Time Line of Exit Poll Irregularities Election 2004: Dopp and Freeman

[Stevepol]

With a normally alert media, this pdf file from Election Archive would be front page news.

http://electionarchive.org/ucvAnalysis/US/Presidential-Election-2004.pdf

Here's Kathy Dopp's email:

NEDA has prepared a historical summary of important facts and events
related to the ongoing debate over the validity of the 2004 presidential
election results and presented recommendations to ensure vote count
accuracy.

Beginning with the discovery that 2004 presidential exit poll numbers
were altered late during election eve and forced to match the official
vote counts, through subsequent proposed explanations for the
unprecedented, statistically implausible disparity between vote counts
and exit polls, to the latest analyses of Ohio's precinct-level exit
poll results which purports to demonstrate vote fraud, the report
presents brief, simplified summaries of relevant milestones.

Ensuring fair and accurate vote counting is an essential element of
American democracy and is fundamental to each American’s right to vote.
This historical overview is provided to aid media and the public in
understanding what might be regarded as the most important debate of our
era.

NEDA's purpose is to facilitate research of and the formulation of
policies dealing with procedures that will ensure accurate vote counts;
and making detailed election data and information publicly available.
NEDA places particular emphasis on and will devote all of its initial
resources to creating a “National Election Data Archive” project with
the goal of ensuring the accuracy of U.S. elections.

The full paper "History of the Debate surrounding the 2004 Presidential
Election" is available at
http://electionarchive.org/ucvAnalysis/US/Presidential-Election-2004.pdf

[Febble]

1. Oh Lord.

Well, as I seem to feature as a principal in this, I suppose I had better summon a response.

All I can muster is:

I dispute the characterization of some of my arguments, and and the validity of the math used to "refute" them.

Nonetheless, I remain utterly convinced that the last election was both corrupt and corruptible, that the electoral system needs fixing from top to bottom.

Elizabeth Liddle

[Melissa G]

5. Poor Lizzie (said with friendly wry amusement) They hit you pretty hard

in the conclusion which is where I jumped to...
http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=132&topic_id=2175336&mesg_id=2175352

[FogerRox]

8. LOL--- >wink

Funny --

[bleever]

9. Dear Lizzie,

At least have the consolation of knowing that after all the scrums we've had here at DU, I think most people in this forum take you to be an honest and sincere proponent of your arguments.

You've taken your share of licks and always done your best to respond to well-reasoned posts, and kept your arguments centered on the math in a rigorously professional manner.

Hence you're earned respect among those you've often disagreed with, and I give you credit for that.


And if, from the remove of Scotland, you can't make the jump from "I remain utterly convinced that the last election was both corrupt and corruptible, that the electoral system needs fixing from top to bottom" to "Of course the most corrupt and corruptible administration in a hundred years took advantage of a corruptible system," I can't blame you.

Frankly, sometimes I wish I were in Scotland too.


:toast:

[Febble]

15. Bless you, Bleever

Coming from you, that means a lot to me.

I'd actually love to bleeve what you bleeve. But better to be in Scotland (actually England these days).

[Land Shark]

24. There've been some additions to the paper, some may want to

make sure they have the lastest and greatest version of the paper, and not the original draft.

[Febble]

38. A major problem

with this paper, as with many papers from this source, are changes made after the original is "published". It makes serious debate of its content virtually impossible.

But an even greater problem, in my view, is that this paper, has been "published" in the format of an academic paper, with a number of authors, who, it is implied, have endorsed it. It also bears the names of a list of "reviewers". It thus gives the impression of being a rigorous, peer-reviewed paper.

In a peer-reviewed article, all authors would be asked to sign off on the final version, and an independent editor (not an author) would make the final decision as to whether the published version paper met the reviewers requirements.

In this case, the editor is an author; two of the authors have informed me that they have not read either the draft published yesterday (don't know about today's version) and did not know it was ready for "launch". One of the reviewers emailed me to say he was "stunned" to see his name on the front page.

The paper may be good; it may not be. I disagree with a great deal of it, particularly the parts that reference me.

But it needs to be in the public record that its contents cannot be assumed to represent the views of either all its authors or all its reviewers.

I am rather angry about this.


(edited for accuracy)

[Febble]

42. And I see it's

acquired an author and lost a reviewer since yesterday.

Strange.

[Melissa G]

47. Like a certain sharky DUer as an Author and Freeman no longer listed

as a Reviewer... Congrats Land shark! Thanks for all you do for the cause!

edit clarity

[Melissa G]

2. This looks like great stuff! Was on the another site reading it and

decided to make sure it was on DU front and center! Vote this up folks! Nominated!

[Melissa G]

3. Just cross posted this to General Discussion Politics to get it more

visible to folks on DU!

Edit to say authors are actually
Dopp, Baiman, Simon and Mitteldorf
Freeman is among the reviewers as is Robert Koehler

[Febble]

14. I've just been in touch with

Baiman, Simon and Mitteldorf. Neither Simon nor Mitteldorf have read it in its entirety, and Simon did not seem aware that it had yet been publicly "launched".

Both are going to review the current draft, and, if necessary, seek to make changes.

I am going to go out on a limb here and say that if I have one criticism of USCV (aka NEDA) as an organisation, it is its methodology for producing papers with multiple authors. Unlike DU, where all posts are stamped if edited, and posters are diligent about reporting where they have made changes, NEDA papers are often updated without re-dating (although of late, the re-dating has become more diligent). This means that where there is a debate about something in one of their papers, there is no guarantee that both readers have downloaded the same edition.

However, as I say, this has improved of late. However, with multi-author papers, any change needs to be run by all endorsers, otherwise there is no guarantee that all the original authors endorse the changes. Here we have a document that apparently has not been read in final form by two of its authors, and yet it is publicly linked.

I do appreciate the generous words that have been said about me in this thread, especially as I know that my views are not what people would rather hear. And I do try to be both courteous and honest (though I know I sometimes blow the courteous part). And I have been circumspect about criticising NEDA directly (I am happy to criticise math, facts of logic at any time, but not an organisation or people). But I do think it is simply irresponsible to issue a public paper, authored by respected academics, without ensuring that those academics have signed off on the published version. Mitteldorf and Simon may be quite happy with it, in which case I can only note that I personally take issue with it.

But it should not have been published without their full endorsement of the published version.

I have not heard from Baiman as yet.

[bonito]

4. kick

[kster]

6. Off to the greatest.nt

[bleever]

7. Thanks, nominated. nt

[Peace Patriot]

10. This, to me, is the most critically important information ever to be...

...provided to the American people. Page one of the report:


"Kerry Wins Presidential Election according to the Exit Polls (November 2 nd )

"Buried so deep in the methodology statement of the National Election Pool (NEP) that nobody but the serious sleuth is likely to find it, the protocol of Edison/Mitofsky, the private company that conducted the National Exit Poll, calls for the gradual Election Night replacement of genuine exit poll data with incoming vote counts. As vote counts become available (that is, as the polls close) they are used to "adjust" or "force" (the term of art used by Edison/Mitofsky) the exit poll results to conform with emerging final vote tallies -- basically the exit polls morph from being Exit Polls at 9 pm to being virtual carbon copies of the vote tallies a few hours later. The Exit Poll adjustments are purportedly designed to help the media clients get a leg up on the results (send A team to winner's ballroom, B team to loser's, etc.) and to supply useful demographic information to analysts. For these and other purposes of their paying clients, the exit polls can be most helpful if they are in line with the vote totals.

"On Election Night 2004, Dr. Jonathan Simon 1 who had learned about the Edison/Mitofsky plan to displace the exit poll data, downloaded and printed time-stamped screenshots from CNN showing pre- and post-adjustment exit poll results for 44 states and the national sample (i.e. the popular vote). Because of a computer problem at Edison, these late-night screenshots of normally weighted but unadjusted exit poll data remained posted several hours longer than intended -- and thus presented the most accurate, complete and authentic weighted exit poll tallies for each state and the national sample. Each of these screenshots also included the number of respondents, as well as a demographic breakdown of the poll results.

"As recorded and publicized by Dr. Simon and ultimately acknowledged by Edison/Mitofsky as their "Call 3 Weighted" data, these exit poll results revealed the discrepancies between the exit poll results and vote counts both in key states and in the national popular vote, giving rise to the critical debate over the cause: either inexplicably skewed exit polls or outcome-altering mistabulation of the votes.

"Pollsters Alter Exit Poll Data to be the Same as Election Results (November 3 rd )

"The normal demographic weighting of the raw exit poll data produced the CNN numbers up until about 12:24 a.m. November 3 rd 2004. At this juncture E/M's glitched servers finally kicked in and began replacing these results with the "adjusted" (or "forced" or "super-weighted" or "nondemographically weighted") results that matched the vote tallies. Once the full-sample authentic exit poll results were replaced in each state and for the national sample between midnight and 1 a.m., the intention was never to post or publish the authentic exit poll results again.

"It immediately became clear to Dr. Simon that highly significant and unprecedented adjustments were being made to the exit poll percentages with little or no change in the number of respondents, confirming that the Edison/Mitofsky forcing protocol was happening in real time.

"'Adjusted' exit poll results, because they are forced to congruence with the tabulated vote, will be the same as the official election results no matter what the actual exit poll data has been, and therefore they bear no relation to the exit polls themselves. However these new results, that are simply election results based on progressively larger samples of the tabulated vote, continue to be put forth as exit polls, which of course they no longer are."

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

Me:

Secret, proprietary exit polls, mysteriously weighted with secret formulae to "fit" the official results, and...

Secret, proprietary vote tabulation, conducted by Bushite electronic voting companies...

equals...

tyranny.

----

Throw Diebold and ES&S election theft machines into 'Boston Harbor' NOW!

[Peace Patriot]

11. Why is this so important--the DOCTORING of the exit polls? Because...

...THAT'S what squelched the protests and calls for investigation. The American people were denied major evidence of election fraud.

In others democracy in the world, exit polls are used to verify elections and check for fraud. They are the standard tool for determining the honesty of elections and the correctness of the results. In this country, in 2004, we were testing out a new electronic voting system for the first time nationwide--a system that had already been called into serious question by many experts--for its demonstrable and proven insecurity, unreliability and hackability. The situation CRIED OUT for independent verification of the election results, and that is the very thing that was denied to us by pollsters Edison/Mitofksy and the consortium of war profiteering corporate news monopolies who hired them: honest exit polls.

It is the worst journalistic crime ever committed, bar none--worse even than Judith Miller's and the NYT's propagandizing of the war--because it denied the American people any corrective mechanism, any means of righting this leaking, sinking, bankrupted ship of a country.

We can't even trust them on numbers any more, on basic facts, on basic fairness. The entire monopolistic media establishment has gone over to the Dark Side. They are now fascist operatives, not journalists. They FIDDLED their own numbers to HIDE evidence that BUSH DIDN'T WIN!

And the impact on the public was dramatic and devastating. Most people don't know to this day that there is MAJOR EVIDENCE OF ELECTION FRAUD.

And for that we have to thank not only the useless and corrupt "news" establishment, but also the corruption of Democratic election officials and other Democratic leaders and representatives in the $4 billion electronic voting boondoggle. Republicans are corrupt by their nature these days. That's what the Bush Cartel has done to them. But the Democrats remained SILENT while two Bushite corporations took over our election system with SECRET, PROPRIETARY programming code. They could hardly bring themselves to criticize the massive, visible vote suppression against black, poor and other Democratic voters in Ohio and elsewhere--let alone warn the American public that our votes were going to be tabulated in secret by major Bush donors.

Edison/Mitofsky has now promised that, NEXT TIME, nobody but nobody is going to get their hands on the REAL exit poll results.

So we had better support Kathy Dopp and UScountvotes.org, and others, in devising new ways to monitor the next elections. It CAN be done. The process of reforming this election system is going to take time and is not going to be easy. But it has to start with independent verification--statistical monitoring and analysis, parallel elections, independent exit polling, and/or election challenges. It has to start with information--gathered by the people, and not by any corporation or corporate shill of an election official.

I say again, it CAN be done. And NOW is not too soon to start.

[OnTheOtherHand]

17. unfortunately, the report is gibberish

(EDIT: or rather this portion of it is; I don't have time right now to read the whole thing)

Yes, E/M said in advance that they would adjust their exit polls to official returns as returns became available. That's how most exit polls work. To say that the adjusted results are then "the same as the official election results" overlooks the salient fact that the vast majority of the data are not "who did you just vote for?"

U.S. exit polls have been weighted to official returns at least since 1976 (the earliest dataset I have access to). To assert that results from these weightings "continue to be put forth as exit polls, which of course they no longer are," shows either utter ignorance about the field, or an insistence upon unilaterally redefining terms for rhetorical effect.

If -- I don't know, but if -- USCV intends to be taken seriously by survey professionals, it absolutely must lay off the purple prose. If USCV intends systematically to alienate survey professionals and see if it can win the argument anyway, well, good luck with that.

[Febble]

18. I don't know what kinds of sleuth

you guys have in America, but far from being:

"Buried so deep in the methodology statement of the National Election Pool (NEP) that nobody but the serious sleuth is likely to find it"

the information is on the FAQ, linked to from the E-M front page, and hyperlinked under the heading: "how are the projections made":

How are projections made?

Projections are based on models that use votes from three (3) different sources -- exit poll interviews with voters, vote returns as reported by election officials from the sample precincts, and tabulations of votes by county. The models make estimates from all these vote reports. The models also indicate the likely error in the estimates. The best model estimate may be used to make a projection if it passes a series of tests.

(my bold)

[Chi]

12. Thanx for posting Stevepol!

Nominated for exposure.
:popcorn:

[autorank]

13. Kick, Recommend

Great post, great information.
:kick:

[DemReadingDU]

16. Recommended

[Febble]

19. OK, here we go:

I'll take it in bits. This bit is boring, but it's wrong (in my view) so I'll put it on record:

Flaws in Liddle's Analysis included:

a) By looking at a simulation of an exaggerated ratio of Kerry-to-Bush voters' exit poll response (Liddle assumed a 2:1, K/B ratio giving an "alpha" of 2) 64 Liddle produced an “inverted U” shaped WPE/WPD graph that seemed asymmetric enough, so that it appeared that it could
approximate the E/M reported WPD outcomes. The asymmetry of the “inverted U” WPD curve – which gives a slightly larger WPD in high Republican precincts seemed to be consistent with the (much higher) WPD of high Bush precincts in the E/M data. The asymmetry is a mathematical result of linking an absolute difference (WPD) measure to a ratio measure (alpha=K/B). This “mathematical nit” cannot possibly explain the dramatic asymmetry in the E/M data. If an absolute difference “differential partisan response” measure is used (Bush - Kerry voter response rates), this small asymmetry disappears altogether. Only with highly magnified levels of alpha (such as a 2: 1 ratio of K:B representing alpha=2) will this small effect look significant.

The authors claim that a "mathematical nit" (whatever that means) cannot explain the "dramatic assymmetry in the E/M data". The aggregate mean WPEs for five arbitrarily drawn "partisanship" categories do indeed look "dramatic". They certainly alerted me when I saw them in the E-M report. However, when the full set of data points is seen in the scatterplot displayed by Mitofsky at AAPOR, a) the assymmetry is seen to be very much less dramatic, and b) also disappears when my proposed measure is substituted for WPE. My measure may not be the best measure of Precinct Level Discrepancy - indeed I think it is not, although I think it is very much better than WPE, which has two major problems. I am not sure what measure is being proposed in the USCV document, but "Bush minus Kerry response rates" begs a huge question - you can only compute Bush and Kerry response rates if you know the numbers of responses for each candidate in each precinct and the number of votes counted for each candidate in each precinct. Clearly USCV do not know this - it is why they advocate for access to the data. Moreover I would argue (though not here) that it is not a better measure than my measure, even if computed using the actual data.

The measure I proposed (and would still defend) was not the difference between the two response rates, but the ratio. There may be an argument for using a difference measure, and indeed Mark Lindeman and myself have spent an inordinate amount of time attempting to derive a measure that is unconfounded by vote-share. It has been an extraordinarily difficult (but fascinating, to geeks) task. However, I think Mark will support me when I say that our conclusion is that a ratio measure is probably the way to go. But I digress. The point is that neither can be produced without access to the actual data, and what Mitofsky did, using the actual data, was to compute a ratio measure. (And if by a difference between "Bush and Kerry response rates", USCV mean the difference between "K" and "B" values computed using their formula, a function of completion rates, then I would maintain that their difference measure is simply wrong.)

b) Liddle’s analysis at this point was based entirely on hypothetical simulations. In response, NEDA did some simulations – trying to match E/M WPD and response outcomes with constant alpha. NEDA's simulations showed that matching E/M reported mean and median WPD levels, and over-all response rates with constant alpha was highly improbable to impossible. In other words, NEDA’s simulations showed that even for the K:B ratio that worked best, the rBr hypothesis was not a likely explanation for the discrepancy between official results and exit poll results.

Well, before Mitofsky did his analysis on the actual data, all we had was simulations. I maintained that the aggregate category data was compatible with a flat linear regression line; NEDA maintained (as I understood it) that it was not. In fact, it turned out it was. There may indeed be non-linear relationships between bias and Bush's vote share, but there does not appear to be a linear relationship. Thus the hypothesis that fraud had a tendency to be greater in precincts in which Bush's share of the vote was greater, is not supported. This does nothing, however, to disprove fraud - it just says there wasn't a generalizable tendency for it to be more prevalent in precincts with high Bush counts. It could have been concentrated in Kerry strongholds.

c) Defenders of Liddle's analysis claimed that the only unusual thing about the scatter plot data are four high Bush outliers in the precincts that voted over 80% for Bush, that are not offset by any high Kerry outliers, and that these Bush outliers should be dropped. However, four outliers represents 10% of a sample of 40. If 10% of all of the high Bush precincts in the country were corrupted, this could represent a very serious problem. Moreover, even if the four outliers are removed, the other ways in which the E/M data are not consistent with constant mean bias hypothesis need to be addressed. The NEDA reports have shown, for example, that the E/M hypothesis is not consistent with the high median WPE in precincts with 80% or greater Bush vote, and with the very small mean and median WPE in precincts with an 80% of greater Kerry official vote, that would not be affected by removing a small number of outliers in high Bush vote precincts.

Well, I can't speak for my defenders (and a major problem with this paper is that this, like many similar statements, are not referenced) but I have certainly never suggested that ANY outliers should be dropped. Outliers are important. All I have suggested is that the four highly discrepant precincts in the high Bush category are no more nor less discrepant than many similar precincts in other categories. In other words there is nothing particularly generalizable about high Bush precincts - no reason to look in those rather than in any others for fraud. What I HAVE pointed out, and the paper also points out, is that there are a fair number of similarly discrepant precincts in which the shift is BLUE. And moreover that one of these just misses the arbitrary category boundary between mod and high Bush. In other words, had the Bush's vote count in that precinct been a single percentage point higher, the mean of the "high Bush" category would have been in line with the rest (except for the low Kerry category). Moreover, the mean of zero in the low Kerry category can be seen to arise not because there is are no discrepant precincts in that category, but because there are highly discrepant precincts in both directions. So my point (and Mitofsky's) is a very simple one: there is no generalizable tendency for red-shift to be greater where Bush's share of the vote is greater. That does not mean (sigh) that there is no fraud. It just means there is no special reason to look for fraud in high Bush precincts. I have NO idea why this finding has become so ridiculously controversial.

I'll post more stuff as I find it.

On edit: missed a bit. I'd just point out that a 2:1 ratio is not particularly exaggerated. Mitofsky's plot indicates that many, many precincts were discrepant at that level or greater.

Link to Mitofsky's plots are given in this DKos diary:

http://www.dailykos.com/story/2005/5/24/213011/565

[Name removed]

20. Deleted message

Message removed by moderator. Click here to review the message board rules.

[Febble]

21. I understand, Kathy

that you believe I am lying.

My own belief is that I am not.

[Name removed]

22. Deleted message

Message removed by moderator. Click here to review the message board rules.

[Febble]

23. Kathy, you just

published a paper which contains things with which I disagree. I have just posted, for the record, my response to those things.

I do not claim to read yours, or anyone's mind, only my own.

If you'd like to be specific, I'll do my best to respond.

Edit: on re-reading I realise you probably were referring to my statement that I thought you thought I was lying. I was not attempting to read your mind. One of your statements in your post was: "Don't you ever examine the truth of your own statements?" From that I inferred that you thought I was lying. If you weren't, that's fine. I wasn't.

[kster]

25. Oh come on

unless Mitofsky demands that Freeman go on live PRIME TIME television,Freeman has won the debate. They want to silence to INTERNET team, but it isn't gonna work this time, Mitofsky did the polling for Cable news, tell Mitofsky to Demand that they(cable news) put the debate on PRIME TIME television so all Americans can decide.

[Febble]

29. Well, that is the strangest

way of resolving a debate that I have ever heard. The loser is the one who fails to demand that the debate is aired on prime time?

WTF?

[kster]

43. Both should demand

that it run on prime time, I just thought it would be easier for Mitofsky to get it on Television, being as he WORKS FOR THE NETWORKS!

[Febble]

44. Well, I really hoped it would be on C span

apparently it wasn't even that well attended. Shame.

[kster]

45. MITOFSKY EXIT POLLS

[kster]

26. Mitofsky vs Freeman

PRIME TIME did you ask Mitofsky yet? I would like to see it, Mitofsky vs Feeman prime time,for all to see, Mitofsky isn't worried about his NUMBERS is he? Mitofsky was working for the cable news channel, Right? He shouldn't have any problem defending his numbers for all to see, on prime time TV with our side in attendance, right?

[Name removed]

27. Deleted message

Message removed by moderator. Click here to review the message board rules.

[kster]

28. Question's

What do you think of the exit poll debate? Do you think it should have been televised?

[Name removed]

31. Deleted message

Message removed by moderator. Click here to review the message board rules.

[Febble]

30. Some in accuracies here, Kathy

"Liddle attacks anything that the National Election Data Archive does - her work is never logically correct but unfortunately she is very convincing to those who aren't deeply involved in the work."

Does this include work I did for USCV/NEDA, and is still hosted on its website? Or is that work also logically incorrect.

Kathy, we disagree. About math. People who disagree about math do not have to disagree about the importance of reforming American Democracy. My own view happens to be that the exit poll debate has actually damaged the movement for the simple reason that I think the exit poll evidence have never been good evidence of widespread fraud, and may at this stage actually be evidence against widespread fraud.

I think, nonetheless, that the electoral system is corrupt, and needs fixing. I think we need to find evidence of exactly where it is corrupt. And I think the exit polls are a red herring that may be preventing us from looking in the places where electoral injustice or fraud really lurks. I may be wrong. As I say, I'm not wedded to the math.

But I do think that the failure to find a linear relationship between the magnitude of the exit poll discrepancy and the magnitude of Bush's increase in vote share since 2000 is a major problem for the argument that the discrepancy indexes fraud. I am happy to debate this, and I am doing so in a civil fashion on many other threads. I hope I am helping a number of hard working investigators look for fraud in places, and of a type, that is compatible with what we now know. I am willing to be persuaded I am wrong.

But I am not willing to be portrayed as a Troll (which I believe is against DU rules) even as a "sophisticated, intelligent-sounding" one (should I be flattered?).


(edited for typo).

[Name removed]

32. Deleted message

Message removed by moderator. Click here to review the message board rules.

[Febble]

33. OK, Kathy, this is more like debate

If you would just stop impugning my motives.

No I was not talking about Ohio. I was talking about an equivalent analysis done on the whole dataset, with the same finding, and presented by Mitofsky at the Freeman-Mitofsky debate.

Mark Lindeman has posted an account here:

http://inside.bard.edu/~lindeman/slides.html

I'l respond in more detail later, but I don't actually think your piece on Ohio refutes the main finding, although it may deduce a second finding.

But please let's keep this debate civil. The DU rules require it, apart from anything else.

On edit: I'd appreciate you editing your otherwise substantive post to keep the debate impersonal.

[Name removed]

34. Deleted message

Message removed by moderator. Click here to review the message board rules.

[Febble]

37. In an earlier

(now deleted) post you took me to task for attempting to read your mind.

Please do not attempt to read mine. My motivations are not as you portray them.

And you are repeatedly breaking DU rules, not by disagreeing with me on the issues, which you are of course entitled to do, as I am entitled to disagree with you, but by posting ad hominem attacks against me, other posters, and now against the DU moderators.


(edited for typo)

[sunshinekathy]

48. That entire analysis is based on a faulty premise

Febble,

IMO it is irresponsible (mathematical thuggery) to blithely throw out such analysis as Lindeman/Mitofsky/ESI has done without first taking the time to check it first and make sure it is accurate, both logically and mathematically.

In this particular case, the hypothesis that both the OH and national exit poll analysis of ESI/Mitofsky/Lindeman and I suppose your own analyses are based on is provably illogical.

In fact it is so easy to disprove its validity that any mathematician on the planet, even those completely unfamiliar with the algebra of exit poll analysis will be able to see in a few minutes that it is illogical bunk.

I therefore challenge you to figure it out yourself BEFORE I publicly release my proof on November 2, the anniversary of the November 2, 2004 election.

In the future, PLEASE act more responsibly so that you do not mislead the public or those who are mathematically challenged.

It is easy for anyone with any mathematical or logical talent to come up with counter-examples to disprove the basis upon which all of ESI/Mitofsky/Lindeman's latest analysis is done and you should all be ashamed of yourselves for not using mathematics in a more responsible way, by throwing out such stuff without any due diligence to make sure it is correct first, before you do.

American democracy deserves more due diligence and responsible use of mathematics than has been shown since January 20th Mitofsky report, and there is no excuse for the June ESI report's public release of an illogically incorrect hypothesis and then analysis which is completely based on it. And you are far to bright Febble, to join in with them without doing the due diligence of a logic analysis on the hypothesis.

I am angry, deservedly so, at the misuse of mathematics by those who do not even take the time to logically analyze the basis of their calculations BEFORE publicly doing analysis and making conclusions based on them.

You are a very smart person. Analyse the logic of your basic hypothesis. It is VERY easy for a smart person like you. Certainly you have taken at least one course in math logic. Apply what you know next time BEFORE publicly releasing incorrect mathematical analyses and conclusions.

This latest hypothesis on which ESI/Mitofsky and now Lindeman is basing analysis on is grossly misleading the public and you are smart enough to figure that out for yourself.

If you don't figure it out yourself by November 2nd (which I know you are capable of), you can read my paper then.

[Febble]

49. I will be very interested to read

your paper when it comes out, Kathy.

I am not sure what part of what I have said is what you find illogical, however. But let me say what I am saying, and perhaps you can tell me where the lack of logic lies.

What Mitofsky showed was the correlation between a measure of "swing" (extent to which Bush increased his vote share" and a measure of "redshift". If "swing" and "redshift" are due to the same cause (fraud) they should be positively correlated. In fact the correlation was tiny (and,ironically, slightly negative).

Now this certainly does not indicate that there was no fraud in the election. But it does suggest, fairly strongly, that fraud was not the cause of redshift.

There may be "loopholes" in this argument, and Land Shark has a fun thread where some of us are thrashing this out. I've thought of a few, though none that have got me very excited yet.

Why don't you post your argument on that thread? (Perhaps you have - I'll have a look). Or alternatively, I have attempted to lay out my own logic on the Calculatus Eliminatus thread.

(In fairness, I should point out that Mark Lindeman is only responsible for the development of the measure of redshift used, not the analysis. He has, however, made the plots available on his website.)


On edit: If the logical flaw is so obvious, Kathy, I do think it would be responsible of you to point it out here and now. Shame to risk anyone becoming convinced of a fallacy when a little proof would send it to flight....?;)

[sunshinekathy]

51. The basic premise for your analysis is logically invalid

You said that




The basis for this analysis is logically flawed. I took the statement of the basic hypothetical inference upon which all this analysis is based, and converted it to logical inferences and analyzed it logically. It is logically invalid. Everyone is wasting their time and ours to base any analysis on it.

Do it yourself - to disprove A implies B one must just find a case where you have A and not B.

It is easy to do. I'll be releasing my proof of the logical invalidity of your analysis by November 2nd - writing anything up so people can understand it always takes forever.

[Febble]

53. Not so

You seem not to have allowed for variance.

If A is "swing" and B is "PLD" ("Precinct Level Discrepancy -I like Steve's term), there will be some variance in A that is due natural variance in the success of the campaigns, and some variance in B that is due to natural variance in polling accuracy.

Variance in A due to success in the campaigns will, let's say, have a normal distribution (most things do).

Variance in B due to polling error will also, let's say, have a normal distribution.

But we are postulating that in addition to variance in the success of the campaigns, some of the variance in A is due to variable degrees of fraud. In precincts with a lot of fraud, Bush's swing will be larger than it would have been without fraud (that's what the fraud is for). In precincts with no fraud, the swing will be whatever the real swing was in that precinct.

Moreover, where there is a lot of fraud, there will be extra red-shifted PLD in addition to whatever polling error there was in that precinct. Where there is none, there will be no extra redshift.

Therefore: variance in A that is due to campaign success alone will be independent of variance in B that is due to polling error alone. However, variance in A that is due to fraud will be shared with variance in B that is due to fraud.

So we partition the variance and determine what percentage of variance in A is shared with variance in B. If it is a lot we can say probably deduce fraud. If there is very little, then we cannot say much about fraud at all - in fact we may even be able to say that fraud was unlikely to be the cause of error in the exit polls.

And it turned out that the percentage of variance shared between A and B was .009%. Not a lot. And worse - the correlation coefficient was actually negative.

Back to your logic:

There will certainly be precincts in which you get a lot of swing and not much redshift. Also precincts with not much swing and a lot of redshift. And precincts with a lot of both, and precincts without much of either. That's what happens when variables have normally distributed values.

And it's why we partition variance to figure out the extent to which variance between to variables is shared. It tells us whether there is a common factor affecting both. The evidence suggests that in the case of the PLD and swing, there wasn't.

[Febble]

50. A post-script

as you've been a bit out of the DU debate lately. Had a conversation with Land Shark on my Calculatus Eliminatus thread, and inadvertently found myself with an opportunity to state my personal position on the exit poll issue, so you might like to take a look:

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=398267&mesg_id=398406

The thing is, Kathy, that although what I've been up to might seem like "mathematical thuggery" to you, I'd argue that there is a role in every campaign for someone to find out the weaknesses in your own side's argument. Far from being a "thug", my own sense is that I've been wielding a scalpel - trying to separate out the weak arguments from the strong. Now we may disagree as to what constitutes a strong argument (clearly we do) but these are somewhat subtle differences in the grand scheme of things. We both know that the election was unjustly won - it was why I contacted you nearly a year ago, when you posted your Florida data, and I was poking around the web trying to find out whether Bush had cheated in Florida for a second time.

Now, if that exit poll proof you are keeping under wraps is so darned obvious, how about coughing it up now, eh? Because have to say that the proposition "if C causes both A and B, values of A will have a relationship with values of B" seems pretty logical from where I'm standing.

But of course there may be something obvious I'm missing....

[sunshinekathy]

52. Definition of "mathematical thuggery"

My definition of mathematical thuggery is bandying about mathematics to prove ones' point without doing the necessary logical and/or mathematical checks of the correctness of one's mathematical statements BEFORE misleading the public with incorrect analyses and conclusions.

That is exactly what you and your group have been doing. American democracy deserves it mathematicians to behave more responsibly than that.

[Febble]

54. I plead not guilty.

See above.

[Name removed]

35. Deleted message

Message removed by moderator. Click here to review the message board rules.

[yowzayowzayowza]

36. No, Kathy. I saw that post...

... and it looked ripe for moderator deletion due to personal attacks, imho. No :tinfoilhat: required.

[Name removed]

39. Deleted message

Message removed by moderator. Click here to review the message board rules.

[yowzayowzayowza]

40. Naw, Kathy, I didn't memorize it.

Most of it looked ok... jus a sentence er two that were definitely over the top. The post I'm responding to now, however, is another matter entirely, as I'm quite sure you know. (I saved it locally this time if ya need sum more re-minden.)

[Febble]

41. Well

before this post is deleted as well, here are some specifics:

You accuse a DU poster of being "offensively dishonest and manipulative". Accusing a fellow DU poster of dishonest is against the DU rules, and has been for as long as I've been posting on DU:

"Do not call another member of this message board a liar, and do not call another member's post a lie. You are, of course, permitted to point out when a post is untrue or factually incorrect."

You have also called me a Troll:

"Do not publicly accuse another member of this message board of being a disruptor, conservative, Republican, FReeper, or troll, or do not otherwise imply they are not welcome on Democratic Underground. If you think someone is a disruptor, click the "Alert" link below their post to let the moderators know."

You have posted personal information about me:

"Do not post personal information about any other person, even if that information is publicly available."

You have also suggested, in many of your recent posts, that my underlying motivation is reprehensible. You are entitled to this view, but not, in my understanding of the rules, to post that view on DU:

"Do not post personal attacks or engage in name-calling against other individual members of this discussion board. Even very mild personal attacks are forbidden."

I consider calling someone delusional, and asserting that that person's motivation for posting is to gain "attention here on DU", or "attention in the U.S. media", or "to show everyone how smart you are" is, at the minimum, a "mild personal attack".

I am more than happy to debate logic. I do not accept that my arguments are illogical. You disagree. That is fine. But I will not debate logic on DU if it is laced with personal attacks, which are against the rules. There is no point in debating with you anyway if all your posts are deleted.

http://www.democraticunderground.com/forums/rules_detailed.html

[autorank]

46. Thaks for posting Stevepol! Shakin' things up are you;)