Gallup Polls: No Longer Credible in the Least

[Beetwasher]

I have come to the conclusion that Gallup can no longer be considered credible in any way. I always suspected that they had a conservative slant, but now I believe it is actually more sinister than that. They are truly part of the propoganda arm of the WH. A couple of recent things have lead me to this conclusion.

The new WP/ABC poll confirms the overall downward trend of Bush's approval and is backed up by several other polls. Gallup's latest poll that has him up significantly, is at the very least anomalous, however I think it's more than that.

Ignoring the suspect methodology of the poll for the moment, if you look at Gallup's last three polls you'll see something very interesting. The poll taken 9/19-9/21 has Bush at 50%. The next one taken around 2 weeks later on 10/6-10/8 has him at 55%. Around 2 weeks time is a normal break between their polls if you look at the history. However, they did something very interesting. They took another poll 2 days later 10/10-10/12, that had him at 56%. Now why did they do that?

The effect that's created is three data points that seem to indicate and UPWARD trend. They are contemptably trying to creat this illusion and at least to me, it's transparent.

http://www.pollingreport.com/BushJob.htm

The second recent thing that convinces me Gallup is merely a tool is their bogus poll in Iraq. They have put their supposed "credibility" behind this farce, but to claim this poll can be in any way credible or representative is nothing but load of tripe.

http://www.newsday.com/news/nationworld/nation/wire/sns-ap-baghdad-poll,0,3149623.story

There is no way in hell they can vouch for the accuracy of this poll. Their methodology must be suspect. How did they choose their sample? I'm assuming they only questioned people who weren't shooting at them. Other than that, I think the fact that a rep from an American Organization like Gallup (who the fuck else would be running a poll in Iraq? I mean come on, like the people didn't know just because they didn't tell them? Yeah right), asking someone questions about how they feel about the occupying power, while that power has soldiers with guns, tanks, planes and bombs everywhere, might make one inclined to say what they thinkg the Gallup rep wants to hear. This is an absurd poll.

This bogus poll in conjunction with their fabrication of an upward trend for Bush Approval, completely destroys any semblance of credibility the once might have had. They are pure, unadulterated propoganda.

Fuck you Gallup. You were once a respectable organization but now you are transparently corrupt.

[diamond14]

1. Bush’s Approval Rating Stablizes (link)

Bush’s Approval Rating Stablizes
Poll Shows 53 Precent Back President After Months of Decline


Heartened by opinion polls indicating President Bush's six-month slide may have ended, the White House has launched steps to reassure supporters before the 2004 campaign becomes fully engaged.

Leaving nothing to chance, Bush-Cheney campaign manager Ken Mehlman summoned a handful of trusted lobbyists, pundits and pollsters to the Arlington headquarters last week for a pep talk and strategy preview. Attendees said Mehlman contended that Bush's standing is reassuring by historical standards and that Bush will enjoy the most elaborate national campaign structure ever built.

Among other troubling signs for Bush: 46 percent said they would reelect him if the election were held today, while 47 percent would vote for the yet-to-chosen Democratic nominee. The percentage that thinks the war in Iraq was worth fighting dropped to 54 percent, from 61 percent a month ago. Majorities said they do not believe the Bush administration has a clear plan for handling the situation in Iraq and said they consider the number of U.S. military casualties in Iraq to be unacceptable.

http://www.washingtonpost.com/wp-dyn/articles/A25667-2003Oct14.html

[Old and In the Way]

2. Much harder to steal elections if the polls aren't cooked.

One needs the other......

[berry]

14. Exactly. So the polls and the "good press" are critical.

This excellent point needs much repeating. We can't ignore it. The propaganda is part of the problem, because even when it's clumsy and transparent it serves their purposes. Aargh.

[Beetwasher]

36. Indeed

If they are in fact planning to steal the election, they will need polls to back them up that shows it's at least close...

[snoochie]

3. I'm with you, Beetwasher.

All part of the plan to continue right on 'manufacturing consent'.

Pay no attention to that man behind the curtain.

[incapsulated]

4. Zogby says the same thing, though

And I trust Zogby.

Bush Numbers Climb; Leads all Democrats; Still Behind on Re-Elect and Against Generic Democrat; Clark and Dean Share Democratic Lead, New Zogby Poll Reveals

http://www.zogby.com/news/ReadNews.dbm?ID=739

[snoochie]

5. I don't trust any of 'em.

Zogby was off by as much as 14 percent in 2002 races.

[Eloriel]

38. Yes, in certain Diebold areas

Like Georgia, where the swing between Perdue and Barnes (Gov.) was 16 percent, and the swing between Chambliss and Cleland (Senator) was 13 percent in pre-election polls versus post-election results. These were not Zogby polls, btw.

There was also Colorado's swing and Minnesota's swing for Senators, big but not quite as big as Georgia.

Eloriel

[Beetwasher]

42. I think Gallup may very well be their "cover"

for rigging the Diebold machines. They'll point to Gallup polls that have the Chimp in the running and use that as evidence to support fishy election returns...

[jeter]

8. Dude, that Zogby poll is almost a month old

I'm interested in what their new polls say, not what they said one month ago.

[incapsulated]

10. I'm not a dude...

And even though it's from 3 weeks ago, it showed the upward trend.

[oasis]

15. huh?

:crazy:

[incapsulated]

17. what?

I'm nuts because Zogby showed that his rating was on the upturn 3 weeks ago? Which makes a current poll putting it even higher now understandable?

:eyes:

[Beetwasher]

26. No, it does NOT show an upward trend

It shows one data point that ticked up. That's NOT a trend. The trend is decidedly down for every poll except Gallup.

[Beetwasher]

25. That Zogby poll is still out of whack w/ the Gallup poll

Zogby's one of the better pollsters IMO. But that one poll doth not a climb make. The overall trend is down, including for Zogby. One data point does not make a trend.

[diplomats]

43. But that was three weeks ago

and 50 percent actually puts Zogby in line with other polls.

[WillyBrandt]

6. No: We must never look away from numbers

Sorry, but this isn't a systematic analysis. Sometimes the numbers go our way; sometimes they don't. They aren't polling DUers: they are polling the general population, much of which (alas) still does not think know that Bush is a lying bum. This is Gallup, not Fox News--they may not be perfect, but they aren't part of the BFEE.

We will win by facing and changing facts, not by retreating into comfortable unreality. The numbers show we have a way to go, but also that huge potential is there for a Democratic victory if we're smart about it.

Let's work.

[w4rma]

13. I totally agree (n/t)

[Beetwasher]

27. Huh?

Never said it was a systematic analysis, however there is certainly evidence that something's fishy w/ Gallup and I stand by that. I don't know about your background, but I have a bit of experience in social epidemiology and I am somewhat familiar with similar polling methodology...

[KoKo]

7. Yes, your assessment is interesting. Gallup Highest. Even Imus (pig)

said this a.m. that the Iraqi Poll was "bogus." The reader (whoever they guy who does news on that show) read the Gallup Poll and said: "and the pollster said Iraqi's were following the Gallup people around 'begging' to be counted in the final tally."

If even "Whoreman" Imus can't accept that the Gallup Iraqi Poll is bogus....given that this is Bush's "Explain the War Week" then why should we believe that Chimp Popularity Plus Iraqi's" giving us honey and flowe petals isn't completely a a Chimp Pump?

I'm not going to get depressed over that Repuke Poll....(well, I was sorta) But NO! Not going to let it get me down!

[Mariner]

9. Too much whining

Some of you people remind me of Freepers. You don't like a poll so you say it's biased, without presenting much in the way of evidence other than that you don't like the results.
Excuse me for being blunt, but we on the left need to be smart and not act like a bunch of whiny Rush Limbaugh dittowads every time we see a poll or a news story that doesn't validate our view of the world.

[oasis]

16. "without presenting much in the way of evidence"??Check original post

BTW, some folks distrust these polls because they show a sudden, unexplained support for the Chimp. That's not "freeper" whining, that's being suspicious of corporate run, corporate influenced polls.

[Cheswick2.0]

20. we on the left?

I love it when people are so obvious

[corarose]

34. Thank You!

:kick:

[ConsAreLiars]

11. Gallup rep on msnbc

spun the Iraq "poll" violently. The question about US occupation forces staying more or less than 2 months was described as indicating that "most" wanted US troops to stay for an "indefinite" period. Methodology issues aside, this alone was clear evidence of political whoring by a gallup operative.

[TruthIsAll]

12. The peripatetic CNN/Gallup poll was a farce in 2000..

Recall that the numbers gyrated madly week after week. They must have picked good days for Bush to poll. Every week he jumped 10%, then Gore came back.

CNN Schneider and Woodruff made sure to snuff any positives for Gore..

Trust Zogby and Harris.

[DeepModem Mom]

18. Wonder if AEI Schneider has input into Gallup process --

or does he just interpret and broadcast the results on CNN?

It's an outrage that CNN's "senior political analyst" is an American Enterprise Institute fellow, and, I assume on AEI payroll, as well as CNN's. A "trusted news source," as CNN advertises itself, should, at the least, make viewers aware of Bill Schneider's agenda. Of, course, MSNBC's pollster Luntz is much, much worse, and MSNBC hides his political activities and agenda.

[Beetwasher]

35. Yes, it's unfortunate, but CNN's agenda is becoming more

clear w/ each passing day. It's sad really...

[Polemonium]

19. Most polling is suspect

Polling is "soft" science. Results are almost impossible to verify. The only polls that have any credibility are ones that have been duplicated many times using a variety of methodologies. Lone polls fail to achieve a random sample, fail to achieve adequate sample sizes, and frequently ask questions that fail to capture true opinions. i.e. Do you think Gov. Davis has done a good job - well for most of us the answer is more complex than yes or no, and the way the question is phased can alter our response. When lone polls fail at these things the validity of statistical assumptions go out the window. When 30 independent polls state that Shrub is popular again, I'll believe it. Until then just ignore it.

[Beetwasher]

28. Correct. That's why the trends are important over a range of

different polls. And that's why something's fishy about Gallup. When one pollster bucks the trend they need to be scrutinized more closely. Gallup consistently polls Bush's approval significantly higher than other pollsters and is now bucking the downward trend. Something's either very wrong w/ their methodology or with all the other pollsters methodology. Occams razor suggests it's Gallup that's fucked, not everyone else.

[Lexingtonian]

21. I wouldn't go that far

We do have a person here who used to work for Gallup and left, and s/he says they have (1) new management that doesn't care about quality anymore, and (2) a demographic model that skews conservative. S/he repeated herself a couple of times and probably got bored of telling the same story every time some bozo who hasn't recieved the memo complains about Gallup.

There is an analysis up at Pollkatz, which puts Gallup at 4th of 16 polling outfits in relative Rightward skew.

My own observation is that Gallup used to deviate from Zogby's numbers by 5% conservative lean two years ago. Today the conservative skew is higher, usually around 7%-8%. It seems to skew higher with time (or maybe money?). They also seem to blip about in their MOE of 3% alot when Zogby can usually get numbers within 1% surprisingly often. I used to joke that Gallup has a little town hidden in the hills of western Pennsylvania from which they get all their numbers, a town in which the inhabitants never get farther than 10 miles from their homes ever.

On the one hand, their numbers can- with this correction- be read to say that W's approval numbers have stalled or even continued to fall in the 49%-50% range. Or, if other polls showing some increase can be believed, risen about 2%-3%, which can be explained by the amount of bad news emanating out of Washington and Iraq slowing alot and a lot of people trying to maintain optimism throwing one last investment of emotion into the thing.

[Beetwasher]

30. Bozo?

Was that meant for me? Please clarify.

Your account of what that poster said confirms my suspicions that Gallup is corrupt.

You also don't address the Iraq poll. A reputable organization would never conduct such a poll with any notion of confidence in it's outcome. They're polling in a warzone! Good grief, how do you even begin to calculate a MOE in that situation? You can't. There's too many variables to take into account. It's a sick joke and blatant propoganda and says much about the organization that's behind it.

[LunaC]

22. I wondered about Gallup

and the people behind the curtain who were running the organization but I came up with zip, zero, zilch. Knowing who the sponsors are and their affiliations helps determine the intended slant but there isn't ANY info on them (that I could find) that would offer any clues. That, in and of itself, makes me say "Hmmmmmmm....."

[Snellius]

23. Josh Marshall must have read your post, Beetwasher

A bounce back in the polls?

That might be a bit of an overstatement. Here are a number of recently-released presidential approval polls (with the most recent listed first) and how far the president moved up or down from the last time that news outlet did a poll. ABC/WaPo: 53%, down 1; CNN/USA Today: 56%, up 1; Newsweek: 51%, down 1; Ipsos-Reid/Cook Political Report: 51%, down 4.

One extra bit of info, CNN/USA Today Gallup did two polls in rapid succession: one at the beginning of last week and one over the weekend. If we go back one more poll, to the one they did September 19th through 21st, that one had Bush at 50%. So if you bend the measure a bit in the president’s favor, you get one poll with a six point bump.
...
More at: http://talkingpointsmemo.com/

[Beetwasher]

33. Well, he certainly hits the same points!

Hi Josh! :hi:

Love your stuff!

[Q]

24. Sometimes I wonder...

...if the Bushies have a Nixonian Slush Fund where certain people are paid to make them look good at just the right time?

- Don't think it's possible? There are trillions of dollars that remain 'lost' in the system. Perhaps they don't really 'lose' these trillions.

[Beetwasher]

31. Yes, I've often thought the same thing

If the missing trillions aren't in the Cayman's and also being funneled into the stockmarket ;-)

[Starpass]

29. Let me share something that came out of mouth of CNN anchor

Do you remember a few weeks ago there was a Gallup Poll done in Baghadad regarding our presence there, etc?? The Gallup guy 'over there' was broadcasting from abroad on the poll and was on CNN explaining their "finds". Remember that the Gallup Poll is usually listed as CNN/Gallup Poll. Anyway, the anchor that morning said how this Iraq poll was done strictly by Gallup without "CNN help" and then inadvertently spewed to the Gallup guy on screen, "and you know how much WE help you with the polling you do here at home". Now, this remark has to have something to do with more than money. They would be glad to also back that Iraq poll so that they could broadcast it first, etc. What the hell does CNN mean???----I think if you can get to the root of that, you probably will have an answer of why they seem to be the "cheerleading" poll!!! Any ideas????

[Beetwasher]

32. Not sure what to make of that, but you're right

Gallup is partnered w/ CNN. I suspect that Gallup's funding to be in Iraq though isn't coming from CNN. I suspect it's part of a gov't contract and that's why it's not a CNN/Gallup poll...It would be interesting to follow that money.

[arwalden]

37. I Feel Like Roger Rabbit... You've Just Tapped Out "shave-and-a-hair-cut".

And I'm COMPELLED to respond with TWO-BIIIIITS!

What exactly ARE my "two-bits" you ask?

Okay. Here goes: "ALL POLLS ARE BULLSHIT". Every time I see a thread or a single post about any poll---no matter if the results are in our favor or against us---I feel compelled to remind everyone that polls are crap.

You've heard my reasons before. So I won't bore you with them here. (I'll repeat them again soon enough.)

-- Allen

[Beetwasher]

39. LOL!

Well, I don't necessarily agree. Polls are a bellweather and an indication of trends. They can be used correctly or abused like anything. There's no such thing as a totally accurate poll. However, it seems to me that Gallup is almost totally corrupt. Others are IMO at least somewhat concerned with conducting sound polling. Gallup OTOH seems concerned with pushing agenda and tailoring their polls to fit that pre-ordained agenda.

[Nederland]

40. The Iraq Poll

I actually think the Bagdad poll is perfectly plausible. It makes sense that a large number of Iraqis there would want the Americans to stay because everybody knows that if the Americans were to leave there would be a civil war in Iraq with much more violence than we are seeing today. The people there are just voicing a concern for their own safety. Sure, they would love to see the Americans gone and a stable, independant Iraqi government in their place, but sensible people know that that is not an option right now.

[Beetwasher]

41. I'm sure there are Iraqi's that hold that opinion

However, any polling organization with any sense of credibility could never vouch for the accuracy of that poll, nor should they. Gallup does. They put their name to it, they assign a MOE and maintain that it's credible. I'm sorry, but that's bullshit. The conditions under which it was conducted (in a war zone) leave too many variables to accurately conduct a poll that has any significance whatsoever. The results of this poll can in no way be considered a real reflection of the attitudes of the people in Iraq. It's just not possible under the circumstances and conditions in the country. As just one example of the problems they face in constructing this poll, they claim the sample was random. Well, how did they choose the sample? It couldn't really be random could it? They went to people's houses and conducted face to face interviews. What if the persons house was blown up and no longer there? What if a person picked for an interview was among the uncounted dead or too sick to be interviewed? Did they interview people in Tikrit? See what I'm getting at? People they supposedly interviewed at random were already biased to some degree in that they probably weren't a part of the resistance and shooting at Americans, they were still alive and had a place of residence etc.

It's nonsensical to even try to conduct a poll under such conditions. But to do it and maintain that it's valid speaks volumes about the responsible organization.