so wrong on Nov. 2.

Cheers!

[Jersey Devil]

5. This is the Zogby Poll I watch - the battleground report to the WSJ

and yes, he does move more slowly and carefully than the others imo. The battleground poll probably will be updated after this past weekend's polling.

http://online.wsj.com/public/resources/documents/info-battleground04.html

[sonicx]

8. there's no way that people's presidential peference is that fluid

The media knows this, but they said nothing since it helped bush. I bet they'll question it now. :)

[Proud2BAmurkin]

10. Gallup had Gore LOSING by 2, down by 13 in mid October

They suck

[DaveinMD]

13. they are saying the same thing

a tie or a three point deficit is statistically insignificant.

[NewYorkerfromMass]

14. Gallup correctly weighted this poll as per 2000 party affil.

according to kos....

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

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

http://www.dailykos.com/story/2004/10/4/174936/830

[Thrill]

15. All these polls are full of crap

I'm now convinced all these polls are full of crap I remember when all the other polls had Bush ahead by 6-11 points. And Pew had it Bush only ahead by two. And now the others have it Kerry ahead or tied, And the Pew Poll has Bush way ahead. Puh Please. Just show up and vote