The Exit Poll Controversy Continues

[BeFree]

There has been much written about the 2004 Presidential Exit Polls. And still, after all those words and arguments there remains much disagreement and a huge controversy.

Let us take a step back and look this over from the beginning.

The early numbers that came out on Nov. 2, 2004, all pointed to a Kerry victory. These early numbers were the product of a $10 million operation paid for by the media, and was conducted by a well known and trusted poller; his reputation is why the media people paid out those big bucks.

All day, up until around midnight, Kerry led in the exit polls by about 51% to 48%. At that time almost all the polls had been closed and official numbers came streaming in.

Once the official numbers had come in, to the surprise of the exit pollers, there appeared a great discrepancy between their numbers and the official numbers.

So the exit pollers began revising their numbers to match the official numbers.

But thanks to the internet, the exit poll numbers before midnight were captured and released to the world.

Then the world saw those numbers being tossed aside and replaced with the official results. In the process of replacing their own numbers with the official numbers, the exit pollers came up with several reasons for the replacements.

But those reasonings brought about a new controversy. It was pointed out that the new numbers were mathematically unsound.

So what did the exit pollers do? Did they stand behind their numbers? Did the use their $10 million product to say that questions need to be asked about the official numbers? No, they did not.

Instead, they began to claim that it was they who made a mistake. That as far as they were concerned, the official numbers were correct.

Well, friends, that left us in quite a quandary. We knew the machines were quite capable of producing altered numbers. We knew the Blackwells of the election offices would do what ever they could to keep democratic votes from being counted. And we have been proven correct.

We knew this administration would do everything they possibly could to keep and hold power, heck they unabashedly stole the election in 2000.

So we questioned everything. We, the netroots netizens, began exploring all the ways that the discrepancy could have occurred. We came up with hundreds of ways that could explain the situation.

We did so with almost no media help, very little government help, and no help from the exit pollers.

In fact, the exit pollers have denied us an examination of their product. They claim that there are certain secrets in their numbers that can not be divulged, leaving us asking questions with no good answers, and that is why the controversy about the exit poll continues.

[rock]

1. Right on!

If you're gonna replace your exit poll numbers with the actual numbers, why not save yourself some trouble and expense? In fact, can't Diebold just publish the numbers before the election and save everybody a lot of trouble?

[unpossibles]

85. it would not surprise me if Diebold announced they were getting

into the exit polling business. And maybe Haliburton and a few others too.

[OnTheOtherHand]

2. fact check

1. BeFree somehow manages to leave out the background information that the 1992 exit poll interviews were almost as far off as the 2004 interviews. The rhetoric about "reputation" is calculated to confuse people about the historical record.

2. "All day, up until around midnight, Kerry led in the exit polls by about 51% to 48%." This statement is meaningless. "The exit polls" were over well before midnight. BeFree knows by now how the projection process works.

3. "At that time almost all the polls had been closed and official numbers came streaming in." Actually, "official numbers came streaming in" long before midnight (as anyone who was paying attention on election night knows).

4. "Once the official numbers had come in, to the surprise of the exit pollers, there appeared a great discrepancy between their numbers and the official numbers." Actually, at 4:30 PM -- before the polls had closed anywhere -- the pollsters warned that they suspected "sizable discrepancies" in the exit poll results from nine states, simply because the results were so far from their prior estimates.

5. "But thanks to the internet, the exit poll numbers before midnight were captured and released to the world." In other words, CNN.com posted estimates as the polls closed.

6. "It was pointed out that the new numbers were mathematically unsound." Whatever that means.

7. "Instead, they began to claim that it was they who made a mistake." Actually, they said that apparently Kerry voters participated at a higher rate than Bush voters.

8. "In fact, the exit pollers have denied us an examination of their product." It's the same-old-same-old double-think: the insistence that the exit pollsters have provided the evidence of fraud and that they are concealing the evidence of fraud.

9. "So we questioned everything." I wish. As far as I can tell, BeFree never, ever questions his assumption that he knows what really happened in 2004, and he probably never will.

[BeFree]

3. Is that your best shot?

2. "All day, up until around midnight, Kerry led in the exit polls by about 51% to 48%."

This statement is meaningless. "The exit polls" were over well before midnight. BeFree knows by now how the projection process works.

The exit polls showed Kerry with a lead up til midnight, well after the exit polls indeed had ended, but before the official results had all come in. My claim is that after midnight is when the final exit polls began to shift from estimates and be altered to match the official returns. That is a fact, and integrity requires that such a fact not be twisted like you tried.

3. "At that time almost all the polls had been closed and official numbers came streaming in."

Actually, "official numbers came streaming in" long before midnight (as anyone who was paying attention on election night knows).

That's a fact. No one has claimed otherwise. The claim here is, again, that after midnight, two hours after the west coast polls had closed, enough numbers from the results were available to begin showing the discrepancy between exits and results. Again, why twist my words?

4. "Once the official numbers had come in, to the surprise of the exit pollers, there appeared a great discrepancy between their numbers and the official numbers."

Actually, at 4:30 PM -- before the polls had closed anywhere -- the pollsters warned that they suspected "sizable discrepancies" in the exit poll results from nine states, simply because the results were so far from their prior estimates.

Yet, at that time, the purchasers of the exit polls were claiming Kerry was headed towards victory. Here we see confusion reigns. The use of the word results vs. estimates is partly the reason. Better that it be said the estimates did not match the prior estimates. But we do not hear what that mis-match was, no, we are just told that the pollers found problems in nine states, not which nine states or what the problems were, just problems. Again, true integrity demands more info, not less, when such claims are made, especially when coming from professionals, don't you think?

5. "But thanks to the internet, the exit poll numbers before midnight were captured and released to the world."

In other words, CNN.com posted estimates as the polls closed.

Actually, CNN posted estimates well before the polls closed. But you have a point: I should have written it thusly: The exit poll numbers were leaked to the public before the numbers had been altered (after midnight) to match the official returns.

6. "It was pointed out that the new numbers were mathematically unsound."

Whatever that means.

Therein lies the controversy, eh? TIA has covered this element quite extensively, and a post of his will relating to that will be included later on this thread.

7. "Instead, they began to claim that it was they who made a mistake."

Actually, they said that apparently Kerry voters participated at a higher rate than Bush voters.

Yes, that is the hypotheses. A hypotheses that remains un-proved to this day, again feeding this controversy.

8. "In fact, the exit pollers have denied us an examination of their product."

It's the same-old-same-old double-think: the insistence that the exit pollsters have provided the evidence of fraud and that they are concealing the evidence of fraud.

No, and that is an ingenuous statement. To be honest such a reply would say that "You are right, and as long as we can we will keep you from looking at the data." Or "Yes, the estimates are evidence, duh, they showed a discrepancy, right?"

9. "So we questioned everything."

I wish. As far as I can tell, BeFree never, ever questions his assumption that he knows what really happened in 2004, and he probably never will.

There is an element of truth in that. We never will know what really happened in 2004 as long as certain data is kept from us. And another big part of that is that is no paper trail to examine. I wonder, however, why someone such as yourself has to stoop to such personal attacks to make his point?

[Melissa G]

4. Hi Be Free, As Dan Rather said in another thread..

'a diversionary technique'
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=364x1632741

In one of the more interesting exchanges, Rather talked about 60 Minutes's controversial broadcast about Bush's National Guard record.

King: "You're saying that was a fair report, I mean that was--you believe that report to this day?" Rather: Do I believe the truth of the story? Absolutely.

Rather added, "...We had a lot, a lot of corroboration..it wasn't just the documents. But it's a very old technique used when those who don't like what you're reporting, believe it can be hurtful, then they look for the weakest spot and attack it, which is fair enough." But, he added, " It's a diversionary technique."

[BeFree]

5. Hi Melissa G

Why? Why would they want to divert us from the truth? Is it because DU - ER has some power, and people read this little ol' forum? That ER is like a media source that needs to be diverted, like they did Rather?

I couldn't argue against that, heck we did make Salon a time or two, eh?

[mod mom]

7. DU is used by quite a few activists in the ER community to get information

out to others outside the list servs, as well as to the general public who is interested in what is being reported. The GOP utilizes distractors and mis-informers to undermine their opponents. We have seen this with campaigns and also in the election reform community.

[Melissa G]

50. so sad, but true, mod mom and Be Free.. n/t

[OnTheOtherHand]

56. could it be plausibly deniable swiftboating?

Seems to be flavor of the week.

[OnTheOtherHand]

6. well, yes, the exit polls are the weakest spot

You seem to be trembling on the brink of awareness. Don't fear the learning!

[Melissa G]

8. No , not the exit polls themselves, just the nits some people pick..

there is probably a fine future for those folks helping out school nurses though!

[OnTheOtherHand]

12. OK, but if you don't have an argument, then you don't have an argument

And I'm pretty sure that you don't have an argument, because a while ago I laid out what? eight or ten problems with Freeman's exit poll arguments, and you had no substantive response.

I have to admit that I don't understand how people can post on the topic of exit polls when they have nothing to say about it.

[Melissa G]

13. Otoh, you have been refuted..I've argued with you and find that you waste

a lot of time in non substantive arguments that don't illuminate much so i no longer waste my energy..But Dopp has a paper that she maintains you have never refuted at this link
http://electionarchive.org/ucvAnalysis/US/IncorrectElectionDataAnalysis-06.pdf
and hey, i just saw another thread refuting you over on PI where someone very bright and beloved did yet another refuting of you there.
Me, I'd rather go play with my family or actually do something about election fraud such as prepare to testify for the hearing on monday morning...

Have Fun!!!:hi:

[OnTheOtherHand]

14. well, no, I haven't

We knocked down that "paper," or whatever it is, point by point right here on DU.

For that matter, I have two working papers on my website that knock down the content of the first page, and she doesn't reference or rebut either one. Her "critique" of my AAPOR paper misrepresents and misquotes it. I am not impressed.

So if there is any point in that paper that you find convincing, why don't you state it in your own words -- or hers, if you prefer -- and I will respond to it?

Better yet, maybe you would take the time to read my paper for yourself and form an opinion of your own. Here it is again:
http://inside.bard.edu/~lindeman/beyond-epf.pdf

Meanwhile, as I said, if you have no argument, you have no argument.

[Melissa G]

15. I trudged though it a while back..not convincing..

There is a revision needed on your very first assertion which reads to me, like SPIN.
It is not a 'vocal minority' who believe the 2004 election was stolen. Since early May, at least, we have been the Majority.

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=364&topic_id=1155523

Unless, of course, we have touched on WHY you believe what you believe, which might be that your main news source is Faux news and that is why you draw the conclusions that you do..If so, THAT could explain a lot...

[OnTheOtherHand]

17. evidence?

Melissa, merely asserting that something is "not convincing" basically amounts to empty words. As I said, if you have no argument, you have no argument.

You do have one claim: if I understand it correctly, it's that most Americans(?), other than those who watch Fox News, think the 2004 election was stolen. I do not believe that, I don't think that an online opt-in panel poll of Pennsylvanians proves otherwise, and I'm not even sure that you really -- deep inside -- believe it yourself. (I can say that if as much as 39% of New Yorkers think the election was stolen, they are keeping very, very quiet about it.) If a reputable survey returns a similar result, I will be interested to learn of it.

And you have one ad hominem smear. Your posts have really gone downhill. You've shaken my faith in human nature and our common future, I have to admit.

[Melissa G]

18. You would be wrong, I do believe it...Even Here In RED Texas..

In reading your posts, I watched you attack Dopp saying she did not use a quote instead of a paraphase. Her paraphrase did in fact say much the same thing.

You tried to ding me for saying cite instead of footnote. These are BS kind of arguments. You just play word games which is why folks do not choose to argue with you. I did read your paper. I am capable of commenting. But it is not an interesting enough exercise to do so and you just bait people anyway.

For evidence of that check the thread the other day when you repeatedly did nasty summations of several posters who did not want to engage with you. They are all capable of doing so because I have watched them do so on many occasions. There is just no productive point..

I stood up for you when you first came here. I took a lot of heat for it. I had hopes we would get some good discussion from diverse viewpoints. In your case, my experience was this was not the case.

nighty night...

[Febble]

19. Let's be fair here Melissa

Dopp is playing a pretty nasty word game herself here. She spent an entire paper dissecting a statement she attributed to ESI and which they didn't in fact make (she amended it, then dissected it). She has repeatedly attributed similar statements to Lindeman and myself, even though those she attributes to us are not statements either of us have made either. Sure, she thinks they are paraphrases, but if you are going to use Math Logic to rebut a statement, you have to get it right. Dopp appears to have misunderstood a particular argument, and on the basis of that misunderstanding has posted copiously over the internet allegations that both of us are liars, and worse.

And yet although I have repeatedly asked her to correct her misquotations (not confined to this document) they still stand.

I confess to getting fairly angry and frustrated with Dopp. Yet I have never accused her of lying, and probably never will. I don't tend to assume that people who disagree with me are lying, although I am sometimes driven to conclude that they are willfully shutting their eyes and ears to evidence and argument that runs counter to their case.

I will post, FWIW, as reply to this post, a response I wrote to Dopp's latest document. It's heavy going, so it may help you sleep.

[Febble]

20. A response to the NEDA handout

Pages 1, 2 and 5 reference either Lindeman, myself, or both; I have also commented on pages 3 and 5. Page 6 refers to a largely methodological paper by Michael McDonald.

Page 1.

Dopp uses quotation marks to frame a statement attributed to Lindeman and myself (footnote 1): "If there is vote fraud, then there will be a positive correlation between Bush vote increase from 2000 to 2004 and the exit poll discrepancy." This statement does not appear in the source given, I am not an author of the source, and it is not true. It is crucially not true. As Dopp correctly demonstrates, it is perfectly possible for there to be vote fraud and for there to be no positive correlation between swing and shift. The relevant question is not whether such a counter-case it is possible, but how probable it is, given the variance in each variable. That, of course, is what is tested when a correlational hypothesis is tested. Indeed it is exactly what Dopp advocates testing when she suggests that fraud is indicated by a negative correlation between WPE and Kerry's exit poll share and/or by a positive correlation between WPE and Kerry's vote-share. It's how correlations work.

My argument is therefore not invalidated by Dopp's because she simply invalidates (correctly) an argument I do not make. What I conclude (by calculating the confidence limits of the regression line) is that the maximum shared variance between swing and shift is very small. However, translating this into votes is less easy, as the amount of variance shared will depend on the variance in fraud. If fraud was uniform in extent and magnitude then there would be no variance, and thus no shared variance. However, by the same token the correlations between shift and other factors from which others have deduced fraud would also be absent or meaningless. Fraud would correlate with nothing. Moreover, one would have to ascribe virtually all variance in redshift over and above sampling variance to non-sampling error in the poll. There is no problem with doing so, but it flies in the face of most fraud arguments, including many of those advanced by Baiman and Dopp. Practically pertinent, however, is the fact that a range of voting methods are represented in the poll, rendering substantial variance inevitable. In order to estimate the maximum number of votes likely to be stolen (for a given probability threshold) we therefore have to make some assumptions as to the maximum likely proportion of corrupt precincts, and also the likely variance in fraud magnitude. I have been modelling various scenarios, and I cannot squeeze anything like popular-vote margin scale theft into the plot without making heroic assumptions about the distribution and uniformity of the fraud, and the more generous my assumptions the more all other fraud arguments (Freeman's "neglected correlations" for example) are invalidated.

Dopp states on this page that "ESI's analysis method is equally invalid as a statistical model or trend <....>". She appears misunderstand the term "statistical model". The way correlational analysis works is that a "model" is fitted to data. This model is sometimes called the "model hypothesis" to distinguish it from the "null hypothesis". If the data is explained better by the model hypothesis than by the null hypothesis (as tested, for example, by an F test), the model is supported. If it is not, the null is retained. When the null is retained, it could well be that the hypothesis is true - in other words that what we are observing may be a "counter-example", to use Dopp's phrase. This is why, when we retain the null, we compute confidence limits for the best model fit. In other words, we do not claim (ever) that the null is "true" (actually a two tailed-null is never true). What we do is to compute the largest effect that is probable (to a given degree of probability, say 95%) given the data. As the statistical power in the Ohio sample is small, the confidence intervals are wide, and could accommodate fraud without stretching probability unduly. However, as the statistical power in the nationwide sample is large, it places strong constraints on the likely prevalence and variance of fraud, as I argue above. (Elsewhere, Dopp has accused Lindeman and myself of using the term "statistical model" as some kind of retreat from the word "hypothesis" - this, clearly, is not the case. All hypotheses in statistics are tested using a "statistical model". It's how inferential statistics works).

As for the word "trend" - this term is usually reserved for a model that narrowly fails to achieve signficance. There is no "trend" for swing and shift to be correlated, either in the ESI's Ohio dataset or in the nationwide dataset.

(I should also point out that the other apparently verbatim quotation on this page, attributed to ESI: "there is no exit poll evidence of fraud..." also does not appear in the source given, although unlike the statement attributed to myself and Lindeman, it is a reasonable paraphrase of ESI's conclusion.)


Page 2.

The idea behind the measure is not to "eliminate or 'straighten out' the influence of exit-poll-partisan-response bias". I refer you to my original paper (www.geocities.com/lizzielid/WPEpaper.pdf) and to the Lindeman, Liddle and Brady paper here: http://inside.bard.edu/~lindeman/ASApaper_060409.pdf. The idea was to find a way of quantifying "bias" (whether in poll or count) that would be invariant with regard to vote-share. It was not easy to do, but I think we managed it.

What Dopp's simulations demonstrate is the nature of the problem we were trying to solve. WPE has greater variance in the centre of the plot than at the end. For a given magnitude of "bias" (in count or poll) the WPE value is smaller at the extremes than in the centre. Our suggested measure ("tau") is uniform for a given degree of bias. However, because of the asymmetric nature of the Poisson distribution for all vote-share proportions except 50%, sampling error results in a non-symmetrical distribution of both WPE and tau at the extremes of the vote-share distribution, and in the case of tau, the variance due to sampling error a) increases at the extremes and b) the expected value of the mean departs from zero (the magnitude of the deviation being a function of sample size). This was why we developed "tau prime" and recommended that a Weighted Least Squares regression technique be used to tame the remaining heteroskedasticity. It has nothing to do with eliminating "the influence of exit-poll partisan-response bias". I can only think Dopp has not read our papers. I would also note, while I am on this page, that the paper she cites in footnote 5 as being a "mathematically correct way to evaluate exit poll data" advocates, inter alia, correlating WPE with exit poll share; as any error in the poll (including sampling error) will then appear on both sides of the regression equation, absolutely nothing can be deduced from such a correlation. Far from being "mathematically correct", this is a fundamental statistical error.

Dopp also alleges on this page that "Liddle also drops the most suspicious (and most indictative of vote miscounts) precincts from her calculations as "outliers". I am not sure what she is referring to here, and she has not provided a citation. I have not dropped any outliers from any analysis, although I have certainly conducted leverage diagnostics and outlier analyses, as any competent and diligent data analyst should.

Page 3.

NEDA's equations to derive Kerry and Bush exit poll response rates unfortunately do not work, for several reasons. One of these relates, again, to the asymmetry of the Poisson distribution when events are rare (such as encountering a Kerry voter in a high Bush precinct), and it can be readily shown that using the equations given, even if overall response rates were known with accuracy (which they are not) and for each precinct (the plot merely shows aggregate values), that sampling error alone will commonly generate impossibly high response rates (>1) for the group of voters who are in the minority.

It is simply not possible to infer differential response rates from these, or any, data. However, differential participation (or "representation") rates can be computed directly from the data (tallied responses for each candidate/tallied votes for each candidate), and the ratio between them is what I termed "alpha", from which "tau" and "tau prime" are derived. (See Lindeman, Liddle and Brady for further details.)

I would agree with NEDA that differential non-response may be a relatively minor factor in producing the overall discrepancy. The exit poll data are more strongly supportive of selection bias as the key to the discrepancy, and this will be reflected in the ratio between participation (representation) rates for each set of voters, as of course will vote-miscounts favoring one candidate. However, both differential non-response and selection bias may arise from the same latent variable - a greater willingness for Kerry voters than for Bush voters to participate in the poll.

Page 4.

The confidentiality issue is far more serious than Kathy appears to realise, and "vote-counts" can't be blurred, at least by Scheuren's method. "Vote-share" can be blurred by collecting data from a large number of precincts (all, I think), banding them by vote-share, and assigning each NEP precinct to the mean of the band in which it falls. I know of no method of blurring a quantity like vote counts that would preserve the statistical properties of the data, and yet render precincts unidentifiable. The more variables provided, blurred or otherwise, the greater the chance of precincts being identified.

As for the claim that

no scientific hypothesis can be considered proven before the supporting data are released to competing investigators

Firstly no scientific hypothesis is ever "proven"; it is merely supported. Secondly, in general, in science, investigators do not "compete" over the same data. If a hypothesis is supported by one group, then the challenge is for the same or other groups to replicate the finding on new data. In the social sciences, data are not generally released because of confidentiality issues. That is why methods and results sections of papers are so important - other investigators need to know exactly what was done and what was found. They do not need to do the same thing again, although reviewers might legitimately demand that other analyses be carried out on the data. However, I will agree with Dopp that the E-M evaluation document was not couched as a scientific report, and lacked the kind of information, such as t tests, F tests and standard deviations, that would be required to critically evaluate the document. That was one of the reasons I wrote my original paper.

Page 5.

Regarding the swing shift correlation, see my comments to page 1. Curiously, on this page, Dopp avers that

Baiman and Dopp never did any of the 'vote share/red shift analysis' Lindeman refers to because Dopp mathematically proved that “vote share/red shift” analysis is useless for analyzing exit poll data

even though at least two of the referenced papers (footnote13) plot linear trend lines between voteshare and redshift (the latter measured by WPE) and one states that "evaluating exit poll data" involves, inter alia, "Plotting the pattern of actual and significant precinct discrepancies according to their official vote share and exit poll shares" (my italics). I can only think Dopp has misunderstood Lindeman's statement, which remains perfectly valid. If a single counter example invalidates a correlational hypothesis, Dopp's method itself would be rendered invalid, as would her advocated correlation between redshift and exit poll share, were it not invalid anyway owing to the error term confound mentioned above.

I will take the opportunity to point out that Lindeman's paper does not characterise "anyone who believes that exit polls are correct as 'fundamentalists'". From his abstract:

Note well: “exit poll fundamentalism” does not refer to the hypothesis that Kerry received more votes, nor the belief or hypothesis that the exit polls evince fraud. These are empirical issues amenable to rational debate, and reasonable people may disagree. Still less does it refer to any and all criticisms of the 2004 election or of election systems. Exit poll fundamentalism as I have encountered it, and as I define it here, amounts to a closed belief system that forecloses further discourse and discovery"

(my bold).

which is a direct contradiction of Kathy's paraphrase.


edited for errors - I expect some remain

[BeFree]

31. Wow

That's a lot of big ten dollar words... it must be right.

:sarcasm:

[Melissa G]

44. Et tu , Febble... so sad..

I'm going to assume for the sake of generosity that you don't know that you are participating in something that is not held to be particularly attractive to discourse around here.

There is a form called tag teaming ( which BTW, has a rather unfair ganging up nature to it) that certain undesirable elements engage in to have a conversation no one is interested in having to bring up talking points that otherwise would be ignored. Said tag teamers brings their conversational partner to the board with them...

I hold, after argumentative experience with your good friend Mark L., that he is more interested in twisting words than having good discourse so to that end he is correct. I have no argument with Mark L, because it is my experience and opinion that he makes neither good arguments nor good discussion. There are plenty of other folks on the board for him to argue with so that is neither here nor there.

My experience of Truth is that it tends to be pretty self evident and does not remotely need the massive explanations and ten dollar words you and Mark L tend to come up with.

What you have just done is support him with a tag team data dump. I hope that was not your intention. It makes for bad, obscuring conversation.
You maintain that you are two separate scientists in a quest for truth. Yet the two of you jump on almost any poster either of you are conversing with and reframe and redirect back to your same old refuted arguments.

My experience is that you are two folks with the same agenda who try to bury any post about exit polls in massive, obscuring verbosity..Truth can be simple and truth can stand alone yet somehow in the posts that you and Mark have been coming up with simple Truth is not evident. And few, if any others are having an enlightening "AAAAHA" experience.

As I told OTOH, you two have been refuted. I personally, find it an uninteresting waste of time to point by point repeatedly refute you two but apparently, a lovely man that I admire does not. The last, easily understood, not at all complicated, refuting was done on July 13th on another site that you can easily find. It is titled
My response to OTOH and Febble's latest post on DU...

Have Fun!:hi:

[Febble]

48. Are you serious?

For starters, BeFree started this thread, not OTOH or myself.

OTOH disagreed with much what he posted.

You cited a paper (post 12) that claims to refute arguments made by Lindeman and myself. Not only that, but you actually took issue with OTOH's objection to a misquotation in that paper attributed to both of us (post 17). So I came in with a response post 18.

And you accuse us of being a "tag team"? What are we supposed to do - pretend that it is pure coincidence that Dopp makes up stuff and attributes it to us both?

You cited Dopp's paper as a rebuttal of Lindeman. Well, if it was a rebuttal of Lindeman, it was a rebuttal of me as well. So I posted a response I'd prepared to that paper. Is defending myself (and my colleague) against misinformation posted about both of us on the internet somehow against forum etiquette? And what the hell is a "tag team data dump"?

Of course we are separate scientists. I am not OTOH, he is not me. But we have made no secret of the fact that we have worked together, and indeed wrote a paper together.

If you can tell me what it is about Kathy's "refutation" of our work you find convincing, I'll attempt to explain simply why I don't think it is. But as you don't seem either to be able or prepared to do that, there seemed little point in editing my response for you. It can stay there for the benefit of someone who is actually interested, which at least you are honest enough to say you are not.

But if you aren't, I completely fail to understand why you would even post on the thread. Or are you content just that if an argument gives you the answer you want it must be right? And that anyone with a counter argument must "have an agenda"?

And no, I don't particularly feel like having "Fun", thanks all the same. Frankly, I'd rather have an apology. Failing that, an actual argument.

[Melissa G]

49. The last time we were discussing how you and OTOH worked together

and I was noticing that you and OTOH were sitting in for the Mitofsky viewpoint (you remember where someone got your way paid to the AAPOR conference, and as a doctoral student, you also amazingly had the discussant position, and Dopp had the email that the same financial was not done for Ron Baiman..) you were, in that discussion, as I understood it, not only distancing yourself from Mitofsky but putting great distance between your and OTOH's work. I was trying to treat you as separate entities BUT you keep constantly appearing together making the same points.. so, if you look at it from my perspective, perhaps you can see my confusion.

You can review your posts and see why I might come to the experience that you and OTOH tag team arguments. You can look at the amount of points that you want any individual to address when conversing with either one of you (which almost invariably becomes both of you) and see what I am talking about.
Tell you what, in good faith, if I find you have become aware of this behavior and stop persisting in it, I'll be happy to apologize.

As for counter arguments, I was quite specific as to which ones I found counter productive... Until you unhook from OTOH, You and I also have no arguments. (BTW, I did not address OTOH before he addressed me.) My impression of how you feel about arguing with TIA (which is that it is not your idea of fun) is how I feel about arguing with either of you on this topic.

Have a lovely weekend!:hi:


edit: clarity

[OnTheOtherHand]

51. but, Melissa, you apparently don't have any arguments

You can blame it on Febble and me if you like. Whatever gets you through the night.

For the record, it is a fantasy that we represent "the Mitofsky viewpoint." What I suppose you mean is that we don't think the exit polls indicate that Kerry got more votes -- and this is surely the majority viewpoint among social scientists who have stated any opinion on the issue. Your impulse to align us with Mitofsky, perhaps as an excuse to avoid engaging our arguments, does you no credit.

Also for the record, Ron's conference fee was waived; mine wasn't.

[Febble]

53. Well, it appears you didn't take in

what I said on that thread.

You make it sound as though five people were picked for the panel and I (undeservedly) got the plum. Well, that's not the way it worked. Four people submitted papers on exit polls and had them accepted, so they were put on the same panel. So the next step would have been to find an appropriate discussant. And I was invited to be the discussant. It doesn't matter whether I am a doctoral student or an alien from Mars - what matters is whether I was competent to be the discussant, and I was and am.

But I wouldn't have been able to do it without a contribution to my expenses, because I don't live in North America. So the convenor found me some funding.

Ron Baiman would also have been unable to attend without funding, so funding was found for him too. Rather than snipe at who got the most funding, I think that both Ron and I owe the panel convenor a debt of gratitude for enabling both of us to attend. Most attendees at the conference would have had to have paid their own way.

I have no idea what you mean by saying that in "that" conversation I was "distancing" myself from Mitofsky". I'm not a politician. I don't position myself. I'm a scientist. I make arguments. I cite data. I say what I think the data show. But as a scientist, my views are always provisioinal. If someone presents me with a convincing counter argument, or discrepant data, then I am likely to change my mind.

As for "appearing together" - it's certainly true that OTOH and I tend to end up on the same threads when it comes to exit polls. There is nothing sinister about that. We have worked so long together on the issue that at times I think we each tend to forget which of us had any particular insight at any given time. But we are certainly separate. I didn't even meet OTOH until AAPOR, and I am aware that he has been doing a lot of work offline on various issues that I am only tangentially involved with, and much more that I am not involved with at all. As you know, I'm not a political scientist, so my expertise is with nuts and bolts of the numbers, rather than with their wider contextual interpretation. Which is one of the reasons I have appreciated working so closely with a political scientist of Lindeman's calibre.

And as for making "the same points". Well, some of those points are important, we largely agree on them, and believe they are worth making. We both consider that Steve Freeman's analyses and interpretation of the publicly available data is seriously flawed. We also both consider that the complete lack of a correlation between swing and shift at nationwide level presents serious problems for his case that many millions of votes were stolen. He has not refuted this argument - indeed, when both Lindeman and I presented the argument to him in Montreal he conceded that he would have expected to see a correlation between swing and shift. The only explanations he has produced to account for its complete absence, to my knowledge, is that either the data are fake, or I somehow did the correlation wrong. Well, if the data are fake, then his own work founders; and there really is a limit to the scope for error in a simple partial correlation.

If you don't think it matters whether 8 million or so votes were stolen, or whether it is more likely that any theft was several orders of magnitude lower, then you are entitled to your position, of course. I happen to think it matters. And the only attempt at a "refutation" of my view that the most probable scale of vote theft is not in the millions is Kathy Dopp's Logic Proof paper, cited in the paper you cited. And for reasons I made clear in my earlier, rather heavy post, to you, I do not consider that paper a refutation. Lindeman has also written two papers explaining why he considers it is not.

I will explain again why not, as simply as I can. However, before I do so, I will note that as of yesterday, Kathy has removed the quotation marks from her footnote on page one. This is not an improvement. It now reads:

The ESI hypothesis was restated by Liddle and Lindeman: If there is vote fraud, then there will be a positive correlation between Bush vote increase from 2000 to 2004 and the exit poll discrepancy. http://inside.bard.edu/~lindeman/slides.html

She therefore still attributes the statement in question to Liddle and Lindeman, although her citation for it is to a piece by Lindeman (not Liddle). Her footnote now simply alleges that we restated the ESI hypothesis in words that neither of us has used. The footnote as it stands is now, therefore, simply a lie. Before its emendation I was prepared to accept it as an extremely careless mistake. It is now just false. We did not restate the ESI hypothesis as she states we did. The words in which she says we "restated" the ESI hypothesis do not appear in the source document attributed to us; and I am not an author of the source document. I am tempted to legal action.

And the reason the lie matters is this:

The restatement (which we did not make) implies that if there is vote fraud, there will necessarily be a positive correlation between swing and shift. This is not the case. Vote fraud will not necessarily lead to a positive correlation between swing and shift. If we had said that there would necessarily be a positive correlation between swing and shift, then a single counter example, whether hypothetical or real, would prove the statement false. But the statement is not ours.

And there are two important reasons why neither of us would make such a statement. Firstly, it is not how statistical inference works. We do not make statements of necessity. We make probabilistic inferences, and we calculate maximum probably effect sizes. Kathy appears to completely misunderstand this, to the extent of not even understanding the huge difference between what she alleges we have stated and what we have actually stated. And she takes what she thinks we must have stated, and "proves" that it is false. It is false. That is why it is not anything either of us have said. Or would say.

What Lindeman (not Liddle) actually says in the cited source is this:

Just as "correlation doesn't prove causation," lack of correlation doesn't prove non-causation. But the failure of Bush to do better in red-shift precincts than blue-shift precincts is a conundrum for analysts who see strong evidence of fraud in the exit polls. If fraud lurks in these scatterplots, it is very well disguised.

With which I am happy to concur.

But there is an important second reason why we would not make the statement Kathy states that we made. There are types and patterns of vote fraud that would not produce a correlation between swing and shift. These are worth exploring, and indeed both Lindeman and myself have spent a considerable amount of time and effort exploring them. My provisional conclusion is that they are highly implausible. I am willing to engage debate about this. But no-one from NEDA has even attempted to engage that debate.

In short, all Kathy has done is refute an argument that neither Lindeman nor myself have made. Our argument is not therefore refuted. Continuing to state that we have made a refuted argument that we have not ever made, in the full knowledge that we have not made it, amounts to lying. The lie, like a great many of Kathy's public comments about us, is a malicious smear.

To turn to your last point: "counter-argument" is not the same as "counter-productive". It may be a legitimate view that making counter-arguments to the case that millions of votes were stolen is "counter-productive" - perhaps it is more productive for activists to believe that millions of votes were stolen, despite the fact that there are strong arguments that the scale of vote theft is unlikely to have been on anything like that scale. I don't agree with that view, but I can, I suppose, see its force. For myself, I like to be grounded in reality. I don't find unfounded and hyperbolic claims useful, or energising.

What you have not not done is given me any counter-argument to my argument that multi-million vote scale theft is inconsistent with the complete lack of correlation between swing and shift. You have merely implied that to make my argument is counter-productive.

Well, I disagree. I'm happy to agree to differ. What I am not happy to do is to put up with the assertion that my argument has been refuted on the basis of a citation to a sloppily referenced document that contains lies about my own statements and those by others, by an author who seems to think that it is appropriate behaviour to devote entire threads on public forums to attacks on the moral and intellectual integrity of those she disagrees with, despite the fact that she has clearly failed to understand the most basic principles of inferential statistics, nor even the most basic academic standards regarding referencing and attribution.

[OnTheOtherHand]

21. huh?

OK, good to know that you believe it; now someone needs to demonstrate it. I could explain in detail why I don't believe it, but I wouldn't want to bore you.

Dopp quotes ESI as saying "there is no exit poll evidence of vote fraud," but they didn't say it. She quotes Liddle and Lindeman as saying "If there is vote fraud, then there will be a positive correlation between Bush vote increase from 2000 to 2004 and the exit poll discrepancy," but we didn't. She misquotes and miscites ESI another time just on that first page. She quotes me as writing "ESI never claimed to rule out vote fraud," which is an incorrect quotation (but more importantly, she is dead wrong on the substance there, as you know if you ever read ESI's work). She claims never to have done "vote share/red shift analysis," although I could point you to quite a few charts that have vote share on the X axis and red shift on the Y axis. That's a lot of mistakes to cram into just two short pages -- not that I don't think there are more.

I get spitting mad when Kathy Dopp makes up words and puts them in quotation marks, because it's totally unethical. It's even worse when the quotations bear no resemblance to people's actual beliefs, but it's bad even if they do. Paraphrases can't be presented as quotations. I get spitting mad when Dopp insinuates that Mitofsky has tried to rule out fraud, even though he clearly said that fraud may or may not have occurred. That's nothing to what I get when I read some of the stuff she has written about Febble on That Other Site. No wonder she was banned from here.

If you think I'm being nasty, complain to the mods, or call me on it in context. But to say, again, that well, people could make substantive arguments, but there's just no point -- well, again, if you have no argument, you have no argument.

[OnTheOtherHand]

35. why TIA's latest post is wrong

He thinks he read my PDF, but something important didn't register, or he is lost in profound denial.

If lots of people who voted for Gore say they voted for Bush in 2000 (and, to a lesser extent, people who voted for Bush in 2000 say they voted for Gore), then you can't use the rate at which the declared Gore voters defected to Bush to estimate the rate at which actual Gore voters defected to Bush. The declared Gore voters are likely to be more loyal than the actual Gore voters.

Frankly, this should be pretty obvious -- but I wrote a whole paper to spell out the details. He claims to have read it. What's the problem?

TIA actually writes, "THE PAST VOTE QUESTION IS IRRELEVANT" -- but he also writes, "All 4 National Exit Poll timelines had Gore and Bush 2000 voters switching at virtually the same 10% rate." TIA, is the question irrelevant, or isn't it? Please choose one.

[Bill Bored]

16. Self-delete. (Can't believe we are still wasting time on this.) nt

[BeFree]

25. Sorry to have bored you, bill. oh well.

[BeFree]

28. Too,

it's not like RFK was practically crucified for his representation of the exit polls, eh? Or that many others have likewise attempted to prove via the exit polls that the machines and republicans in general stole millions of votes.

Nah, none of that happened, it's not important. Nothing to see here, move along.

:sarcasm:

[OnTheOtherHand]

9. nope, sorry

"The exit polls showed Kerry with a lead up til midnight"

No, they didn't. You aren't referring to the exit polls -- you aren't even referring to the projections -- you are referring to the tabulations that happened to be posted on CNN.com. Words have meanings.

"after midnight, two hours after the west coast polls had closed, enough numbers from the results were available to begin showing the discrepancy between exits and results."

No, well before midnight, enough numbers from the results were available to begin showing the discrepancy between exits and results. Probably by 8 PM.

"Yet, at that time (4:30), the purchasers of the exit polls were claiming Kerry was headed towards victory."

Who?

"But we do not hear what that mis-match was, no, we are just told that the pollers found problems in nine states, not which nine states or what the problems were, just problems."

Surely you've read the evaluation report by now? Do you really think it's fair that you get to make everything up?

"Actually, CNN posted estimates well before the polls closed."

Well, they posted a national tabulation before the polls closed in all states. But elections are determined by state results, and CNN posted results for each state after the official poll-closing time in each state. (At least that's what they said they would do, and that's what they did in the states that I was monitoring on election night.)

"The exit poll numbers were leaked to the public"

I repeat: CNN posted results for each state after the official poll-closing time. There were earlier leaks, also, but they weren't crucial to the controversy. How many times have we gone over this, BeFree?

"TIA has covered this element quite extensively"

OK, good luck with that.

"No, and that is an ingenuous statement. To be honest such a reply would say that 'You are right, and as long as we can we will keep you from looking at the data.' Or 'Yes, the estimates are evidence, duh, they showed a discrepancy, right?'"

Huh? I have already set out dozens of times why the discrepancy doesn't evince fraud, and you refuse to look. It has nothing to do with withheld data -- that is just your cop-out.

If you want to argue that the exit polls evince fraud, you need to engage the arguments that they don't. I am not holding my breath.

[BeFree]

10. I refuse to look?

Just because I don't buy your theories - your used car, as I call it - you think I am not looking? I've looked it over, and I ain't buying it. But hey, if you think it's such a good deal, put it up out front; post a thread with your ideas, and let us have at it.

I imagine it would go something like this:

We know what we are talking about. The exit polls don't show anything but that they agree with the official, machine made results, showing that 62 million people were happy with bushco and he is so convincing that of the 16.8 million new voters, nearly two out of every three of them voted for the same old crap. Yes, we are the masters at this and the rest of you should just STFU and take our word for it, because we looked at data that you will never see, therefore we win!

Prove me wrong. Put it up front, take your best shot.

[OnTheOtherHand]

11. dude, I've been putting it up front for over a year now

You don't have to invent what I believe. You could read it.

EDIT TO ADD: As I just said in another post, it has nothing to do with "data that you will never see." There are half-a-dozen working papers on my home page now. Nothing secret in them.

[BeFree]

24. Up front?

That is dis-honest. You have never, that I can recall, put your opinions up front as an opening thread. What was it you used to say about intellectual integrity? I have seen very little of that from you.

What I wrote in the previous post is a summation of your story as I recall it. Honestly. Oh, you threw out a few other facts here and there but basically, your attitude is just this: You are a know-it-all.

Anyway, still, your response to my thread is mainly a personal attack, void of any coherent and reasonable response to the main idea that the exit polls before midnight detailed a Kerry victory. And that the alteration after midnight detailed a cover-up. What you have done is nit-picked your way through the complicated matter in an intellectually dis-honest excursion.

The OP has not been refuted, not by you or anyone else.

I challenge you again: put up a new thread with your arguments if you feel your's are worthy of discussion. That's pretty simple and straight forward, and a time honored tradition for DU. Why do you continue to duck such a simple practice unless you just wish to continue to hide down thread and mainly nit-pick?

[OnTheOtherHand]

26. what's wrong with yours?

I'm sorry, I think this is absolute crap. My posts have the same merit, or lack of merit, whether they open a thread or not. It's not as if you have any trouble finding me. It's not as if I don't answer questions when you ask them.

Why don't you write a 40-page paper and post it on your home page? Well, why should you? But I did that. I won't accept the canard that I haven't put my ideas out there. And I've put them here, too, and you know it. There is no need for you to imagine(?) I ever said that "of the 16.8 million new voters, nearly two out of every three of them voted for the same old crap." Heck, if you really think I said that, just link to it. Easy.

I don't bother to archive personal attacks, so I can't throw all of your attacks on me back in your face, but I assure you that my conscience is clear on that score.

Now, for those who haven't been through this at least a dozen times before: the tabulations are periodically updated to conform with the estimates, because that is assumed to yield more accurate results. You might find the fourth paragraph of http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=439786&mesg_id=440133 helpful in this respect. I put a worked example at http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=132&topic_id=2490768#2499695

And I say again, if updating the tabulations on cnn.com -- after having posted them for hours on end -- was a "coverup," then it was one of the most inept coverups in history. Not even worth talking about.

Now, you accuse me of having the attitude that I am a know-it-all. But consider this, BeFree: you couldn't find it within you to say that Febble is honest. Heck, I will go out on a limb and infer that you don't think Febble is honest. And why is that? Apparently, because she disagrees with you. So, I ask you, who is the know-it-all?

"The OP has not been refuted" = "I disagree." OK, fine, you disagree. You have no answer for 1992. You have no answer for New York. Actually, I don't know if you have an answer for any of my comments. How would that be different if it were on my thread instead of yours? It wouldn't. Don't kid yourself. Don't start a thread about the exit poll controversy if you aren't prepared to engage in it.

[BeFree]

27. There you go again

Another incoherent rambling crusade to save your reputation.

You, sir, are the supposed intellectual, professional scientist, yet your words read as if written by some high school kid making his last argument on the debate team..

Geeez. Just look at this! Nananana!

""The OP has not been refuted" = "I disagree." OK, fine, you disagree. You have no answer for 1992. You have no answer for New York. Actually, I don't know if you have an answer for any of my comments. How would that be different if it were on my thread instead of yours? It wouldn't. Don't kid yourself. Don't start a thread about the exit poll controversy if you aren't prepared to engage in it."

WTF?

So, you are ducking my challenge to be up front with your opinions? Ok fine. Its alright, really.

[OnTheOtherHand]

29. well, thanks for quoting me accurately

and, indeed, not answering -- which seems to make my point.

[BeFree]

30. What is your point anyway?

Is your point that bushco won the election fair and square? Because that is precisely what your effort, to me, boils down to.

My effort, on the other hand, was a simple narrative of the situation as I see it. And one which incorporated several items, the likes of which, you have helped me to understand better over the last year. We thank you.

You obviously disagree with my simple narrative, but I wonder if anyone else, besides your sidekick, Febble, (who, I guess might be honest, dunno, don't care, does it matter?) understands the point you have tried so hard to make?

I give you, at last, an 'A' for your effort. You deserve it, profoundly. Take it, embrace it, sleep with it, and you will feel better in the morn'.

But what is your point? That, is the 64 dollar question.

[Febble]

34. Yes, it matters

whether I am honest matters for two reasons:

One is that I mind being accused of being dishonest, when I'm not.

Two is that if I'm honest, it might be worth evaluating what I can tell you about the exit poll data.

But what matters rather more is your apparent belief that anyone who thinks the exit polls aren't evidence of fraud must be arguing that "Bushco one the election fair and square".

I do not believe the exit poll data are evidence of fraud; in fact I think that they actually contra-indicate massive (multi-million vote scale) fraud.

But I do not believe that "Bushco won the election fair and square".

These two beliefs are not contradictory. What makes you think that they are?

[BeFree]

36. No, it doesn't

What matters is that your theory is unproven. And that just because you haven't found something doesn't mean it doesn't exist. It has nothing to do with honesty, it has to do with facts, science and provability.

You have not proved your case, and as long as we can't see how you arrived at your opinion, it will remain so.

What we saw is what we saw. That's a fact.

We know how you feel about the stolen election. Very good. But my question was for your sidekick, where did he go? Is he ducking the question?

[Febble]

37. I've told you exactly how I arrived at my opinion

and I actually showed you the data. No, I have not "proved" anything. What I have done is to demonstrate what the data show - that redshift is not correlated with swing - and to try to explain the implications of this finding.

They are that if fraud of the type that would cause redshift (essentially vote-switching, or differential destruction of Kerry votes) were prevalent in the election, it is difficult to postulate any plausible mechanism that wouldn't also produce a positive correlation between swing and shift.

Not impossible, but difficult. I'm happy to discuss with you possible ways it might have occurred, but they do present major credibility issues. Perhaps you can find a way out of them. That's how science works.

And I think if you do a search of OTOH's posts you will find an answer to your question.

[BeFree]

38. That's all fine and dandy

And I am sure you did your best. But, after all you were working against impossible odds. Those odds? The early estimates which showed a Kerry victory.

You did a study using data otherwise unavailable to other researchers. You did not show me that data. Honestly, did you think I'd let you slide on that? "I actually showed you the data". The data in question is that which you have said can't be seen by anyone but, well, you, you and Miscountski. Not even your sidekick has seen it. I have not seen it. Neither has Freeman, Dopp, Baiman, etc.

Off hand, I see the data from places that in 2000, bushco lost on paper ballots, but won overwhelmingly in 2004 on DREs. In those same places senate races showed the same correlation from 2000 to 2004. It was a redshift from paper to machines for bushco alone. Them's the facts. I've asked you to look at those cases and all you said was that is was a good idea to do so. Hmmmph.

How nice of you to protect OTOH. But I prefer a man who can speak for himself, up front.

[OnTheOtherHand]

39. I already did. Doh. n/t

[BeFree]

41. But you can't repeat it?

Cat got your keyboard?

You can only type on line at a time? I see a lot of those from you.

Its OK, we can guess why.

[Febble]

42. Well, while you are waiting, how about

an answer to my question in 34?

[BeFree]

45. See # 36

Copied then pasted below. See, OTOH? If you already did answer, you can just copy and paste. It's easy.

But then, maybe you are still looking for the "Great White Whale" amongst the papers crowding your desktop?

-----------------
Answer to Febble.
36. No, it doesn't

What matters is that your theory is unproven. And that just because you haven't found something doesn't mean it doesn't exist. It has nothing to do with honesty, it has to do with facts, science and provability.

You have not proved your case, and as long as we can't see how you arrived at your opinion, it will remain so.

What we saw is what we saw. That's a fact.

We know how you feel about the stolen election. Very good. But my question was for your sidekick, where did he go? Is he ducking the question?

[OnTheOtherHand]

55. how does that answer the question?

Febble asked, "These two beliefs are not contradictory. What makes you think that they are?"

And your "response" is, "No, it doesn't"

You apparently consider this response so cogent that you brag about your ability to copy and paste it.

Do you really wonder why we get sick of repeating ourselves to you?

[BeFree]

57. Tag teamers again

First, I asked YOU what you believe about fair and square, you never replied, but your tag team partner scribbled the following:

I do not believe the exit poll data are evidence of fraud; in fact I think that they actually contra-indicate massive (multi-million vote scale) fraud.

But I do not believe that "Bushco won the election fair and square".

These two beliefs are not contradictory. What makes you think that they are?

So, I figure that is the question? That is what has your panties in a twist? Hell, son, of course I don't think the two statements are contradictory because, mainly, they deal with two different subjects. Thats why I avoided that part and focused on her honesty. Honestly.

But that's just Febble talking. I asked you, otoh, if YOU thought bushco won fair and square. Easy, simple question. One you have mightily avoided.

Anyway..... why don't yall tag up and tell us how it is yall believe that bushco did not win the election fair and square. Either you believe he won fair and square, or you don't.

[OnTheOtherHand]

58. so you meant to answer "No, they aren't"?

Because I daily have to ask myself, why bother to attempt to communicate with a person who apparently doesn't care if he can be understood?

"No, it doesn't" isn't a response to "What makes you think that they are?"

"Hell, son, of course I don't think the two statements are contradictory because, mainly, they deal with two different subjects."

Excellent. So, why are you bringing up an apparently irrelevant question on your own exit poll thread? Because I daily have to ask myself, why bother to attempt to communicate with a person who seems to change subjects at random?

No, I don't think that "bushco won fair and square." Easy, simple question; easy, simple answer. I bet you won't be satisfied with it, which is why I didn't bother in the first place. Because I daily have to ask myself, why bother to attempt to communicate with someone who considers me the enemy?

[BeFree]

59. Good. Thanks

I figured you didn't think bushco won fair and square. It just took a few posts to get that out, but it was worth it. I have patience.

Ya know, you do have so much knowledge and wisdom that you are all too willing to share, that I figured: Hey! Why not get that point clear? Because, honestly, I just wasn't sure you thought that way, ya know, that bushco didn't win fair and square.

So, if you don't mind, could you relate just how you don't think bushco won fair and square? Don't go out of your way just for me and peel off some rambling 1,000 word essay on how. Really, a small bulleted list would be enough.

Just as an opener, and Lawd knows you think I'm as wrong as could be about this, I think bushco mastered the machines to steal the election, so of course that won't be one of the bullets, I know.

Heck, maybe you have just one theory on how it wasn't fair and square? I just don't know.

Too, if you don't mind, how about starting a new thread with your theory(ies). I know, I've asked a few times already for such a minor favor, and please excuse me for asking again, but I'd treasure seeing your name up on the big board.

[OnTheOtherHand]

60. how about you defend some of your args on your own thread?

I posted a response, you and althecat responded, I knocked you both down, he disappeared, and you seem very eager to change the subject. Guess how I am scoring this.

1992? New York? Bush doesn't do better where the red shift is bigger?

I'm actually patient too, believe it or not. And in this case I strongly suspect that I will be waiting forever.

[BeFree]

61. He shoots! He scores!

Is this just a game to you?

Because, if it is, then it all makes sense.

So here we go again, sorry to bore you bill, but what reasons do you have to think bushco did not win fair and square?

But I will try to reply to this post, but first I am wondering what you mean by this sentence:

1992? New York? Bush doesn't do better where the red shift is bigger?

Here is my immediate response to that sentence:
1992? It was a good year.
New York? It is a northern state, in the US.
bush(co) doesn't do better where the red shift is bigger? Now that, that is a hard one. Because from where I sit, bushco does do better where the red shift is bigger. Isn't that what red shift means? That there is a noticeable shift from dem votes in 2000 to bushco in 2004 in same precincts.

Say, about your scoring, have I got any points yet?

[Febble]

62. Terms need explaining

"Redshift" was Jonathan Simon's term for the degree to which the count was "redder" than the poll.

So "shift" in the exit polls is the shift towards Bush from the poll to the 2004 count.

"Swing" was a term I used (it's a British one) for the degree to which the 2004 count was "redder" than the 2000 count.

So "swing" is the swing to Bush from the 2000 count to the 2004 count.

If fraud was responsible for both "swing" to Bush from 2000, and "shift" to Bush from the poll, they should be correlated. And they aren't. So the question is: how could this be, if both "swing" and "shift" had the same cause - fraud?

There are some interesting possible answers.

on edit

For clarity, rephrase: if fraud was responsible for "swing" to Bush from 2000, and also for redshift in the poll, you'd expect to see bigger swings where there was bigger redshift. And you don't. That's what the plot is about.

[BeFree]

65. Thanks

See, that's why it is is important for you two to keep playing.

Yall cleared that one up, in double time, I might add!

I stand corrected. Swing and shift are not the same, but you got my gist.

So how would one explain this hypothesis:

A place with paper - punched - in 2000, Gore won. But in the same place with a DRE- no paper - in 2004, bushco won by by 10%.

Is that a red swing, or what?

And if I could get a peek at the raw data, I could tell ya if there was red shift or not.

Are there real life examples of other places that had a 10% swing either way? 'Cause 10% is a lot, and comparing the swings (red and blue) with changes in voting systems seems like a way to prove the case.

I imagine I am not the first to think of this, heck no, I'd be willing to bet yall already did such a study, eh?

[Febble]

70. That's a very good thought.

But it would be SO nice if it came without snark.

I think it would be an excellent idea to see whether precincts or counties that changed from paper to DREs between 2000 and 2004 had greater swing to Bush than other places. I think OTOH has offered to help with that if you really want to pursue it. I think it is exactly the sort of work that needs to be done, and FAR more informative than anything the exit polls can tell you from a handful of precincts in each state.

[OnTheOtherHand]

63. OK, that's something

Yeah, obviously this isn't a game to you, accounting for these content-free responses. I actually care about the exit polls; I am still foolishly waiting for some indication that you do.

"1992? It was a good year."

Not for the exit polls, it wasn't. I hear over and over again about how accurate exit polls are, but when 1992 comes up, people mumble something and change the subject. althecat's version was, "The 1992 poll was nowhere near as large and as sophisticated as 2004s effort." Hmm. I always thought the talking point was supposed to be that the exit polls have a record of uncanny accuracy, not that we're pretty sure they became accurate sometime after 1992. (By the way, the 1992 national subsample has n = 15,490, so it's actually bigger than the 2004 national subsample.)

"New York? It is a northern state, in the US."

Where the exit polls projected Kerry winning by about 30 points, contradicting both official results and pre-election polls. So, did Kerry have a massive last-moment surge wiped out by widespread rigging of the lever machines? I don't think so, but if you do, won't you kindly make some effort to convince me that you take this exit poll result seriously?

"bush(co) doesn't do better where the red shift is bigger? Now that, that is a hard one. Because from where I sit, bushco does do better where the red shift is bigger. Isn't that what red shift means? That there is a noticeable shift from dem votes in 2000 to bushco in 2004 in same precincts."

No, that's not what it means. That is not remotely what it means. Red shift refers to the discrepancy between the exit polls and the official returns (or, rather, discrepancies in which Bush did better in the official returns). This isn't something I made up. Actually, I'm not sure who first came up with it -- maybe Jonathan Simon* -- but here is Alastair Thompson using it back in November 2004: http://www.scoop.co.nz/stories/HL0411/S00270.htm

This is the starting point of the exit poll controversy. How can you not know what it means?

*EDIT TO ADD: OK, I guess Febble is confident about that one. I think she paid a lot more attention to the exit polls back in 2004; I was mostly looking at Florida returns.

[Febble]

64. I think Freeman told me it was Simon's term

but now you mention it, I'm not so sure. It's a neat term.

[eomer]

77. Febble, have you looked at the following?

Is there a correlation between swing and change in red shift? In other words, instead of using just red shift, try the change in red shift (from 2000 to 2004, for example) instead.

If historical fraud is causing some of the historical red shift then you might expect that when fraud bumps up in a precinct (a perpetrator moved to a different county due to personal reasons) then the vote would swing and the red shift would swing too. Official count swing and red shift swing would travel together, to some extent.

(By the way, if "red shift swing" becomes a recognized term, did I just invent it? Or was there a song in the 1930s by that name?)

[OnTheOtherHand]

78. PMJI, I think Febble has left the building

(I.e., I think she is traveling for a few days. Given the state of the VTUSA threads, she is escaping just in time.)

My understanding is that Mitofsky sometimes reuses a state's master sample in two consecutive federal elections (presidential/midterm), but doesn't reuse the master sample in consecutive presidential elections. I really doubt that he reuses the interview subsample. So, unfortunately, I don't think it is possible to conduct that analysis at the precinct level.

We can try it at the state level, sort of. The only historical measure we have for red shift is mean WPE, which I don't love (and of course it isn't very "historical," but it does go back to 1988). It's flatline for 2000 to 2004 (an insignificant r < 0.1, actually flips to insignificantly positive if you remove VT). It might be interesting (although I don't know how revealing) to extend the analysis backwards.

Even though I think the exit poll data are full of noise, I still harbor the hope that they might contain _some_ signal along the lines you mention. But the data are so noisy, I'm not sure anyone would believe me if I found the signal. (I look now and then, on the thought that the process of looking in one place will spark new ideas about how to look in other places.)

I felt certain that you had invented "red shift swing," but I guess some Norwegians got there before you: http://www.boomkat.com/item.cfm?id=15315 (I don't know if they were first).

[eomer]

79. What about when you roll it up to county level then?

Under the theory that fraud can be implemented for a whole county, sometimes. Maybe this one could be done for only the more populous counties.

Or a slightly different one: is there a correlation between red shift in 2000 and red shift in 2004, at the county level? I'm not sure what it would tell us if found but, like you say, any clue might lead us somewhere useful.

I was mindful also of the noise problem but, if we can search for signs of alien life in the radio noise of the cosmos, why not keep trying to get a glimpse into the exit poll black hole?

That reminds me that red shift is of course a term in astronomy, and here's discussion about who invented it:

The earliest occurrence of the term "red-shift" in print (in this hyphenated form), appears to be by American astronomer Walter S. Adams in 1908, where he mentions "Two methods of investigating that nature of the nebular red-shift"<2>. The word doesn't appear unhyphenated, perhaps indicating a more common usage or its German equivalent, Rotverschiebung, until about 1934 by Willem de Sitter <3>

http://en.wikipedia.org/wiki/Redshift

I like Adams' phrase: "Two methods of investigating that nature of the nebular red-shift".

[OnTheOtherHand]

80. nebular, nebulous....

In fact, I've encountered quite a bit of alien life while peering into the exit poll black hole, so one just never knows. ;)

Red shift at the county level: serious problem of micronumerosity. In Ohio there are many fewer precincts than counties. Even aggregating into regions (which is probably not for the faint of heart), well.... Dunno, it might just be emotional exhaustion on my part. It does seem to me that if there is a "huge" story here, there should at least be hints of it in the historical state-level analysis (and maybe there are, if I go back further). Do you think I'm missing something? I cheerfully admit that I haven't defined "huge." Yes, conceivably fraud-purveyors could move from county to county, but wouldn't that show up strangely in the election returns even without trying to fold in the exits? (Don't just get stuck on the part of that q. that deals with the exits; you/we might think of some analysis of the returns that no one ever did before.)

In that vein, something I would like to do (to the extent possible) is to look at correlates of change in county-level residual voting. Getting a maximum-feasible-complete database of that seems to be a project in itself, unless Charles Stewart posted 2000 data and I didn't notice (quite possible).

BTW I swear I still haven't forgotten Ohio; I'm just juggling too many projects (way too curious for my own good). I've made a start.

[eomer]

81. Yes, I'm sure you're right (about micronumerosity).

I've been keeping in mind the issue of how to distinguish redshift that is due to repeated fraud and would therefore not correlate with swing for a given 2-election comparison. Some of the posts in this thread gave me those ideas, but I guess you're right that the data is insufficient to support the tests I came up with. I'll keep pondering it. Or maybe I'll try pondering the cosmos and see if it will just come to me.

[OnTheOtherHand]

82. just to repeat myself ;)

keep thinking about shift, but also think about other thing that might be expected to covary with swing. Change in residual vote rate is definitely an interesting candidate. There are some folks working on that sort of thing; I honestly haven't looked closely enough to know whether they seem to have picked the bones clean.

[eomer]

83. Yes, I'll think along those lines.

I wonder if the ratio of votes to signatures can be captured at either the precinct or county level.

If so, then we could check whether the change in that ratio has covariance with swing.

[OnTheOtherHand]

84. check with Time for change re: Ohio

My impression is that the Ohio investigators found it slow going to get the signature figures even for a relatively small sample of precincts in Cuyahoga in the last election -- not so much because the BoE was being uncooperative (although many say that it is), but because there is no SOP of tabulating those figures county-wide, so someone has to do it by hand. But check with him, and maybe ask around about some other jurisdictions. (Some states and counties gather much more data than others; I find myself very skeptical that signature counts are widely available, but I don't know.)

At the same time we think about what we wish we knew about the past, we are thinking about what we could hope to know about the present and future... umm, wow, I seem to have tuned into the Cosmic Channel. But seriously, I think the signature counts are on a certain sunshine's list of data to collect in the future, and it makes sense.

[Febble]

86. Yes, I did look at that

I could only baseline it to the state mean WPEs, but that at least gave me some sort of ballpark as to how far the 2004 precinct level discrepancies had swung from a 2000 baseline. But "change in redshift" was no more correlated with swing than redshift was.

But I'm glad I got to that one first! I am still tinkering with your last suggestion....

[BeFree]

66. Two points for snarkiness!

I didn't teach you that, how'd you get so good? Gosh, if I did, I sure am sorry.

New York, again? You still haven't figured that one out? What did Mitofsky say? Did he beat himself up over that one too?

Who did that poll in 92, because that would have a lot to do with the quality, and how do we know it wasn't stolen? Or had high levels of fraud? Maybe the CIA wanted bush out? Just cause billy won doesn't mean squat, wait, it might mean squat because the polls didn't do a red shift did they, a shift like 2004's where it all went from the winner to the loser, which is, I am sure you would agree, highly improbable, ya know, Kerry won!, wait, no, he lost, whereas in 92 Clinton won, no, wait, yes, he really won!

What are the odds an exit poll would get it ass backwards? At least in 92 it was bubba all the way! What are the odds an exit poll would get it right?

So, hate to be such a bug, but what reasons can you give to support your opinion that bush did not win fair and square?

[OnTheOtherHand]

68. that's not snarkiness, that's reality

By the way, this is a thread on the exit poll controversy. At least, that's what it says in the OP. But you apparently don't know or care about the exit polls. So I am pissed, of course. And no, you will not waste my time by changing the subject.

Even though, predictably, you didn't answer any of my questions, I will answer one of yours. We don't know that there wasn't fraud in the 1992 election. However, no one who claims to believe that exit polls are accurate has done any work to show that there actually was, or might have been, not only fraud in the 1992 election, but fraud that would explain the exit poll discrepancy.

Even weirder, there hasn't been much attempt to prove that fraud in the 2004 election could explain the exit poll discrepancy. Which brings us back to New York.

You said something smart in #65: if you identify the counties that switched to DREs between 2000 and 2004, you can compare the election returns to see if they are anomalous. I will help, if you want.

[BeFree]

69. Going for a three pointer, eh?

Oh, I know snark, and that was snark.

You are so off base with most all of what you just wrote that the officials called a walk on you. Its my ball now....

And I'm taking it and going to get something to eat. C'ya.

Oh, wait, what is your reasoning for thinking the 2004 elections were not won fair and square?

Gotta run.

[OnTheOtherHand]

71. oh, yeah, right

(Note: the preceding was snark.)

"Gotta run."

Well, that part rang true, at least.

EDIT TO ADD: Oh, by the way, the exit polls did a red shift in 1992, and they did a red shift in 2004. The main difference is that 2004 was close, and 1992 wasn't. The fact that the red shift in 1992 doesn't alter the outcome is one reason to think that it probably doesn't evince CIA rigging.

What really strikes me is, as I said, when one points this out to folks who claim to believe that the exit polls are accurate, they change the subject, exactly as you are trying to do. People who claim to believe that the Bible is inerrant, when confronted with 'problem texts,' often do the same thing. Often they don't seem to care about the meaning, they just want to defend the dogma. That's how I see people who claim to believe that the exit polls are accurate, but aren't actively looking for massive fraud in 1992, or in New York. That's why I finally started referring to exit poll fundamentalism (which could be seen as a rip-off of Bill Bored).

And you seem to have skipped over the most important part, which was dead serious: if you want to look for election fraud on DREs, you can do it, and you don't even need access to the exit polls. Just find, systematically, the places that switched over between 2000 and 2004, and see if their results went screwy. I don't have all the right data, and you wouldn't believe me anyway. Do it yourself. Walk the talk.

[OnTheOtherHand]

54. now, here's an odd sort of tag team

Melissa is trashing Febble for using too many words, and you are trashing me for using too few. Very compelling stuff there.

You want the exit polls to be accurate, but you can't account for their inaccuracy in 1992, and you can't account for the evidence that they were inaccurate in New York -- or, more generally, the evidence that Bush didn't do better in states with bigger exit poll errors. That's just two points. You rarely deal with any points. When confronted with evidence, you go back to complaining about secrets. Which is why I say, again, that the problem isn't the secret data, it's the data you already have that you refuse to look at, or to face up to.

But you knew that.

[BeFree]

67. Melissa!

Dang it Melissa, I'm gonna have to get a new tag team partner, you are lousy. You can see how it is done, this thread with these two is a good lesson, lesson up girl or you are fired!

[Febble]

72. That's right, BeFree

You tell her. Not much fun playing doubles if one of the foursome doesn't show up.

But I'd really like to see you and OTOH on the same team, figuring out whether voting-method change correlates with swing.

If you get a result, I shall be the first to cheer!

OK, I'm off for a couple of days. See you when I get back.

:hi:

[Melissa G]

73. LOL! Be Free!! I like talking to you but I'm not inclined to waste my

keystrokes/ breath talking to certain people about their imaginary points in what I consider irrelevant arguments that we had OVER A YEAR AGO... I told them this and sent them off to play with folks who did want to pursue playing their word games..they used the opportunity to post a bunch more of their stuff that I had no interest in. They seem desperate to get it out there.:shrug:

I told said persons where it was refuted not because i was attached to anything but rather to redirect them to folks who were ACTUALLY INTERESTED in talking to them about this topic WHICH WOULD NOT BE ME, (maybe if i cap this they will get it this time)...:shrug:
They latched onto Dopp and wanted to engage ME, WHO HAS NO INTEREST IN TALKING TO THEM about this topic since i think their construct is a big bunch of hooey that is not even worth shredding anymore since i did talk about the same stuff with them A YEAR AGO....

A Certain Brave Patriot with the patience of Job has something like 15 points they left unanswered. If you wanna go play with them over there i'll send you the address but I'd rather prepare to shred Hart inter civic voting machines at the hearing in the morning..It does more good for fair, free elections that talking about certain silly arguments that IMHO don't even merit argument..

But whatever is fun for you...:shrug: :hi:

[BeFree]

74. Yeah,

I missed the discussions a year ago, so I guess I just wanted to get in my thoughts, but really, it isn't fun at all, it actually disgusts me.

What you said pretty well wraps it up.

[Febble]

76. In that case

as we were.

There was me thinking you were sincere a couple of posts ago. Stupid me.

Bye.

[Febble]

75. Well, if you don't want to talk to us

why the heck bother?

[Febble]

40. I did show you the data

I showed it to you here:

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=437075&mesg_id=437615

That's the data. Every cross on that plot is a precinct. If you read off the vertical axis it tells you how many percentage points Bush did better (or worse) in 2004 compared to 2000, and if you read off the horizontal axis it tells you how many standard deviations off the exit polls were.

And where do you "see data from places that in 2000 bushco lost paper ballots, but won overwhelmingly in 2004 on DREs"? Can you give me a link? If "them's the facts"? Where can I see those facts? I'm happy to look at them if you can show me that data.

[BeFree]

43. That picture is your data?

You know better than that. I am surprised at you. Is that really a good example of the "Intellectual Integrity" and honesty yall pride yourselves on?

That picture is your data, the whole kit and kaboodle? Now wonder you couldn't find anything. All joking aside, you know that is not what we want to see. And you can't show it to us. Are ya gonna make me find the link quoting you on that? Puuleezze.

And had you really looked at things all over the place, say, even at MCM's "Fooled Again" you saw the same address I saw where I found the redshifting machines.

[Febble]

46. That picture is data

real raw data, from real precincts. It shows an extremely important relationship between three important variables: Bush's vote count in 2000; his vote count in 2004; and his share of the 2004 exit poll. Think about what it means.

What you asked for was the data from the analysis on which I based my opinion that it is very improbable that vote theft occurred on a multi-million vote scale. I've actually shown it to you. Sure, there are lots of other variables in the data set - if you read the Edison-Mitofsky evaluation you can find out what a lot of them are.

But there really is no point in continuing with this BeFree. I have explained the analysis, I have shown you the plot, and I have invited you to comment. All you can do is make snarky remarks about my integrity. Well, I've had enough.

Bye.

[BeFree]

47. Bye, bye

For everyone else... there is a set of data that the exit-pollers have said they can't release to the public. Febble, not being public, did see that data, the whole set. Others have asked to see the same data but have been refused. Those others have been so noted up thread.

Thanks for reading, hope it was not too boring?

[althecat]

22. Nitpicking and not very good nitpicking...

OTOH... you seem determined to undermine exit poll analysis so intently that one cannot but wonder whether you are doing so with clean hands.

1. The 1992 poll was nowhere near as large and as sophisticated as 2004s effort.

2. Sophistry.

3. Sophistry.

4. Do you have evidence of this.... are you saying that before their were any official numbers the exit pollsters already had doubts about their numbers. Because why? And if this was the case then why was the data released raw on all the networks without a warning to this effect. And since it was isn't it more than probable that this is yet another post-facto BS explanation from Mitofsky.

5. Actually thanks to Jonathan Simon and someone else (never disclosed who) the data was captured before it was scrubbed. And not only was it scrubbed but Mitofsky and the networks refused to release the very data that they had the previous evening posted on their own websites. IS THIS NOT A CLEAR SIGN OF BAD FAITH - NOT ONLY FROM MITOFSKY BUT FROM THE NEP AS WELL.

6. The new numbers contain numerous anomalies which the massaging could not remove. For example.... they say that a clear majority of new voters voted for Kerry. For that to be true and for Bush to have won somehow a large number of democratic voters must have switched to Bush.... which seems exceedingly unlikely. This is just one of numerous problems with the massaged data.

7. They actually said nothing for days weeks infact. And then they eventually came up with this BS explanation which has no evidence to support it apart from an assumption that the official count must be correct.

8. THIS ONE REALLLY REALLLLLY REALLLY ANNOYS ME. THE EXIT POLLSTERS AND NEP REFUSED ACCESS TO THE DATA - ALL ACCESS TILL WELL AFTER THE DUST HAD SETTLED. AND EVEN NOW THEY HAVE NOT RELEASED DATA WHICH ENABLES A PROPER ANALYSIS OF WEP TO BE CONDUCTED.... I.E. THEY HAVE NOT ENABLED INDENTIFICATION OF PRECINCTS. THEY HAVE REFUSED TO DO SO ON THE BASIS OF A SPECIOUS ARGUMENT THAT THIS COULD INFRINGE PRIVACY. THE ANALYSIS THAT WAS ALLOWED IN OHIO - EVENTUALLY & IN OH ALONE 0 INVOLVED A GREAT DEAL OF UNNECESSARY OBFUSCATION WHICH SHEDS DOUBT ON THE RESULTS AND IS NON-TRANSPARENT. THEREFORE WHAT BEFREE SAYS IS PRECISELY CORRECT THEY HAVE "DENIED AN EXAMINATION OF THEIR PRODUCT". MOREOVER WITH ALMOST CERTAINTY THEY WILL NEVER AGAIN ALLOW PUBLIC ACCESS TO THEIR DATA IN THE IMMEDIATE AFTERMATH OF AN ELECTION.

IF THESE POLSTERS ARE SCIENTISTS - AS THEY CLAIM - AND NOT PARTISAN HACKS WHO ARE IN FACT PARTICIPATING DIRECTLY IN THE FRAUD THEN ONE WOULD THINK THEY WOULD DO THEIR BEST TO DEFEND THEIR SCIENCE RATHER THAN OBFUSC, FRUSTRATE AND GENERALLY GET IN THE WAY OF REASONED ANALYSIS OF THIS PHENOMENA.

9. OTOH you have never questioned your assumption that you know what happened in 2004. So how can you criticise Befree. There is copious evidence of motive, opportunity and execution of a massive election fraud in 2004 and yet you resolutely refuse to examine or discuss any of the facts and instead choose to snipe unceasingly from the sidelines. If you have something to contribute apart from snide remarks please do so.

[OnTheOtherHand]

23. oh, golly

Damn straight, you can help but wonder whether I am doing so with clean hands. You just didn't help it. Your choice. Noted and logged. But whenever you are willing to return to civility, I will be there with you. Your choice.

1. You might want to think really, really hard about this one. Do you really want to argue that U.S. exit polls shouldn't have been expected to be accurate prior to, oh, 1994? 1996? you tell me....

2, No, it isn't. The estimates are updated continually from the times the polls closed. BeFree appears to be hung up on the timing of the CNN.com tabulation updates, which really is immaterial as far as I'm concerned.

3. Ditto. Same issue.

4. Why? According to Slate, the early numbers put Kerry up 10 in New Hampshire. Wouldn't you be suspicious?

The evaluation report at p. 17 states that they warned sponsors about nine states at 4:30 pm, and apologized for not warning subscribers at the same time. Given what we know about the early numbers, I find it totally plausible.

"why was the data released raw on all the networks without a warning to this effect"? I have no idea what you are saying. (Well, for one thing, I am getting the idea that you don't really have a clear idea of what you mean by "data.") Again, think carefully about this one. Do you want to argue that all the networks ran the exit poll results, and then CNN.com sneakily tried to cover them up?

5. The evaluation report contains two sets of Call 3 estimates. What's your gripe?

6. I don't think we can say that "(D)emocratic voters" _must_ have switched to Bush, but some Gore voters must have. Whether you and I like it or not, there is nothing inherently "exceedingly unlikely" about that. The Pew poll is one among many that showed Kerry well ahead among first-time voters but Bush nonetheless ahead overall -- which of course entails (not logically, but practically) that Bush picked up some support among Gore voters.

7. Mitofsky was interviewed on News Hour on November 5.

"no evidence to support it apart from an assumption..." -- I swear, that is exactly what creationists say about evolution, and I am so incredibly sick of hearing it from my side. Read my paper. Read any three of Febble's posts. Get over it, already. You can be unconvinced, but don't just gabble talking points. It's undignified.

8. The privacy argument isn't specious. Moreover, the explanations of what, specifically, analysts would like to do with the precinct identifications have been staggeringly weak. There were over 11,000 precincts in Ohio; there were under 50 exit poll precincts. Even if you assume that the Ohio exit poll was perfect, why would those precincts hold some distinctive key to fraud?

I don't think it's too big a stretch to say, based on your statement here, that every disclosure just makes you more suspicious. In fact, I would suggest that if you are honest with yourself, if NEP released a complete dataset (whatever that means) tomorrow, and it seemed to refute major fraud theories, you would argue -- rightly -- that you can't be sure they didn't fake the data from the ground up sometime after the election. So, I'm not at all convinced that we are really any longer in the realm of evidence.

If you are trying to impress me with your "reasoned analysis," then you had better turn off the caps lock. Meanwhile, in what way do you think that the pollsters have failed to "DEFEND THEIR SCIENCE"?

9. You may believe that, but you are utterly wrong, and your unsubstantiated assumption in this regard is at the root of much confusion. Again, it strikes me as quasi-creationist confection, this strident insistence that because I disagree with you, I must simply be committed to the assumption that you are wrong. It may please you to believe that I always know the answers to my questions before I ask them, but it isn't true. My challenge to you is to stop making up stories about me, and look at the evidence. It's the same challenge I've posed to BeFree, but he has ducked it, and that is why I criticize him.

"you resolutely refuse to examine or discuss any of the facts" -- hello?? what on earth do you think I've been doing for the last year and a half?? what do you think is in my papers? what do you think is in my earlier posts to you? Wow.

[autorank]

32. Remarkable post althecat. And a heads up on ERD history, important.

althecat is an historic figure in our movement. He was there, with "Scoop," Jonathan Simon, and TruthIsAll in the early, early hours and days after the theft.

Here are the two historical articles. Book mark them, they're priceless:

Complete Exit Poll Confirms Net's Suspicions 11.17.04 (althecat)
http://www.scoop.co.nz/stories/HL0411/S00227.htm

and

47 State Exit Poll Analysis Confirms Swing Anomaly (althecat & Jonathan Simon)
http://www.scoop.co.nz/stories/HL0411/S00142.htm (with TIA Red Shift original graph)

This post is excellent on it's own. I particularly like your point about "motive, opportunity and execution of a massive election fraud in 2004."

It starts with "motive." Who are we dealing with, the Little Sisters of the Poor (bless their hearts), a crew of wandering monks learning humility OR

THE VERY WORST COLLECTION OF LEADERS THE WORLD HAS EVER KNOW...people who have brought us to the verge of extinction as a species due to climate change...ignored, enabled, and executed with complete indifference to science and the outcome to humans. This is not to mention the vile foreign adventure they have committed this country to called Iraq, the human rights abuses, the joyful advocacy of death by multiple means (execution: of individuals through court sentencing; through neglect of the environment; through an unnecessary, immoral war).

For the intellectually honest, analysis of the copious evidence of fraud is always informed by these facts.

IF YOU'RE WILLING TO RISK THE ENTIRE SPECIES TO MAKE YOUR ENERGY PALS HAPPY,
IF YOU'RE ABLE TO DENY THE CONSEQUENCES OF NEGLECT TO ALL OF HUMANITY,

HOW HARD IS IT TO JUSTIFY STEALING A FEW ELECTIONS? Not very...

Election fraud is the key element in propelling the messengers of death.

Ignoring this axiom is the height of intellectual decadence, a casual dilettantism that justifies the perpetuation of tyrants.

[Febble]

33. This is getting SO silly....

We have a thread, again, about the exit polls, and whether they are evidence of fraud.

OTOH, and myself (Elizabeth Liddle) have spent a fair amount of time, as have others, looking into this question, and have concluded that they probably don't, and indeed I have come to the converse conclusion, that they are actually counter-evidence of million-vote scale fraud, although they are not nearly sensitive enough instruments to detect fraud on a smaller scale. Nonetheless, there is some evidence in the exit poll data that the exit poll discrepancies, at precinct level, were higher in urban areas where older technology was used, which may suggest that some kind of miscounting associated with older technology differentially benefitted Bush.

Now, I may be wrong. OTOH may be wrong. When it comes to these kind of data, all anyone can offer is probabilities, not certainties. And there may be ways of interrogating the data that I have overlooked. I have certainly spent the best part of the last couple of years figuring out new ways of interrogating the data that might shed light on the issue. So I quite accept that there are arguments and counter arguments that could be made.

But I do not see them being made on this thread. I see attacks on OTOH's integrity. On other threads I see attacks on mine. They are completely irrelevant (apart from being offensive). If an argument is worth making it is worth making. I am interested in arguments. I am interested in counter-arguments. But I am not interested in posts that seem to conclude from any post that questions the proposition that the exit polls evince fraud that the poster

• thinks the election was fair and uncorrupt
  
• thinks that DREs or other digital voting systems are reliable or secure
  
• thinks that Bush deserved to win
  
• is a Republican plant
  
• wants to "undermine" the exit polls
  
• is not interested in evidence

OTOH made a substantive series of responses to BeFree's OP. So, leaving aside, for a moment, your unsupported inferences about OTOH's integrity, let's look at your counter-points.

Your response to the point that the exit polls were almost as far off in 1992 is to argue that the poll in 2004 was larger and more sophisticated. I don't know what your evidence is for this. What may have been more sophisticated were the models used to make the projections from the various data sources (which, as usual, included the vote-returns) but I am not aware of any changes to the data collection protocols, nor am I aware that the sample sizes were any larger. Indeed, in 2004, the mean precinct sample size was only 80, for a target sample size of 100, so the total sample may even have been smaller. Non-response was a big problem in 2004. 53% is a low response rate.

Your response to OTOH's second and third points is the word "sophistry". I'm meeting this word increasingly. Unless you can say what it means, there can be no response.

Your next point is:

Do you have evidence of this.... are you saying that before their were any official numbers the exit pollsters already had doubts about their numbers. Because why? And if this was the case then why was the data released raw on all the networks without a warning to this effect. And since it was isn't it more than probable that this is yet another post-facto BS explanation from Mitofsky.

Well, Mitofsky is on record as saying he knew the raw numbers were off before polling closed, but Mystery Pollster points to direct evidence that this was not "de facto BS" Mitofsky from Wonkette, who leaked the information at 5.40pm on election day itself:

“*** There appear be problems with exits in the following states that could be tipping numbers toward kerry: MN, NH, VT, PA, VA, CT, DE. described only as ‘serious’ issues we’re looking at. so i would not put too much faith in those results.”

http://www.wonkette.com/politics/campaigning/hot-fresh-polling-024809.php

As for your second point - none of the data was "released raw" - it was all weighted, precisely because that's what pollsters do when they have reason to believe their sample is biased, and is why no projections were made on the basis of the exit poll samples alone. The idea that the numbers CNN put up were "raw" is false, although it is true that those numbers were (mostly) not weighted by the vote-returns as at that point, not many vote-returns had entered the data-stream. But they were weighted by other stuff, because already the pollsters knew that their sample had some "'serious' issues" indicating a pro-Kerry bias.

Then you deduce "bad faith" from the non-release of the data, and claim that it was "scrubbed". Well, I don't call posting "data" (actually cross-tabulations from partially weighted data) for hours on the internet "non-release" - I saw them myself. Sure, they were replaced later by fully weighted cross-tabs, because the assumption would be that the fully weighted crosstabs would be more accurate than the partially weighted cross-tabs. That assumption has been legitimately brought into question, but there is no a priori reason to suppose that the interim cross-tabs were "scrubbed" in "bad faith", and the fact that in their January evaluation (pages 21 and 22) E-M actually published their "Call 3 estimates" - in other words, the "raw" estimates made at close-of-poll, weighted only by geographic stratum and not by factors designed to compensate for sampling bias - looks to me like an act of good faith, not bad. Not only did they publish those Call 3 estimates, they also published additional tables showing how representative their precinct samples were of each state (good), and pinpointing the source of the discrepancy to within-precinct level.

There are certainly some anomalies in the reweighted cross-tabs that are worth examining, although the extent of the reweighting required (because the discrepancy was fairly large) means that both weighted and unweighted cross-tabs need to be treated with some caution. Once you think you have a biased sample you have to make a guesstimate as to where to apply the weights. If the weights are applied to the wrong characteristics, you won't necessarily get a better answer. But bear in mind that they knew before the first results were in that the sample had problems.

However, even that being the case, the anomalies are less anomalous than they appear once examined closely, and mostly concern the past reported vote question. OTOH researched answers to this question from past exit polls, and found a clear tendency for voters to over-report having voted for the incumbent at the previous election, including voters who claimed, in 2000, to have voted for Clinton in 1996. Strange it may be, but it seems to happen. In which case, the responses to the past vote question in 2004 are no longer anomalous (unless you want to postulate massive pro-Gore fraud in 2000).

"They actually said nothing for weeks infact" isn't a fact. Here is a transcript from a pbs interview with Mitofsky on November 5th 2004:

http://www.pbs.org/newshour/bb/politics/july-dec04/exitpolls_11-05.html

Data release: privacy is not a "specious argument". It is absolutely essential. The data were collected under promise of confidentiality. A great number of personal details were collected and are in the public domain. Release of data that would allow precincts to be identified could quite easily result in respondents being identified, and would run counter to the ethical guidelines of the pollsters own professional organisation. People keep asserting that confidentiality is not a problem. It is a huge problem. It would be less of a problem if those questionnaires were not already archived, but they were, and they always are. So turn off the caps lock and consider what you are demanding: access to data that would compromise the confidentiality of detailed personal information given by volunteer participants under promise of non-disclosure.

And as for your last point: it is completely unsupported. OTOH has worked harder than almost anyone I know to find out what the exit polls can tell us about what happened in 2004, and the assertion that "he never questioned his assumption" that he knew what happened is ludicrous. My email inbox is stuffed with questions from OTOH regarding just about any assumption either of us has ever made. And your accusation that he "resolutely refuse{s} to examine or discuss any of the facts and instead choose{s} to snipe unceasingly from the sidelines" is similarly ludicrous. His own homepage is stuffed with papers that examine and discuss facts in copious detail, his posts are full of discussions and facts, though typically responded to by sniping posts that avoid engaging either.

Neither OTOH nor I have any fear of facts. I wish the same could be said of others who, like both of us, believe that Election Reform is an urgent and serious requirement.

edited for clarity

[BeFree]

52. For a much better discussion see this thread

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=371726&mesg_id=371726

Time for Change started a thread over a year ago on this controversy. Check out his arguments. Glad to say, it does reinforce what I have written here.