Blitzer/CNN defend Gallup polling today

[SoCalDemocrat]

Blitzer did a segment this morning on Gallup, interviewing a Gallup organization member. Of course no one with an opposing viewpoint was interviewed.

Wolf raised the issue that Gallup has been using 40% republicans in his recent surveys which some accuse is biasing the results (no way!).

Gallup response was that Annenberg did a study in 2000 that shows peoples views of their partisanship shift around over time. They claim that those surveyed don't know whether they are Democrat, Republican, or Independent from one week to the next. They didn't further try to substantiate this claim. Can anyone locate this study and verify if it's valid? I strongly contest this claim and polling basis. I believe the vast majority of Americans are decided on the issues, and know where they stand come Nov 2nd. Undecideds have been pegged at somewhere between 3% to 4% by most sources. How then can Gallup justify a 40% Republican sampling, when we know that only 34% of voters were Republican in the past two elections (2002 and 2000)? Even if every single undecided in the U.S. did decide they were Republican when Gallup called them up last week, that would only account for a 38% Republican showing.

Blitzer ended by giving his complete endorsement, saying that the CNN/USA Today/Gallup poll is the best poll in the industry.

Wow! Could they of provided a more biased, one sided coverage of a topic?

Wolf also failed to point out that in 2000, just 5 days before the NOvemeber elections, Gallup came out with a poll result showing Bush ahead of Gore by 13 points. 13 POINTS! We all know that was horribly skewed and innacurate now, as Gore went on to win the popular vote, and by some standards should of won the electoral college as well.

I think CNN could use another dosing from DU members through their phones and email.

[Search Party]

1. what a load

"Gallup response was that Annenberg did a study in 2000 that shows peoples views of their partisanship shift around over time."

dumbest. statement. ever.

[cthrumatrix]

25. this "fraudulent arrogance" is going to bite them -- Gallup is flawed

[markus]

2. This is much worse that CBS "Memogate"

It is a deliberate manipulation of national elections in the interest of corporate profit.

[rumguy]

9. good point

this is a serious issue....

it's a fucking travesty...

I am pissed.

[KittyWampus]

38. Democrats In Congress Should Call For An Investigation

[blue neen]

3. The very fact that Blitzer did a segment defending Gallup

means that they KNOW Gallup IS biased!!! If they were not feeling the heat, they would not have tried to put out the fire.

[BeyondGeography]

4. Wolfie had the kneepads on for that one

I guess we'll have to see if the poll that delivers the most extreme result turns out to be the "best."

In the meantime, good to see they're feeling the heat. GallupGate, anyone?

[snoogins]

5. nazi media whores

i expect little else from their filthy holes.

[DemFromMem]

6. Well done Dems!

The fact that CNN felt the need to do a story at all means they're on the defensive.

We're all worried that biased polls will become self-fulfilling prophesies because people won't bother voting if they don't think it will make a difference. And the polls affect the perception of the candidates - people like winners.

If we can get the story out in enough places that the polls are biased to the right it's going to lead to a general mistrust. The right wing has done this number to us over and over again this cycle on the swift boats and the flip flops. Just repeat the same accusation enough and you'll create doubt. In this case, we've got the truth on our side so no need to have a guilty conscience.

Now we just need to get the media to start attacking each other like they did on the CBS story. My poll is fair! No your poll is bunk! Blah, blah, blah.

[gratuitous]

7. Didn't Judy run this segment yesterday?

Several minutes of unrefuted testimony from Gallup about how honest they are. This after Judy got waxed by some Democratic operative knowledgeable about Gallup's fraudulent methods.

I wonder why CNN is so heavily "invested" in defending Gallup?

[dennis4868]

8. Blitzer defending Gallup is like....

Bush talking how great things are in Iraq right now....hahahaha.....they are 2 of a kind!

[oldsneakers]

10. CYA

sounds like they got some heat over these bogus polls-I just emailed CNN and nicely requested that they please include a Dem pollster on to give their take on Gallup-and not that bitch Cadell.

[virgdem]

11. There he goes again...

trying to defend the indefensible. I don't even bother to watch Conservative News Network anymore. My blood pressure is high enough, I don't need to elevate it any more than it is already.

[joanski01]

12. I saw that, too.

Wolfie is a piece of work. Than on top of that, he helped that Republican woman from the American Enterprise Institute argue with the Democratic lady who used to work for Clinton. Two against one!

[SoCalDemocrat]

13. MysteryPollster is the material they are using against us

Their rhetoric, almost word for word, appears to have been lifted off of this site:

http://mysterypollster.typepad.com/main/2004/09/weighting_by_pa.html

The LA Times spin from Blitzer's piece, for example, sounded word for word as reported on the above site.

Interestingly

"There is another good reason to be weary of automatically weighting survey results by Party ID. Most of the national polls ask Party ID toward the end of the survey. Campaign pollsters can tell you that Party ID can vary with the content of the questions that precede it. Include a long battery of items on the health care or the environment and you are likely to get more Democratic identifiers by survey end than if you include a long battery on fighting terrorism. "

[HFishbine]

14. That's so lame (re: party ID)

Why not just ask "With which party are you registered?"

[tritsofme]

17. Some states don't have official party registration

Like Illinois for example.

[HFishbine]

18. I forgot about that. Thanks. (eom)

[SoCalDemocrat]

20. Why not ask at the beginning of the interview vs. the end

Rather than biasing the results and asking them at the end how they feel about their party id, ask at the outset?

Are you a Republican, Democrat, or Independent?

For states that don't have party registration, they can answer that question with the party the identify with. For others, they can respond with their party.

Then give them the rest of the poll.

[tritsofme]

22. Good question

I guess it depends on the outfit.

[SoCalDemocrat]

24. bias polling and presentation of results

IF they asked at the beginning, and tied it more closely to Party ID, then they could make qualified statements about the results such as:

If we had a turnout today of 32% Republican, 35% Democrat, then we can expect Kerry to get X and Bush to get Y of the vote.
However, if the turnout was <blah> then the result would be <blah>.

Gallup could certainly do that with its polling data if it was collected more fairly and accurately, and if it was presented without bias.

[tritsofme]

27. I can send you a complete sample of a recent Gallup

poll if you wish, and you can go fishing for bias in the questions.

And BTW, they do not weight for party ID.

[SoCalDemocrat]

34. Please post it with the results

What I'd like to see is the complete poll from one of their slanted results. One component missing of couse are the pollsters. We don't know how they actually asked the questions over the phone. You can add alot of bias in the emphasis placed on words, the tone of the pollster, etc.

[tritsofme]

35. Its 91 questions,

And I can only go back as far as the 9/3-9/5 poll.

and I'd probably being violating copyright posting it here, so I'll send it in a private message to you.

[SoCalDemocrat]

41. The racial composition of this poll looks skewed

I picked just one problem with the poll, the racial profile. 25% of Americans are Black or Hispanic, but Gallop just interviewed 13% Blacks and Hispanics in that poll.

[tritsofme]

42. They do weight for factors they can check against census figures,

such as race presumably.

What I gave you was the raw data.

[SoCalDemocrat]

44. They don't have sufficient data

They only had 2% Hispanics polled. Hispanics comprise 13% of the population. They would have to poll a whole lot more, or throw away almost all the other data. I don't see how that poll can be considered demographically balanced no matter what type of editing of the raw data they perform.

Also I'm not sure how they would select who to drop from their raw data. That would also tend to bias the results. How can the drop raw data (people who were polled) impartially?

I would expect a pollster to collect demographics such as race and party ID up front. Then they would know whether to continue polling that person, and could accumulate the necessary numbers to have a representative cross section of the voting populace.

[SoCalDemocrat]

15. looking at Oct 27, 2000 polls vindicates Zogby, damns Gallup

We know now that Gore lead Bush in the popular vote, and nearly won the electoral college.

On Oct 27th, Zogby had Gore ahead by 2 points in the polls.

CNN/USA Today/Gallup gave Bush a 52 percent to 39 percent edge over Gore.

Which polling technique was accurate? The answer without question is Zogby.

-------------------------

October 27, 2000
Web posted at: 8:36 p.m. EDT (0036 GMT)
http://www.cnn.com/2000/ALLPOLITICS/stories/10/27/cnntime.poll/index.html

WASHINGTON (CNN) -- Republican presidential nominee George W. Bush holds a 49-to-43 percent edge over Democratic rival Al Gore in the latest CNN/Time poll, conducted Wednesday and Thursday.

The poll of 2,060 adult Americans, including 1,076 likely voters, has a margin of error of plus or minus three percentage points and is thus in essential agreement with a CNN/USA Today/Gallup tracking poll also released Friday. That poll gives Bush a 52 percent-39 percent edge over Gore. More important, both polls show the same snapshot of the current state of the presidential campaign: a solid advantage for Bush.


ABC News and The Washington Post both have daily tracking polls today putting the race at 48 percent for Bush and 45 percent for Gore. The latest Reuters/MSNBC/Zogby tracking poll has the contest at 45 percent for Gore and 43 percent for Bush.

[Gyre]

16. Good point.

Be nice if somebody would throw this in Gallups "face" in mainstream media.

Gyre

[tritsofme]

19. And they showed the race closing nearly every day until election day

And that showed the momentum toward Gore that was appearent in the final weeks of the campaign.

In their last pre-election polls, both Zogby and Gallup were off 2%, Gallup +2 in favor Bush +1 in favor of Nader, Zogby had +2 in favor of Gore, and +2 in favor of Nader.

[SoCalDemocrat]

23. Why were they off SO FAR leading up to the election?

By Nov 1st, everyone had made up their minds.

Why was Gallup off by the distance of a solar system in their earlier polls? There are two possible answers?

A) Their polling methods are faulty or innacurate
B) Their organization is biased

Clearly Gallup is generating biased polling numbers leading into the election. Those polls are cited broadly across media sources, and they have an impact on undecided voters. Most voters want to pick a winner. Who wants to go with a loser? If Gallup shows Kerry behind by double digits a month before the election, there is no question it's going to hurt Kerry.

These organizations should not be allowed to interfere with the election process by pandering their partisan results and representing them as accurate, unbiased, polls.

I would like to see anyone provide a credible defense of Gallup and the extremely innacurate and slanted results they have posted in the past two Presidential elections.

[tritsofme]

26. I think Bush did have a lead going into the final stretch of 2000

not as as big as 52-39, but I think he had a lead.

But the final 2 weeks or so showed a huge momentum shift toward Gore, he had a lot of things go right for him, from DUI, to really hitting his message on the campaign trail.

It doesn't do them any good in terms of their credibility to put out BS polls.

[sonicx]

30. right now, do you think bush leads as much as gallup says?

I think it's about 4 points.

[tritsofme]

31. Right now I personally think bush leads 4-6 points.

Kerry has been at his best the past week or so.

[sonicx]

28. if a poll is only going to be accurate a couple days before the election..

why should we pay attention to it?

[NightOwwl]

21. Wolf is a magnificent icon of journalistic integrity.

If he gives it his complete endorsement.

That's good enough for me. Not!

:puke: on Leslie

[ItsMyParty]

29. Didn't MoveOn Just do a huge ad in the NY Times attacking the

polls like Gallup and their methods???---I'll bet that is why the little terds are trying to keep all this shit floating. It was to be their means of manipulating the voters and it got exposed. So now they are trying to tell the public "trust us"---yep, as far as I can flush you.

[SoCalDemocrat]

33. Wolf cited the MoveOn attack vs. Gallup

Yes it was cited as a reason for the CNN piece.

We need to continue the attack. They have not adequately defending their polling practices. I still firmly believe Gallup is presenting biased information and using innacurate polling techniques.

[Crunchy Frog]

32. I think it's a good sign

that the media is starting to be put on the defensive. We have to keep pushing at them, and not let up.

[SoCalDemocrat]

36. Racial statistics also slant these surveys

12.3% of the population in 2000 was Black and 13% Hispanic according to the U.S. Census. Both populations vote predominantly Democratic, with Blacks voting 90% or more as Democrats.

This comes from a recent Gallup survey where about 10% of those surveyed were Black, but just 2% were Hispanic. So out of a 25% Democrat demographic, Gallup polled just 12%.

------------------------------------

Field Date: 09/03/2004-09/05/2004
Sponsor: CNN/USA Today/Gallup
Methodology: Outbound Telephone
Population: National Adult


D4A What is your race? Are you white, African-American, Asian, or some other race? Mean: N/A Total N: 1018


% N

1 Other (list) 3.56 36

2 (DK) 0.05 1

3 (Refused) 1.20 12

4 HOLD 0.00 0

5 HOLD 0.00 0

6 White 81.50 829

7 African-American/Black 9.75 99

8 (Hispanic) 2.50 25

9 Asian 1.45 15

[dave502d]

37. Has anyone heard of the new defend the polls this much before?

They want this cheat to look real.

[devinsgram]

39. Can't wait. Is egg hard to get off your face?

Then again Leslie might look better with egg on him.

[Bullshot]

40. Wolf Blitzer has a hard enough time to defend his name!

Where did that douchebag ever come up with that name?

Maybe he should change his name to douchebag. It sounds more dignified than the name he's using. It sounds like some type of bar drink.

[NightOwwl]

45. He thinks it's sexy.

He hears women's voices in his head: Oh Wolfie, you are a stud. You are such a big, bad Wolfie.

Ohmygod...I need to go wash my mouth out with soap now!

[spanone]

43. Of Course they Did, they pay a Fortune for it.

[SoCalDemocrat]

46. Do you have info

What is CNN's financial link to Gallup? Does it make money for them when they run a CNN/X/Gallup poll? Do they sell the data?

[tritsofme]

47. CNN and USA Today foot the bill

for Gallup to conduct the poll.