Free Republic
Browse · Search
News/Activism
Topics · Post Article

To: tatown

“We have never weighted polls by party affiliation” - Frank Newport, Editor in Chief of Gallup.

Party affiliation follows the way a person is currently inclined to vote and party affiliation doesn’t drive how someone votes.

“We only ask the person after we have polled them who they currently self-identify with at the end of the poll for informational purposes.”

See here:

http://www.freerepublic.com/focus/f-news/2948176/posts?page=1

Somebody explain to me how such a methodology is better and more accurate than the others.


26 posted on 10/22/2012 10:28:46 AM PDT by SeekAndFind
[ Post Reply | Private Reply | To 1 | View Replies ]


To: SeekAndFind
Somebody explain to me how such a methodology is better and more accurate than the others.

Well let the past results speak for themselves. Gallup has been in the presidential polling business since 1936 and the only time they screwed up was the Dewey/Truman race in 1948. I think that's a pretty good track record.

29 posted on 10/22/2012 10:32:54 AM PDT by SamAdams76
[ Post Reply | Private Reply | To 26 | View Replies ]

To: SeekAndFind

31 posted on 10/22/2012 10:34:07 AM PDT by TornadoAlley3 (Obama is everything Oklahoma is not.)
[ Post Reply | Private Reply | To 26 | View Replies ]

To: SeekAndFind

I listened to a Gallup talking head on some Fox show yesterday, and he said they don’t use party ID weighting at all at Gallup. He said they considered it to be an attitude (hope I’m using the right word), and that it fluctuates with the candidate the person is supporting.

As I understand the logic from those who use party ID weighting, they believe that party affiliation is a long term position held by a person that does not fluctuate in terms of their answering which party they ID themselves with. Those that use party id, like Rasmussen, then adjust it based on other elements of a secret sauce formula that they apply.

The guy from Gallup made a compelling case. I think Rasmussen’s numbers have been closest for a number of election cycles now, so I wouldn’t discount his secret sauce just yet.

They both have Romney up, but Gallup by about 6 and Ras by about 2.


32 posted on 10/22/2012 10:35:06 AM PDT by xzins (Retired Army Chaplain and Proud of It! True supporters of our troops pray for their victory!)
[ Post Reply | Private Reply | To 26 | View Replies ]

To: SeekAndFind

Gallup hx.
http://www.presidency.ucsb.edu/data/preferences.php

Rasmussen hx. 08.

http://www.rasmussenreports.com/public_content/politics/elections/election_2008/2008_presidential_election


34 posted on 10/22/2012 10:37:01 AM PDT by TornadoAlley3 (Obama is everything Oklahoma is not.)
[ Post Reply | Private Reply | To 26 | View Replies ]

To: SeekAndFind
"Somebody explain to me how such a methodology is better and more accurate than the others."

Check their record for the last 60 years or so. Their final polls have been pretty close in nearly every election. They had 49-48 for Ford in 1976 which went 50.1-48.1 for Carter. That's the worst they've been. They've rarely overestimated the GOP candidate.

35 posted on 10/22/2012 10:37:27 AM PDT by CatOwner
[ Post Reply | Private Reply | To 26 | View Replies ]

To: SeekAndFind

While I disagree with that to some degree....would rather weight.....it has some logic to it that affiliation is a changing variable and not necessarily something you can rely on.

There may be a lot of Dems voting Obama or something, who knows.


41 posted on 10/22/2012 10:45:07 AM PDT by rwfromkansas ("Carve your name on hearts, not marble." - C.H. Spurgeon)
[ Post Reply | Private Reply | To 26 | View Replies ]

Free Republic
Browse · Search
News/Activism
Topics · Post Article


FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson