Should We Believe the Polls?

[CrisisPapers]

| Ernest Partridge |

If we are to believe the most recent public opinion polls, this has been a very bad week for the Obama/Biden ticket.

According to Gallup, the Democrats' consistent eleven to fifteen point advantage since January dropped to three points this week. Newsweek, CNN, NBC/WSJ, and CBS all report a tie.

But should we believe the most recent public opinion polls? Today's "dead heat" seems inconsistent with other statistics. Among them:

• New registrations are overwhelmingly Democratic: The AP reported, just last week (September 7) that during the primary season, "more than two million Democrats (were added) to voter rolls in the 28 states that register voters according to party affiliation. The Republicans have lost nearly 344 thousand voters in the same states."


• The same AP article reported that nationwide, registered Democrats outnumber Republicans, 42 million to 31 million.


• As recently as September, Gallup reported that the Democrats had a ten percent lead in party affiliation among voters: 47% to 37%.


• And 80% of the American public is "dissatisfied with the way things are going in the United States." (Gallup, August 23, 2008).

And yet Gallup chooses to survey an even number of Democrats and Republicans. Why? In addition, the pollsters contact users of land-line phones and exclude cell phone users. Presumably, younger and more liberal voters are more inclined to use cell phones. Both factors would surely inflate the GOP numbers.

Moreover, some of the recent alleged shifts in public opinion strain credulity. For example, Jonathan Freedland in The Guardian notes that last week, "the ABC News-Washington Post survey ... found McCain ahead among white women by 53% to 41%. Two weeks ago (before the Democratic convention!), Obama had a 15% lead among women."

That's a shift of 27%. And what could account for it? We can only assume that three days of GOP bombast from Minneapolis and the introduction of a new, pretty, face, convinced a quarter of those white women voters to change their minds.

Sorry, but that's more than I can swallow. Somehow it just doesn't add up.

So, should we believe the polls?

Frankly, I can't offer a simple answer. But I most assuredly have a few nagging questions.

First of all, why wouldn't the polling organizations publish results that are as accurate as reasonably possible? After all, their reputations, and therefore their profitability, depends upon proven records of accuracy. The fate of the Literary Digest poll, which predicted the overwhelming defeat of FDR in 1936, is indelible in the institutional memory of all polling organizations. Soon after that election, the Literary Digest ceased publication.

But an "accurate prediction" of an election presupposes honest elections. Thanks to "paperless" electronic voting, on machines operating with secret software, manufactured and programmed by private firms with Republican affiliations, U.S. elections are "faith-based." Are our elections honest and accurate? Unknown and unknowable. And the corporate media, both political parties, and the Congress are spectacularly uncurious and unperturbed about the insecurity of U.S. elections.

Furthermore, we now know that the corporate media print and broadcast lies (Saddam's alleged WMDs and involvement in 9/11, Al Gore's "invention of the internet") and fail to report essential truths (Bush's AWOL from the national guard, election fraud, John McCain's involvement with embezzler Charles Keating). So why assume that the same media publishes accurate opinion polls? And if the polls are not scrupulously accurate, this does not necessarily mean that their numbers are simply "made-up" on the spot. Deliberate sampling bias will suffice to yield the "desired" results.

So might it not just be possible that the covert function of opinion polls is not to "track" public opinion or to predict the outcome of elections, but rather to validate the predetermined outcome? Likewise unknown and unknowable.

If the major national polls are "in on" another fixed election, it would not be their task to report actual public opinion. Rather it would be to publish a "prediction" close enough to the outcome to make the theft plausible. (See my "The Fix is In, Again," and other essays on election fraud).

In the meantime, absent legal, legislative, and journalistic diligence, it is up to individual citizens and citizen organizations such as these (here, here, here, and here) to raise the question of election integrity, and to cite the abundant and growing evidence – anecdotal, circumstantial, and statistical – that during the past decade at least, the "will of the people" has not always prevailed in our national elections. As Republican Congressman Peter King carelessly blurted out on election night, 2004: "It's all over but the counting, and we do the counting."

Contrary to these dire, and possibly paranoid, suspicions, is this plain fact: There are numerous polling organizations, independent of each other. Some of these are affiliated with and sponsored by the Democratic Party. Thus it is highly unlikely that all of them would be complicit in a grand conspiracy to lie to the American public.

As I said at the outset, I have many questions, some suspicions, but no definitive answers.

But these are questions that all concerned citizens should be asking, even though the corporate media are not.

And if these questions indicate that the polling organizations have lost some of their former credibility, along with the media that publish them, they have only themselves to blame.

-- EP

[gateley]

1. I'm not going to believe or disbelieve them -

all we can do is keep on keepin' on.

I honestly have a belief that there are way more people than the pollsters tap into, who will show up en masse on voting day and handily put Obama over the top.

[RaleighNCDUer]

8. And will HOPEFULLY put Obama over the top.

The other side underestimated our strength in '06 and, despite record levels of electoral fraud, still lost control of congress. I hope they believe their own propaganda and count on women defecting from Obama and low turn out of the newly registered - if so, they will underestimate again and fail to cheat by as much as they need to.

But they got spanked in '06, and will be very leery of underestimating again. The electoral fraud is going to be MASSIVE. I think the turnout will take them by surprise and we will again see bizarre flips that are completely impossible to account for. But maybe it won't be enough.

[Pete2069]

43. Still there is something deeply wrong with our leaders.

Say for example our democrats were in a health crisis and
their life depending on the results which would come from the
electronic voting system , they have forced on our voters for
electing our leaders. 

Do you really believe they would trust the answers they would
receive from the results of our election process they have
force on us.

My question is,, are they being black mailed by this
administration's data mining , have to many ties to
corporations or ties to the Jewish organizations when are
pushing the policies of war against the middle east , immunity
for illegal spying by the telecoms (in which Jewish companies
were one of the top telecoms involved in the illegal spying on
Americans) , funding of these actions against Iran /Syria ,
really are not qualified to understand the legislature &
its effect it has one our country they are passing , afraid of
losing their jobs or like the blue dog democrats they are
really republicans in democrat clothing.

[buff2]

2. The media does this every election

Always close to the election they put the repuke ahead. That way they can sway the public better and then make it easier for them to steal it. It's the same ole shit year after year. :boring:

[kpete]

3. last 2 election cycles

Last 2 election cycles

I barely slept.

I checked every poll, every day.

Most were close, too close for comfort.

Both elections had the same outcome - ???????????????????


What do I seen in polls (remember, I do not have an analytical mind)

I see that symmetry is necessary in order to create the illusion

meant for our entertainment, perhaps, but more to keep us distracted.


peace, kp

[nxylas]

4. 2nd rec, but I have a question

Conservative will rightly object that the 80% figure for those who think the country is headed in the wrong direction includes those who think that it is not conservative enough and for whom the "right direction" includes outlawing abortion, teaching creationism in public schools etc. Do we have any reliable indicators as to what percentage of the population these people make up. My suspicion is it that it is much smaller than the self-styled "silent majority" would have us believe.

[DCKit]

10. The pollsters say people who identify as evangelicals make up 23%...

But they only polled evangelicals.

[polichick]

5. Excellent piece, as depressing as it is.

[katandmoon]

6. Believe them when they're favorable, don't believe them when they're unfavorable

That seems to be the rule.

[Sam Ervin jret]

9. or, you could be reading posts from "believe the polls" when they are ahead and posts from "don't

believe the polls" when they are behind posters and assuming they are from the same people.

Could you be reading the information and drawing incorrect conclusions?

This sometimes happens when you have a preconceived idea as to the meaning behind data.

[RufusTFirefly]

7. K and R. n/t

[groovedaddy]

11. The point on not polling cell phones is extremely important

I firmly believe youth and african americans will turn out in record numbers. This is not reflected in the polling. We will win states that aren't even considered in play AND ignored by the rebub dirty tricks operation.

[AZ Criminal JD]

34. Polling companies call cell phones.

This is not a factor.

[obiwan]

44. As far as I know, polling companies only phone land lines.

If you know different, please prove it. With verifiable facts.

[AZ Criminal JD]

45. I have worked on campaigns that use polling companies

I know how they get data. If living in fantasy gets you through the day, go for it.

[nichomachus]

12. Ignoring the polls is a huge mistake

We can all quote figures from morning to night that tell us the polls are wrong. However, the great mass of Americans know only what they read about the polls. So, if the majority of people go into election day thinking it's a dead heat, they will accept any outcome -- no matter how much we scream and yell.

The whole idea of "not being close enough to steal" is ludicrous. The right-wingers who own and run the voting machines -- without any oversight -- can create any outcome they want. They can make it close enough to steal. Or they can just steal it outright.

I know the Pollyannas will cry and whine and stick their fingers in their ears and shout "LALALALALALA." I'm sorry if this information is "a downer," but ignoring it won't make it go away.

[Joe Chi Minh]

23. Nobody's ignoring them. We know they're bent and we're angry, but

blogs enable us to spread the word to people who are impressionable in the face of the lies promulgated by news organs that purport to be quasi official.

Your first paragraph is the only one with a glimmer of sense. Then you spoil it by your failure to understand the now ever-increasing need for some measure of pretence, even by the Republican "electronic" fraudsters. Sorry to spoil your fantasy. Only even greater fraud could prevent a landslide dwarfing even Kerry's.

[nichomachus]

25. Actually, some people her are saying that they do

[Joe Chi Minh]

33. Unless the rest of their posts suggest otherwise, it's not difficult to

ascribe their import to be that the polls should simply not be believed, because they are palpably fraudulent; not that they should be of no cocern, for all their mendacity. Even when favouring Obama, it's highly unlikely that they would be giving the true picture of the extent of Obama's superiority.

Your post gave a regrettably confused impression of your concern, apparently advocating that Democrats should worry about losing the election, fair and square, on the basis of the picture given by the polls, which is palpably farcical. Confusion between the import of subject and predicate, sentence and sentence, paragraph and paragraph, seems to be a trick of last resort of Republican operatives on here, so you need to be clearer. Rancorous bombast intended to be satirical - the Polyanna nonsense - also doesn't reflect well on your motives.

[bemildred]

13. In a word "No."

They are hard to do right and easy to manipulate, to fudge the results in this or that direction. Unless you know the math and the details of the methodology used, you can't even begin to evaluate them. The very notion that the subject of the poll (say, "the american voter on such and such a date") is something that can be evaluated by such a method, with the attributes being studied, is open to doubt. Even given all that, interpretation is very much a matter of opinion and not fact.

For example, is there really such an entity as "The American People" and can it have opinions in the same way that you or I might have an opinion? The answer is clearly no. "The American People", to the extent it exists at all, is a constantly changing collection of persons, and what it has, if it has anything, is a distribution of opinions or the lack of one about any particular subject. You could theoretically make up a table of what everyone in the country thinks about "X", including an entry for "Huh?", and that TABLE would be "what the American People think about X" at that time, but NONE of the entries could be said to be what the American People think.

[WillyT]

14. Big K & R !!!

:applause::kick::applause:

[barbtries]

15. the polls are full of shit

in my opinion.

for the record, i NEVER believed that gw had a 90% approval rating.

[Lydia Leftcoast]

19. Remember when 90% of Americans supposedly approved of the Gulf War?

I was on my way to attend a teacher training seminar in Hawaii in the summer of 1991, flying on a 747, and a Middle Eastern-looking couple was sitting across from me. The flight attendant struck up a conversation with them and asked them if they were from Iraq. They said that they were from Iran. Well, this flight attendant seemed to think that they were the same, but obviously her heart was in the right place, because she said, "I never thought we should get involved in that whole Kuwait mess, but I kept my mouth shut, because everyone else was for it."

"I wasn't for it," I spoke up.

Immediately, everyone else who was within hearing range spoke up and said the same thing. This was on a 747, mind you, so that was a couple dozen people.

What are the odds of that happening unless the polls were garbage--or unless all the anti-war people flying to Hawaii that day were sitting in the same section of a plane that holds over 400 passengers?

It's also true that before the Gulf War, there were demonstrations with 10,000 or more marchers in most major cities in the U.S.

Ninety percent in favor of the Gulf War? I..don't...THINK...so!

[barbtries]

29. you're right

though i was naive and gullible enough to support that war. i regret it now, and doubt i will ever support another war in my lifetime regardless of how just it might be. i won't trust the people telling me that any more.

i put a little poll on my blog back in 2002-2003. not scientific or anything; it was just that in all the conversations i had with people, they all seemed to agree that gw was a moron. i ended up with about 70% responses in agreement with me. NOT approval for gw. i never believed the inflated numbers. i don't believe them now. we as a people are being played, as always i now understand, but i see it now. i am very suspicious.

[Spritz57]

16. My Observations on Polls and Elections

I'm wondering if the polls are being manipulated to explain
(or dismiss) the undervote or exit polling in November.  If
the numbers are close to skewed polling data then the
discrepancies can be explained away by the main stream media
as in 2000 and 2004.  Just my 2 cents. . .

[bemildred]

17. You will find friends here.

:hi:

[McCamy Taylor]

18. This is my suspicion. Bogus polls now to soften us up for the stolen election.

[Spritz57]

21. Which begs the question. . .

(I haven't seen a thread on this yet which doesn't mean one doesn't exist, I simply didn't find it), Which States have been proactive in creating paper trails for their electronic voting machines so in the event of a challenge a proper recount can be achieved? I know Ohio and California have made changes much to the chagrin of DieBold but what about the other 48? I thoroughly enjoy the forums, and have learned much from all the information shared, but if there still remain doubts about the accuracy of our votes then what is gained by being informed? I would love to see a thread, assuming one does not exist, on the progress or lack thereof taken by the states to assure our votes count.

[Voice for Peace]

20. I don't believe the polls. They just don't make any sense to me.

I think somebody's skewing them for diabolical reasons.

[RufusTFirefly]

22. Polls influence public opinion as much or more than they reflect them

There's a delicious ambiguity in the term "public opinion poll." Whereas most people assume that these polls reflect public opinion, they play an even bigger role in influencing public opinion.

Our current reliance on polling creates a dangerously closed system.

Polls affect a candidate's ability to raise money and play a key role in the determining how much corporate media coverage that candidate will get.

It's a great big Catch-22. Without money or coverage you are unlikely to poll well. But to get money and coverage you need to poll well.

Finally, there's the alarming bandwagon mentality of the American public (remember the stampedes for Cabbage Patch dolls?), which leads many to pull the lever for the candidate that they think everyone else is voting for. How do they know this? Even if they believe their choice is influenced by friends or family, the origin of the decision can often be traced back to polls.

As a result, a handful of people and polling methodologies wind up deciding our elections. Often by the time we cast our ballots, the outcome is already a fait accompli.

[nichomachus]

And the Catch-22 to the Catch-22

If you poll too well, the money dries up. People figure you don't need it. So, it's in a candidates interest too to have the polls close.

This is the same phenomenon that causes some "safe" candidates to get a ringer to run against them -- so they can raise more money.

[RufusTFirefly]

26. Excellent point

So, even though as you pointed out earlier, the election doesn't have to be close in order for it to be stolen (even if the victory storyline is ludicrous, if the MSM repeats it enough, most people accept it), a candidate's efforts to keep the money coming in (by keeping the election close) may ultimately make a stolen election that much easier.

[nichomachus]

27. Shhhhh

You're getting close to the truth here. The Corporatocracy doesn't like that.

[RufusTFirefly]

28. Of course, one way to keep the money coming in even it ISN'T close...

... is to depend on a handful of large corporate donors instead of on smaller, fickler, grassroots donations.

All paths lead to the Corporatocracy.

[Tutankhamun]

24. Either scenario sucks:

Possibility 1: The polls are rigged to soften us up for the third consecutive stolen presidential election.

Possibility 2: The polls are rigged to keep it interesting so we consume more MSM and they make more money.

Possibility 3: The American people really are doornail stupid.

I think we can rule out possibility 2. Most major media outlets wouldn't just blatantly fabricate poll results for a small, short-lived spike in ratings. It's just too great a risk for a puny benefit.

That leaves possibilities 1 and 3.

[tom_paine]

35. Or BOTH 1 and 3...

[Cherchez la Femme]

30. The MSM and polling companies

who, it is my firm belief, all are in collusion with the GOP, are setting up plausibility that it is a tight race
all so they can steal the election.

Again.

The data I saw on the 2004 election alone statistically proved, not only to myself but to many others, that vote flipping to benefit */Cheney had occurred. Prior to the election Bush was unworried and was sure the GOP would prevail. His nonchalance was not because he was in his bubble but because he knew that votes were going to be flipped to benefit the 'pukes.

The only thing is, in most of the places where they were effectively challenged (and perhaps all) they appeared to flip the vote by about 5%
which thankfully wasn't enough.


From what I understand, what the pollsters are doing right now by unequally representing the population (in a biased manner)

but then of course they could just be fudging the raw data. Or both.

[Vattel]

31. Here's an argument for the philosopher

Either the exit polls in 2004 were unreliable or the election was stolen. If the latter, then the polls are irrelevant. If the former, then because the exit polls are more the most reliable voting polls, the current polls are certainly unreliable and so are irrelevant. Conclusion: The current polls are irrelevant.

OK, it's got holes in it. By the way, it's ridiculous to think that all the various polling organizations are "in on it." You yourself seem to realize that. So why even raise the possibility? I know, Bush has made us all a little paranoid.

[dccrossman]

32. I look at the CNN Poll of Polls

Whether accurate or not or using an effective sampling strategy or not, I've liked watching this one.

It pulls together Gallup polls and other random polls from around the country.

In my opinion, even with being somewhat skewed because of the way they choose the sampled population, we've now seen the low point for Obama (42%, I think) and the high point for McCain (46%, I think).

We should now start seeing a gradual shift, on that poll, back to the roughly 48% Obama / %42 McCain, with 8-10% undecided.

And these polls definitely under-represent new voters and youth voters.

Key message: Get out the vote!

If we get out the vote, and the Thugs only fix enough to overcome the numbers in the polls, we should still win by a significant margin.

[Historic NY]

36. Most are being freeped big time....try changing one its almost impossible.

the creeps must have figured that crap out especially on AOL.

[p4poetic]

37. I don't even know what to think anymore

Honestly.

[Enthusiast]

38. The criminal Republicans might do ANYTHING.

Can you even imagine what they would not resort to. Think of the very worst thing. They would do it! Manipulating polls? That's probably an easy one if you control intelligence resources of the U.S.A.

These people are fascists, make no mistake about this. And they know they are fascists. The gloves are off.

[atjrpsych]

39. Interesting point, it would be nice too have a full report of their methodology

If you look at the polls, their sample sizes tend to be at the highest 3,000 people and on the lower end about 900. So, considering there are close to 300 million residents in the U.S (I don't know how many are registered to vote, or are too young etc...) this is an very small representation and may not be an accurate reflection. Especially when you take into account that it is more common for many to use their cell phones, and the high number of Dem's that were recently registered. Makes you question the validity and reliability of the polls statistical procedures?

[elleng]

40. A WASTE OF OUR TIME.

.

[lelgt60]

41. Noooo!!! Don't believe the polls...

1) Don't "believe" anything. Demand background facts, In this case: methodology and affiliation. You wouldn't trust a single medical study done by the company promoting a drug, why trust a poll done by someone with questionable affiliation. It doesn't mean they lied...just you don't know.

2) If you question the polls prior to the election, you also need to question the exit polls - same organizations.

3) The polls no longer seem accurate. There was a time over the last 50 years when they seemed really accurate (Read: predicted the outcome correctly - fair or not). But not in the last 10 years. Something has clearly gone wrong with the methodology. There's no basis to think they predict anything even mildly close. Whatever their motivation, they need to go back to the drawing board.

4) You need to point out both the corruption posiibilities as well as the methodology failures to everyone you can. It's up to the polling companies to prove themselves reliable. It sure would be nice to have statistics on their reliability.

So, forget the polls - work for Obama's election in all cases.

[happygoluckytoyou]

42. IF THE GOP WINS THIS ONE WITH THEIR LAME HORSES.... they deserve the barren country they create