The Red Shift Is A Result of Bias, not Fraud

[mgr]

I realize that many of you may already suspect this, but for those of you that don't, I offer the following argument. I presume that any readers of this thread are not unfamiliar with the issues presented, otherwise this post would take on the dimensions of a book chapter.

A critical assumption made by various proponents that the popular vote was stolen is that the NEP exit poll cannot distinguish between sample bias favoring democrats and a swing within the electorate that favors democrats without resorting to the current election results (a logical inconsistency). I offer an argument below that an internal consistency check is possible without resorting to this circularity. With this the issue of fraud being detected in this year's exit polls falters.

Allow me to depict the proponents’ positions into two camps—the strong fraud and the weak fraud camps. The strong fraud camp would have it that fraud has occurred in all elections that exit polling has been applied, and all of the national outcomes are suspect. The weak fraud camp argues that only the fraud in this election was of such magnitude given the closeness of the race to swing a democrat majority into a republican majority, but that fraud in past elections was did not significantly affect results.

The strong fraud argument would have it that the Dukakis/Bush contest was won by Dukakis as shown in the raw exit poll results, but went to Bush due to election fraud. On the face of it, this argument may appear risible, but its plausibility was addressed at this thread (Mistwell's What Are the Exit Pollers Afraid Of). In this thread, one of the strongest proponents of the stolen popular vote position, TIA, allowed that he did not take this position. I should point out that plausible and possible should be understood as different criteria for evaluating a possible event (both are speculation)—possible implies that one may conceive of something; plausible requires that something is possible, and means are available to accomplish it.

I should note that the critical difference in opinion between TIA and myself regarding the use of the NEP polling is that I do not think that the sampling sufficiently precise to detect voting irregularities. I would like to see sampling at a larger magnitude -- 1% of the total electorate (we disagree on other matters as well). I am agnostic on how NEP’s exit poll plays in detecting election fraud in particular states, but would like closer scrutiny of those where vote suppression measures and voting irregularities were reported frequently (Ohio, Pennsylvania, Florida, Texas, New Mexico, Arizona). In my gut, I feel that we won the electoral college.

However, a characteristic of the NEP polling has not been examined as to how it plays into checking internal coherence of the sampling, and that is how turnout is measured. Each precinct has a set sample interval, one out of five, one out of two, etc. If you already know the precinct’s voting population, then the total numbers to be sampled (response and refusals) will characterize the precinct’s turnout on election day.

Exit polling has to rely upon past patterns of turnout to characterize voter intent—there is a pattern of turnout when democrats win, and there is a pattern of turnout when republicans win. Particular precincts when sampled may be bell weathers to address the consistency of the overall sampling, and would be the internal check on a stolen popular vote. In other words, if a bellwether precinct pattern is consistent with a democratic victory, and there is a shift towards democrats in the raw results, but the popular vote went to republicans, then fraud occurred. If the bellwether precinct pattern is inconsistent with a democratic victory, but there is a shift towards democrats in the raw results, bias has entered into the sampling. I would allow that NEP has identified these precincts prior to the election, and reviews them before making any predictions. One may presuppose these bellwethers when sampled lack the significant Within Precinct Error suggestive of some internal inconsistency in the sampling, because it would not make sense to use them if they did not.

The strong fraud position weakens this ‘consistency check’ with the argument that all historic sampling would be tainted by previous election fraud. However, we can rely on the more ‘anecdotal’ experience of election observers and political scientists and would expect (for example) that a lower than expected turnout among African Americans to be indicative of an election we are likely to lose, where a lower than expected turnout among white males would be indicative of an election the Republicans are likely to lose.

Although TIA claims their support, I suspect that Freeman, Dopp, and others are aware of this likely internal consistency check. Their purposes and goals are different from TIA’s, and I do not see this argument undermining their work. The problem from their perspective is that no internal consistency check been made explicit, and their intent is that the raw results be a part of the public domain.

Mike

[chomskysright]

1. great post: I've been in conv w/ Dopp & Lynn Landes..who....

maintains, I believe I can say this, that many of the elections have been flawed by tainted by exit poll 'malfeasance.'

[TruthIsAll]

10. Elections: flawed by stolen votes; covered up by final exit polls

The exit polls, at least the preliminary ones, show that but they are kept hidden from public view. Only this time we got them because they were downloaded by astute patriots on Election Day. The FINAL exit poll is an attempt to legitimize the flawed recorded vote.

Lynn Landes apparently agrees that the final exit poll is bogus, since it matches to the final vote which she knows is corrupted.
But she is apparently unaware that the early, pristine polls, which are hidden from public view by the MSM, appear to be very close to the truth.

[mgr]

20. What you say?

Which preliminary ones? Those released at Scoop with the term 'weighted' on them, those pristine ones, or the saved CNN screen shots with no attribution as to whether they were weighted whatsoever? I guess one's pristine orange juice is another's filtered concentrate.

Why might one keep them from public view on election day? I remember a President by the name of Carter, you remember him don't you, gave a concession speech before the polls closed on the West Coast. Watch what you wish for, and don't talk out of both sides of your mouth.

Which final exit poll? The state poll, the national poll? We know from Zogby (11/6 release on his website) that there is no issue within the polling community with the changing numbers, so do you understand the matter better than Zogby?

Lynn Landes either agrees or she does not. There is nothing apparent about it. Putting words in her mouth, or thoughts in her head? She has a voice, so let her speak for herself. Stop building yourself up as being bigger in the community than you are by implying assent when it is probably not there. The tactic does not work with me.

Mike

[TruthIsAll]

30. Where have you been, mgr, the last few months?

I don't need lecturing from you, Mr. Misinformation.
What have you contributed other than slick obfuscations?

Surely, you must know which preliminary and final polls I'm referring to. If you do, you are just blowing more smoke.
If you don't, you haven't been paying attention.

OK, I'll assume you don't:

Nation Exit poll:
Preliminary I: 7:38 PM posted on CNN (11027 respondents).
Kerry wins by 3 points.

Preliminary I: 12:22am posted on WP (13047 respondents).
No change. Kerry wins by 3 points.

Final: 2:04 Pm posted on CNN (13660 respondents).
Bush wins by 3.

State polls:
Prelim: 12:22am downloaded by Simon.
Kerry is winning FL and OH and virtually all battleground states.

Final: On CNN, aprox 1am
Bush overtakes Kerry - just like in the NEP.
Bush makes a magic comeback, winning 100% of the final (613) respondents, which are make up just 5% of the total.

THOSE PRELIMIARIES. THOSE FINALS.

And, as usual, you are wrong about Lynn Landes. She has always claimed that fraudulent elections occur. And she also claims that the exit polls are fraudulent. And she is right. The final exit polls, which you alone here on DU cannot distinguish from the preliminaries, have been shown to be fraudulent, indeed.

I will give you credit. You are indeed an expert at obfuscation.
You pretend to seek the truth, while at the same time throwing one false argument after another out there.

You have always been one of the more sophisticated, smoothest naysayers, going way back to 2002. But your arguments have been thoroughly debunked by myself and others.

Your attempts at describing my exit poll simulation as kurtotic and platykurtic distributions, was a laugher. And you have not yet responded to my follow up simulation, where I raised the individual precinct MOE to 14%, by cutting the number of precinct respondents in half to 50 and doubling the number of precincts to 260. The final average of 200 simulations was even closer to the base. The result laid to rest the false argument put forward that precinct MOE's would effect the National Average.

So now you or Nederland or Mistwell, I forget who, claims that 13047 from the Preliminary poll is not a large enough sample.

But apparently you believe that the FINAL 13660 is enough.
Just 613 more. That's all that was required to achieve the sample size necessary to make Bush a winner, going from 3% down to 3% up in a flash. Wrap your non-conspiratorial mind around that one.

And you see nothing wrong with that, do you? You never question anything, except to obsure the facts presented in my posts.

Have you ever contributed even ONE, lonely post which questions the impossible anomalies of 2004?

[mgr]

38. None of these points are valid

1. National exit poll is obtained from precincts or individual responses nested within the state polls. The fact that two iterations and a final adjustment are made, and how they are released to the public by the media is beside the point. The methodology for how they are obtained is still unstated, we assume that the mechanism is the same for the state polls, but that may not be the case. Central Limit Theorem, that you are so fond of throwing around, would have the national exit poll MOE greater than the state polls with larger numbers of respondents, but NEP's design does not do this. This suggests that the national exit poll is built straight from the weightings of past voter registration and past voter behavior, but does not relate to trends in the current election. This would explain why two samples do not vary at all, but that the final sample, meant to characterize the current election, is reworked to characterize a new ratio. And that new ratio will be applied to what new precincts will be selected to address the change.

2. Your exit poll simulations were lame. Laugh of the statistics description, but it is part of the science. First off, you mis characterized the MOE in your random sampling (or assumed a CL of 100%), the mean of your sample distribution approximated 50%, which suggests that the mean of the real population could be at either side of 50%. In addition, I also provided an equation for how the MOE can be calculated if you have strong directional bias (one of many I may add) that pretty much follows what most sixth and seventh graders learn in their introduction to probability when differentiating between independent and dependent variables.

3. I personally have never posted to this site until Fall of last year. I am not exactly hidden either, since I have given numerous clues as to where I live, what I do, where I went to school. Anyone with a bit of motivation could find someone's whose first name is Mike with the initials mgr. You may want to compare these to your mystery 2002 poster. I would like to think my preference for long semi-articulate posts would be singular but maybe I am wrong.

4. I have not made any claims as to how Lyn Landes thinks, but you were the one putting words and thoughts into her mouth. My integrity would not allow me to intentionally misrepresent another's position.

5. A sample of 13,000 is inadequate to characterize the intent of 100 million voters, the difference of a single vote represents a difference in outcome of 8,000 votes, so an MOE of 5% (due to bias, because you cannot throw out too many precincts or responses)would represent a swing of 5 million votes. With a million sampled, the difference is 100 votes, (and one may select those precincts with necessary WPE to eliminate the bias) so the MOE could approach 0.1%, or a swing of 100,000 votes, a scale where fraud is easily resolvable.

6. I would question them if I thought them anomalies, but if they were artifacts of poor design, of poor thinking, or of poor sampling, I would comment on them in the manner that I have--GIGO

7. My final take is that I would not accept characterization of any current voting trends from NEP's exit poll, their results and control for bias are that poor. Does that mean I oppose USC's request for the raw data, no. Do I think that you are statistically challenged and ascientific, yes. Do I suspect your motives, well, I am getting there.

Mike

[L. Coyote]

60. You write: "Bush makes a magic comeback, winning 100% of the final (613)"

Does this mean you have the raw data? Or are you misrepresenting what happened to bolster your case?

[Bill Bored]

2. 2 points

1. Has anyone aggregated the total state polling numbers as a consistency check on the 13,660 popular vote sample to see it they match? The state polls may be more accurate; they sampled more precincts; the overall sample size was much larger, etc.

2. The other possible consistency check is party affiliation weightings which HAD to be changed from 38 R/35 D to 37/37 to give Bush any chance of winning the popular vote in the exit poll. Can't we check the true turnout of Repubs and Dems independently without using exit polls?

[FogerRox]

3. No incumbant has won re-elction with May/ June approval rating

below 49%--ever-- Peace time ---war time ---- ever--untill 2004.

Bush was at 44% may/june.

[mgr]

7. Negative campaigning undermines that effect

Here in California we reelected governors with poor job approval ratings-- Wilson, Davis--all they had to do is make sure their opponent's approval rating was lower.

Let's be honest, the Swift Boat Ads were negative, and it did hurt. I mean, I could not believe that my Mother who voted for Gore, did not like Kerry.

Mike

[merh]

18. You want to be honest - then let us see the inside of the computers

and the tabulators used to count the votes!

You want to be honest, then why did those responsible for overseeing election recount efforts in Ohio violate state law regarding the recounts and why did Triad's employees create cheat sheets and screw with the computers prior to the recounts.

You may try to twist and distort the facts, but the truth is, they stole votes in Ohio, Florida and New Mexico.

Excuse and argue away all you like. We know the truth and we are battling to get the election laws changed so this cannot happen again.

Bias has nothing to do with the election theft. More people are for a change of regime than for the weed and his destructive ways.

[mgr]

21. Read my post again

In no way do I disagree with you.

Mike

[FogerRox]

58. Horse swagle. An incumbant gets his may/june aporoval rating

in NOV. for DECADES-----you know thats so-

Its a fact. trying to change the subject does not alter facts.

[mgr]

5. The consistency check can only be at the Precinct level

By NEP's design all of your higher categories (state, national) rely on a lack of bias at the precinct level. To build these higher categories requires no bias to allow for calculating the MOE (SD)

So, for #1, this would probably not make any sense, as you would be with the two data sets, comparing at some level the same precincts, since the national poll is nested in state poll precincts.

#2. You have either your friendly SOS and BOE, or the NEP to choose. I am sure that is not what you were asking, but is there another way to gauge turnout, and the answer is probably no.

The change in registration ratios may either be real, or compensation within NEP for the democratic bias. I have not seen any numbers on republican registration since 2000, but I live in Condit country, so I know what a blue dog looks like (and it isn't from looking in the mirror).

Mike

[Bill Bored]

25. Mgr, if you and NEP are correct, then the margins should be the same

in both the national poll and the aggregate of all the state polls.
Just because they say there's no bias doesn't make it so. If there is no bias, then Kerry should be ahead by the same percentage he was in the national poll if you add up all the votes in all the state polls (at about the same time of day). Perhaps some state population weightings would have to be applied, but other than that, the numbers should come out the same. Kerry should have a little over half the votes of both total samples. If not, then there's a lack of internal consistency, isn't there?

If this has been done, I don't recall seeing it.

TIA, do you have an extrapolation of the popular vote from adding up the state polls?

[mgr]

44. That is an interesting premise.

What would be more operational would be to take the national precincts, remove them from the state precincts, and compare the two. The one big problem is we don't know which precincts to reassign. And, I suspect that the national poll is from individual responses, rather than precincts, this makes it more problematic (if you look at the NEP report, there are two classes of questionaires).

Mike

[papau]

4. is Freeman's book on the election theft still coming out in May?

:-)

[Peace Patriot]

6. I object to your summing up the USCountVotes group with the name...

..."Dopp."

Here is the full group:

Josh Mitteldorf, Ph.D. - Temple University Statistics Department
Steven F. Freeman, PhD - Center for Organizational Dynamics, University of Pennsylvania
Brian Joiner, PhD - Prof. of Statistics and Director of Statistical Consulting (ret), University of Wisconsin
Frank Stenger, PhD in mathematics - School of Computing, University of Utah
Richard G. Sheehan, PhD - Department of Finance, University of Notre Dame
Elizabeth Liddle, MA - (UK) PhD candidate at the University of Nottingham
Paul F. Velleman, Ph.D. - Department of Statistical Sciences, Cornell University
Victoria Lovegren, Ph.D. - Department of Mathematics, Case Western Reserve University
Campbell B. Read, Ph.D., Professor Emeritus, Department of Statistical Science, Southern Methodist University
Kathy Dopp, MS in mathematics - USCountVotes, President
Also Peer Reviewed by USCountVotes’ core group of statisticians and independent reviewers.
http://uscountvotes.org/ucvAnalysis/US/USCountVotes_Re_Mitofsky-Edison.pdf

They give odds of 10 million to one that the exit polls could be skewed the way they were skewed. So, it's not so much that Kerry won the exit polls (by 3%), or that the Edison-Mitofsky excuse for this (that Republicans were shy of the pollsters) is without foundation, or that USCountVotes found E/M data that completely contradicts this excuse (the exit poll was likely skewed to Bush if anyone)--it's that the high peaks for Bush in 7 to 50 states is virtually IMPOSSIBLE.

----------

The 2004 election and its exit polls occurred IN A CONTEXT. You cannot divorce it from that context. That context included:

--the TV networks CHANGING the onscreen exit poll data, alterering it to fit the official results, thus depriving Americans of this evidence of fraud and preventing early and rigorous investigation of the results

--the election was non-transparent, unverifiable, and thus invalid on its face

--Tom Delay and other BushCons deliberately blocked a paper trail requirement and other auditing provisions for electronic voting

--the votes were tabulated by computers using secret, proprietary software, owned and controlled by major partisans of the Bush regime

--Republican Secretaries of State in at least two states, Ohio and Florida, committed massive violations of the Voting Rights Act and other election laws, to suppress minority and other Democratic voters; there were 57,000 complaints to Congress about it; and virtually all machine malfunctions and vote suppression favored Bush and hurt Kerry

--recent opinion polls show continued, widespread opposition to every major Bush domestic and foreign policy, in the 60% to 70% range

--Dr. Freeman adds up Gore repeat voters, the blowout Dem success in new voter registration (Dem 57%, Rep 41%), and the big jump of Nader voters to Kerry, and concludes that Kerry had a margin of 4 to 8 million votes that just somehow disappeared on election day

--this is what the exit polls are showing us--those disappeared Kerry votes

--and the exit polls don't count voters whom Republican electon officials prevented from voting, with every manner of illegal activity and dirty tricks (novel registration rules, lost registrations, purges of black voters, lost absentee ballots, too few precincts, too few voting machines, disinformation on where to vote, etc., etc.); the exit polls only count voters who made it to the polling place.

The exit polls represent one piece of the picture--a rather powerful one, since it is a method that is used worldwide to verify elections and check for fraud. It is particularly useful when the election is non-transparent and with strong evidence of partisan control of the results--as was the case with the 2004 election. But it is not the only evidence of fraud. For instance, what do you make of the numerous reports of touchscreens changing Kerry votes to Bush votes, and almost never the other way around. What are the odds of that happening by chance?

At some point you have to look at INTENT. Was it the intent of the Bush regime to have an honest, transparent, verifiable election? And if the answer is "No"--and I think the evidence is overwhelming for that--then you have to ask, "What is the evidence for a wrong result?"

And here it all is, STARTING with the exit polls.

[mgr]

8. Can a case for election fraud be made without recourse to exit polls?

I think we can, and it would be a stronger argument.

Someone once described what we are addressing as "a great theory undermined by an ugly fact." Take this as peer review, and consider what might happen to your argument among the disinterested.

Mike

[mgr]

22. For an example, see post #7 n/t

[Bill Bored]

26. blowout Dem success in new voter registration (Dem 57%, Rep 41%)?

PP, you say:

"--Dr. Freeman adds up Gore repeat voters, the blowout Dem success in new voter registration (Dem 57%, Rep 41%)"

With all due respect to Dr. Freeman, what is the source for these blowout numbers please? Does he cite references?

And what about all the Dem registrations that ended up in the trash, were lost in the mail in FL, etc? These would of course, not have affected either the vote totals or the exit polls.

[mgr]

36. This requires further investigation.

One of the things we need to do is filter the partisan claims from the balanced. I believe your personal focus is on Florida, where there is a claim by republicans that they were quite successful in obtaining new registrations since the 2000 election.

One of the things that keep being thrown around are percentages without the real numbers behind them. It may appear to be a blowout success if the total of newly registered were over ten million, and greatest in the battleground states; but if disproportionate to blue states, or only at one million, the effect would be insignificant (and there is that nagging question as to whether they went to the polls).

If I may recommend a research venue in FL, it would be to look at relationships between the Jewish vote and precinct level disenfranchisement. I just do not have the time to look into this other than cursory (there appears to be a pattern in Ft. Lauderdale), but Jewish communities are highly autocorrelated to places of worship; and so Jewish predominant precincts would correlate to synagogues--if long waits or missing registrations correlate to a precinct that has several synagogues nearby....

Can you enlighten me on what PP means? Thank you in advance.

Mike

[passy]

43. This is the best explanation of the situation.

I think you should post the second half of your reply as a new thread.
It is very concise and explains the fraud in simple uncomplicated terms as well as drawing attention to the intent of the administration.
Well written.

[Pacifist Patriot]

52. I agree. Fabulous post.

[TruthIsAll]

9. To say "The Red Shift Is A Result of Bias, not Fraud" indicates YOUR bias.

What a statement.

mgr, you have just given us the FINAL word -just like the FINAL exit poll. You really know how to fire off those talking points, don't you.

1. Swift boats
2. Precinct level
3. Negative campaign
4. strong fraud
5. weak fraud
6. insufficient sample-size
7. lower black turnout than expected
8. uscountvotes.org and TIA goals differ (divide and conquer?)
9. no internal consistency check
10. But you say Kerry won the electoral vote?

How did you come up with that:
JFK won the EV but lost the popular?

There was no fraud?
Just bias?

We might as well all just pack up and go home.

Oh, come on now, mgr.
Do you really want to keep that title for your thread?

That statement reveals your bias, not the exit polls.
Seems to me that you want to bury the whole exit poll smoking gun.
It won't wash.

Keep posting.

[kobeisguilty]

12. Seriously.

Look up http://en.wikipedia.org/wiki/Ad_hominem ... it might help.

[TruthIsAll]

13. "The Red Shift Is A Result of Bias, not Fraud" is a declarative statement

To make that statement with 100% certainty is presumptuous, to say the least. It's pure chutzpa. Look up the definition in your wikipedia.

How can anyone, after all that we now know, claim that the exit poll discrepancies were due to innocent sampling bias and not fraud?
Maybe it was a combination, at best.

Apparently, mgr has excluded the possibility of fraud showing up in the exit polls.

Quite a leap of faith.

[mgr]

16. Not Quite the 'SMOKING GUN' is it?

From someone who made a pastime of universal statements regarding stealing of votes from every state, this constitutes critical thought--"that maybe in my excess at least one state exit poll will indicate fraud." C'mon TIA, when does an existential warrant a universal (I know its philosophy, but its symbolic logic--you know, the meta logic underlying programming) Just what do you suppose an agnostic is anyways? Speak to the reasonable man argument, just once.

Mike

[mgr]

17. 2. By the way--it's called design bias. n/t

[Peace Patriot]

11. OK, here's some more...

(response to reply #6)

Dr. Ron Baiman: Economist/Statistician - senior research specialist, Institute of Government and Public Affairs at the University of Illinois at Chicago.

Dr. Baiman: "I conclude that, based on the best exit sample data currently available, neither the national popular vote, or many of the certified state election results, are credible and should not be regarded as a true reflection of the intent of national electorate, or of many state voters, until a complete and thorough investigation…."

"The United States of Ukraine?: Exit Polls Leave Little Doubt that in a Free and Fair Election John Kerry Would Have Won both the Electoral College and the Popular Vote, "
December 19, 2004

http://www.freepress.org/departments/display/19/2004/997

--

Dr. Michael Haut, & UC Berkeley Quantitative Methods Research Team; Haut is a nationally-known expert on statistical methods and member of the National Academy of Sciences and the UC Berkeley Survey Research Center

"UC Berkeley Study Questions Florida E-Vote Count: Research Team Calls for Immediate Investigation"- 11/18/04 (Florida: 130,000 to 230,000 phantom votes for Bush--paper vs. electronic voting)
Report: http://ucdata.berkeley.edu
Press release: http://www.commondreams.org/news2004/1118-14.htm

Dr. Haut and the UC stats team: "…irregularities associated with electronic voting machines may have awarded 130,000 - 260,000 or more excess votes to President George W. Bush in Florida in the 2004 presidential election. The study shows an unexplained discrepancy between votes for President Bush in counties where electronic voting machines were used versus counties using traditional voting methods. Discrepancies this large or larger rarely arise by chance – the probability is less than 0.1 percent."

--

Richard Hayes Phillips, Ph.D.

(precinct level study of Miami County, Ohio).
" It is my professional opinion that these numbers are fraudulent, and that this election has been hacked."
"Hacking the vote in Miami County," December 25, 2004
http://www.freepress.org/departments/display/19/2004/1038

--

Dr. Webb Mealy: http://www.selftest.net/redshift.htm (Bush vote skewed to the Electoral Votes that were needed to win.)

--

Dr. John Allen Paulos: Professor of mathematics at Temple University; award winning author of several best-sellers; writes columns on mathmatics for ABCNews.com and for The Guardian.

"Final Tallies Minus Exit Polls = A Statistical Mystery!," Philadelphia Inquirer, 11/24/04
http://www.math.temple.edu/~paulos/exit.html

--

Dr. Brian Joiner (of the USCountVotes group), "'Voting Glitches Haunt Statistician," 3/4/05: http://www.madison.com/tct/news/index.php?ntid=30826&ntpid=1

--

Dr. Steven Freeman: Professor, Center for Organizational Dynamics, Univ. of Penn.; Karel Steuer Chair for entrepreneurship, Univ. de San Andreas, Buenos Aires; Professor of Management, Central Amer. Inst. of Business Administration (INCAE); member of the USCountVotes group.

"The Corrupted Election," In These Times, 2/15/05, by Dr. Freeman and Dr. Josh Mitteldorf
http://www.inthesetimes.com/site/main/article/1970

also: "The Unexplained Exit Poll Discrepancy" (11/10/04), and "Hypotheses for Explaining the Exit Poll-Official Count Discrepancy... " (draft 2/05). http://www.appliedresearch.us/sf/epdiscrep.htm

--

And that's not even to mention those without a Ph.D. (or whose credentials are unknown), including people like...

Jonathan Simon, investigator for "Scoop" in New Zealand (early whistleblower on the exit polls) : :http://www.scoop.co.nz/mason/stories/HL0411/S00142.htm

Greg Palast, investigative reporter for the Guardian: "Kerry won Ohio – just count the votes at the back of the bus!"
http://www.gregpalast.com/detail.cfm?artid=393&row=0

Chuck Herrin, Republican hacker, election commentary and easy demo of how insecure voting machines are:
http://www.chuckherrin.com/hackthevote.htm

or groups

"Myth Breakers: Facts About Electronic Elections" (2nd edition): www.votersunite.org

57,000 machine malfunction/vote suppression complaints to Congress:
http://www.votersunite.org/article.asp?id=3961

Ohio vote suppression: http://www.bpac.info

Documentation of widespread machine fraud and dirty tricks in over 20 states: http://www.flcv.com/ussumall.html

----

The CONTEXT of fraud is overwhelming, and we have all those folks with their reputations on the line--people from big universities, and the statistics and mathematics departments--saying the the exit polls are one big, fat piece of evidence of a wrong result in the 2004 election, and are calling for an investigation.

WHY, in that case, should anyone who is discussing the evidence for fraud in the 2004 election LEAVE OUT the exit polls?

That would be like leaving out Wally O'Dell. O'Dell's state chairmanship of the Bush-Cheney campaign--and his written promise to "deliver Ohio" to Bush-Cheney--doesn't PROVE that personnel from his voting machine company, Diebold, fiddled the software. Diebold's refusal to disclose the proprietary software used to count all the votes doesn't PROVE that they slipped in Bush-friendly lines of code. But if you LEAVE OUT these facts that strongly point to fraud you are obviously omitting critically important information.

Any cop looking at facts like these in a criminal financial case would be very suspicious. And if you add in all the OTHER FACTS (the touchscreens changing Kerry votes to Bush votes; the massive vote suppression, etc.) AND a 3% virtually impossible (odds of 10 million to one) anomaly in the perps' favor, hidden in their 2nd set of books, these perps are looking at jail time.

Naturally, the criminals' lawyers are going to assault those facts--particularly that 3% weirdness. "It can all be explained!" Uh-huh.

But the prosecution puts, count 'em, one, two, three…fourteen Ph.D.'s on the stand-, against…who?...well, you and the perps' accountants.

Maybe a little doubt gets cast on that 3%--the standard for conviction and jail time being "guilty beyond a reasonable doubt"--and the perps get off. But we're not talking here about a criminal conviction (although that may come). We're talking about the basis of democracy. The standard is far different for the transparency of elections. It's like the standard for judges—they need to avoid the APPEARANCE of improper behavior.

In the case of the 2004 election, the APPEARANCE of impropriety is huge. And Congress was especially malfeasant in failing to investigate it. Most of them, too, are interested parties, just like the voting machine companies. (In fact, the BushCons in Congress deliberately set up the conditions for fraud by failing to require provisions to audit these machines and the vote totals they came up with.) In a REAL election, in a REAL democracy, with a REAL 4th Estate, there is no question that this election would have been investigated.

In THIS democracy, the electorate was left without recourse—not a very good situation for a democracy to be in.

[TruthIsAll]

14. Keep pounding the naysayers with the facts. Thanks, Peace nt

.

[mgr]

15. You seem to think I am not aware of these

I have read and pondered all of these. Much of these address specific places, and do not address the entire country, other than the exit polls. This is what you have to ground yourself in, is that the patterns in Ohio, Florida, Texas, New Mexico, do not translate to every state in the Union, nor in many cases to every BOE within the state. You are working with a select few that require further investigation, and evaluation. It is within these select few the fraud was perpetrated.

Context means what? Guilt before innocence? It is interesting you refer to the cop but not the judge. Context and circumstantial evidence will not get you before the judge. You need intent, you need affadavits, you need discrepant poll books. THAT IS WHAT YOU NEED!

Regardless of the post modern interpretation of what constitutes reality or context, all of us apply a filter to the information we receive, taking some as significant, and other facts as insignificant to weave our story of how the election was stolen. The fact is that I believe that one can tell the story far better, by omitting discussion of the NEP exit poll; and you can sell it to a skeptical and indifferent public far better.

Mike

[Peace Patriot]

23. "the patterns...do not translate to every state in the Union..."

That is exactly the problem. Any bias in the exit polls would be more or less evenly distributed. That's why I said, it's not so much that Kerry won the exit polls, as that the high peaks for Bush (the discrepancy) in selected states is virtually impossible.

And IN ADDITION, Kerry won the exit polls.

---

"Context and circumstantial evidence will not get you before the judge."

Yeah, but they SHOULD get you a warrant, or a subpoena. Ohio alone should have stopped that inauguration--and inspired serious investigation--and it would have in a fair and just Congress. But we don't have that. And in Ohio itself, we had gross malfeasance and illegality in officialdom.

But when you add in everything else--the weird skew to Bush in selected states, exit poll vs. official result, that Kerry won the exit polls, that Florida, New Mexico and other states were also massively suppressing Democratic votes, the lack of transparency, and all the rest--you have an overwhelming case, at the least, for an invalid election and the need for investigation.

There is also the corroboration of Bush's unprecedented low approval ratings, and Americans' high percentages of opposition to all his policies. I mean, it shouldn't surprise us that Kerry won the exit polls, and very likely the election (very likely by a large margin). Everything pointed that way, and still does.

One of the questions I have is how the Democratic Party leadership fits into this picture. I think there is malfeasance, corruption and collusion there as well. They most certainly should have objected to this election SYSTEM. Why didn't they?

The only thing I can think to do about it all--the same determination a lot of people have--is to continue to expose this invalid election, however I can, and to fight for election reform, and start getting verifiable elections.

[mgr]

24. Read the NEP report

"That is exactly the problem. Any bias in the exit polls would be more or less evenly distributed. That's why I said, it's not so much that Kerry won the exit polls, as that the high peaks for Bush (the discrepancy) in selected states is virtually impossible."

You are confusing sampling error with bias. Sampling error should not be directional, and would not affect aggregation of smaller precinct level samples into larger populations. Slight bias may be well within the MOE to not affect characterization, but extreme bias will. That is what the WPE discussion is all about, NEP identified design bias associated with the exit poller such that under sampling of Bush support occurred with the exit poll, and was extreme enough to undermine the published MOEs anticipated for the exit polls.

I doubt that the Democratic party was complicit in any of this. In politics, declaring fraud is about as close to the nuclear option as you want to get. You had better let a proxy address it.

Let me point out that should the consistency test have born out that there was massive fraud, do you honestly think that network news would sit on it? It would not serve their short term or long term interests-- since it would likely come out anyway on some other network, in the newspapers, or through some other venue?

One may believe in conspiracies, but be rational about it.

Mike

[TruthIsAll]

33. Design bias or final exit poll fudging BS?

You say:
That is what the WPE discussion is all about, NEP identified design bias associated with the exit poller such that under sampling of Bush support occurred with the exit poll, and was extreme enough to undermine the published MOEs anticipated for the exit polls.

IS THAT DISPROVED CANARD YOUR ARGUMENT FOR DESIGN BIAS?

And what does uscountvotes.org say about the Reluctant Bush Responder (RBR) after reviewing Edison/Mitofsky's 77 page report?
That the report shows just the opposite happened:
More Bush voters responded than Kerry voters, especially in heavy Bush precincts.

The exit polls themselves confirms uscountvotes:
In the final exit poll(2:05pm on 11/3) 43% of 2004 voters claimed to have voted for Bush in 2000 while only 37% said they voted for Gore.

The preliminary (12:22 am 11/3) had it at 41/38.

So the numbers were adjusted from 41% in the preliminary to 43% in the final. The only problem with that is that IT IS IMPOSSIBLE FOR 43% OF 2004 VOTERS TO HAVE VOTED FOR BUSH IN 2000.
The proof: .43* 122.26 = 52.57 million.
But the Bush 2000 vote was only 50.5 million.

WOULD YOU CALL THAT DESIGN BIAS? OR JUST PURE BS?

[mgr]

40. Have you simulated precinct patterns based on non response?

I have.

Then you would know that with 40% nonresponse, two variant responses due to sampling error would exceed 5% MOE. If in reviewing numerous precincts the sampling error is directional, then it is bias.

You have oversimplified the exit poll to address only registered democrats and registered republicans. Try using the independents as well, and such sampling error becomes even more likely--what you need is a single additional democratic response (to increase democratic turnout), and a single additional independent voting for Kerry to do this to bias a precinct to wards Kerry.

The Canard is disproved only if you ignore NEP's response. USCounts Votes' inference was addressed by Mitofski, and in essence, his response was that the bias effect was significant in precincts other than those with strong Bush support, and that precincts sampled with 80% Bush support were few (as one would expect, I might add). If one consults the precinct WPE ratios, one might infer that the precincts with WPE greater than five to wards Bush, are those with a preponderance of Bush supporters; just as one would assume for Kerry, except there are almost twice as many of them. This asymmetry should not exist since the sampling design should balance the two. I would not want to speculate on the significance of the sparring between USC and NEP, as there are always hidden motivations behind what appear to be dunderheaded observations. However, I will offer that they may have been trying to get a handle on which type of precinct sampling design NEP pursued, centers and edges (each cluster), or heterogeneous populations (interacting clusters), each which I suspect can be approximated with an appropriate experimental design so further analysis may be performed.

GIGO--43% of those sampled does not necessarily translate to 43% of those that voted. You may also want to calculate how many questionnaires this represents, since I don't believe this question was on a subsample. What it suggests (without other information) overall is nothing--what were the other choices? voted for Nader, did not vote,... and how did these break down? The context cited is incomplete. What it hints at is that the Democratic Party was unsuccessful in persuading previous Bush supporters to abandon Bush.

Mike

[Blue Shark]

19. Man Do you ever Nail it!

Take a raise out of Petty Cash (and make a down payment on a piece of Canadian Real Estate)

[kster]

27. With all we

know, when do we start standing up for ourselves and demand this election gets investigated,not tomorrow or the the next day but now!There does come a time that we have to stop questioning each other and get the answers we are entitled to.

[GuvWurld]

32. We're past investigations...we have no BASIS for confidence

Conditions do no not exist for the results of U.S. federal elections to be accepted without question.

We have no reason to believe the results reported from U.S. federal elections.

We have no basis for confidence in the legitimacy of the results reported from U.S. federal elections.

Read the No Confidence Resolution

Everything else being discussed, while intellectually engaging and stimulating, blows past basic and obvious statements of logic which provide a much better frame for action. Since last April I have been seeking anyone willing to make the argument that there IS a BASIS for confidence. I've had no takers. This is a winning frame.

Think of it another way. Many of us knew before the "election" that we couldn't possibly have a legitimate outcome, if only for the reason that there would be no permanent record to forever sustain unanimous agreement about the outcome. Shouldn't this, by default, be part of the criteria for any contest?

Reforms have to go much further than mere paper trails, and again, in the paradigm of how we frame our work, we are best served by stating that our comprehensive collection of election reform demands is intended to create a basis for confidence where none exists. Is this not a precise extension of the default mentioned above?

[bleever]

28. Rock on, P-squared.

With the elephant stinking up the tea room, some people want us to focus on the tea leaves.

[Name removed]

29. Deleted message

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

[skids]

31. You want a "consistancy check?" OK, try North Carolina.

...where the pollbook demographics don't match the NEP poll, but the NEP poll final weighting is in the wrong direction in many demographic groups, and when compared to the completion rate, the applied weighting was biased against 18-25 year olds.

To say that the exit polls are a dead end in determining what went on is spurious. They are information, just like any other source of information. They can be used as a component in building an agument, and there is nothing special about them that says they are completely meaningless.

I really don't see why you feel so threatened by TIA's stuff and why you think it is worth so much energy to tackle it. You say you want to develop an exit poll that works to detect fraud. Well, got news for you. I hope you can find someone to fund it other than the MSM, because they won't pay for it. It should be amply obvious from their coverage bias that what they want for their money is a nice warm and fuzzy that the election system is just fine and OK, regardless of reality, and factoid fodder for talking points.

No, you'd have to recruit and train a whole lot of volunteers, on a volunteer basis. I don't see that happenening now, and it would have to begin planning real soon now in order to make it for 2006. So unless you have several million dollars up your sleeve, or the magic ability to organize and coordinate a huge force of somehow incredibly eager volunteers, trying to figure out "what would make an exit poll into a definitive fraud detector" is a mute point.

[mgr]

47. North Carolina State Poll or National Poll?

Which are you addressing? Do you have the poll books from precincts where NEP sampled? Are you addressing the demographics characterized by the national poll samples, the same ones that I characterized as crap. Guess how many questionnaires came from North Carolina if the total is ~13,000? Beyond that, I guess I don't follow your point.

As to why I expend so much energy (I really do not)would be the same reason you might if someone completely mis characterized an area of your expertise. My life is built around teaching and properly applying science. But, in this discussion, I would gladly defer to the social scientist that works in this field, since my experience with demography is insufficient to address exit polling as a white box (a gray box, yes).

You are speaking to legal proof, not mathematical or scientific, when you state the exit poll is information (or else we are just conspiracy buffs). Information based on what, biased sampling? Information to formulate an opinion that fraud occurred, information that must be empirical, accepted by experts in the field, and not subject to refutation. Does the exit polling, as presented by TIA, meet this? Or is the argument stronger without it? Mitofski stated that characterizing fraud was not part of the exit poll's design (as I am sure you are aware).

Look, North Carolina's election was fucked over by electronic voting, do you need the exit poll to tell you that?

I really do not know what MSM's interest in having exit polls are other than calling a horse race. But do we have a national interest in having them, and should it be part of government or non-partisan election oversight, funded by the government? I think so, and I would like to see it as part of this years legislation by the Dems.

Otherwise, I think it might be fun to administer an exit poll to a million voters across the country. 2006 is a little tight for a grant, but 2008 has potential.

Mike

[berniew1]

34. Your argument has major problem; widespread fraud was documented

Widespread fraud, systematic (illegal) dirty tricks, and manipulation of ballots, registrations, absentees, provisionals has been documented in many states using information from the huge data base of election fraud and irregularity reports from the EIRS system, Common Cause, TrueVoteMD, county election office complaints, etc. plus the Ohio recount,etc.

Touch screen switching was documented in 18 states; default of straight ticket Democratic votes to Bush or other than Kerry in at least 10 states, ballot manipulation in many counties of Ohio, etc.,
systematic dirty tricks(illegal) and malfeasance/misfeasance of officials to reduce minority and student registrations and cause them to vote provisionally in the wrong precinct in many counties of many states; manipulation of registrations, absentees, provisionals in many counties of many states.

http://www.flcv.com/summary.html
http://www.flcv.com/ussumall.html
http://www.flcv.com/studentv.html

While it is true that an underlying factor in most of this is highly partisan activists, officials, poll workers; the big swing was not just due to "bias". There was real fraud and illegal acts and malfeasance involved in much of it.

[mgr]

35. Where do I deny this?

TIA has staked out the position that fraud was widespread, and intended to alter the popular vote. Everyone else addressing the matter, is focused on the electoral vote issues with the battleground states.

Is the documentation so widespread that it applies to every one of the fifty states? Or is it limited to the battleground states? I've looked it over, and if you consider allometric effects, many of the reports can be reduced to insignificant, or typical of confusion on an individual level that has nothing to do with disenfranchisement. The example I use is for California, over 1,000 incidents are reported for Los Angeles County, but when you review these, they are almost all registration based--most are voters unsure if the BOE has their registration--that when you factor these out, there is little left. Since it is registration based, if they could not vote, it would not show up in the exit poll.

The problem is that if your polling has an MOE of 5%, and if all the fraud was meant to do was swing 3%, then this is hidden in the exit poll MOE.

Does this mean that the exit polling data cannot be used to detect fraud in each state, probably not. But until we are able to ground which and where each precinct was sampled, it will be of little use.

Mike

[TruthIsAll]

37. mgr, stake out these FOUR RED FLAGS.

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x346284

[mgr]

42. Employing the same error to its logical extremis.

Pointless, if the numbers from the start are jiggered to match a predetermined outcome. I have already addressed this many times.

Mike

[TruthIsAll]

46. Your logic is pointless. The Final numbers were jiggered to match

an known impossible result.

I'm talking about the 43% which the final exit poll says voted for Bush in 2000. They revised the weight upward from 41%.

That is proof they overstated the Bush vote by at least 2 million, AT MINIMUM - and probably by 3-4 million

Answer the question: WHY?

You cannot spin it away, so you will just ignore it.
But that is your MO.
You will never change.

[mgr]

48. What is the difference between independent and dependent variables?

Imagine that the colossal multiple regression equation that found that 75% of voting behavior was related to party identification. Other prominent characteristics were gender and ethnicity. How, if you are changing the weights for party identification by changing the questionnaires included, control for the other dependent variables? Gender and ethnicity do not vary independently of party identification, do they? So how is it that if you change the number of respondents, you expect these numbers to remain fixed, or have some howlers as a result? I don't find this problematic, and I do not find this predictive or indicative of fraud. However, I do think you probably wasted part of a fine Saturday generating this GIGO.

Mike

[TruthIsAll]

54. Your post is totally incoherent.

"Imagine that the colossal multiple regression equation that found that 75% of voting behavior was related to party identification. Other prominent characteristics were gender and ethnicity. How, if you are changing the weights for party identification by changing the questionnaires included, control for the other dependent variables? Gender and ethnicity do not vary independently of party identification, do they? So how is it that if you change the number of respondents, you expect these numbers to remain fixed, or have some howlers as a result? I don't find this problematic, and I do not find this predictive or indicative of fraud. However, I do think you probably wasted part of a fine Saturday generating this GIGO."

One demographic is absolutely wrong (I proved it it another thread, Bush's vote is overstated by a miminum of 2 million):
1) How did you vote in 2000?

Two others are entirely inconceiviable based on the last 3 elections:
2) Party ID
3) Gender

I'm sure ALL other demographic weights were also adjusted incorrectly. And the above three are consistent with a bogus Final Exit Poll.

IF JUST ONE WEIGHTING IS ABSOLUTELY WRONG, ALL OTHERS ARE SUSPECT.

IF THREE ARE WRONG, FORGET IT.

If The Exit Poll was bogus, so was the recorded vote, since the Poll matched to it..

So if you want to do regression analysis, include ALL the bogus weights.
And so, I am sure are all the other demographics.

[mgr]

59. Your analysis is beside the point

I apologize the for the apparent incoherence of the previous post, but was rushing to a doctor's appointment. I had some reservations about the clarity of what I wrote, but figured I'd get to it today.

What I am trying to explain is that what NEP did was re-sort the responses to conform with the party ID reweighing, and adjusted other weights from these. I also did not make it clear that I considered NEP's work GIGO, I consider all the trends they identified occurring with the election to be crap. Because all the other demographic qualities nest within the party ID, they will vary with that weighing. I will have to make an evaluation of what you have later, but I suspect it derives from single weighing, not multiple re-weighings. However, it is not three errors, but one.

Because the resolution of all voting patterns for this election are from a bias at the precinct level that increases MOE across the board, and what we are specifically considering is the ability of the NEP data to resolve whether or not fraud occurred, your argument is beside the point. All it does is argue that we should not trust the data at all.

Mike

[libertypirate]

39. Dear any moron who thinks that there is a 'myth' of fraud...

Simple Answer....

If there was no fraud why can we not have a court hearing of the matter, and why can't we have an investigation, everything has been unofficial. In all the brilliant number crunching and illogical logic passed to defeat it, you all forget that fundamentally we have been played as fools. No investigation no crime, officially swept under the rug.

TIA should be able to say this shit doesn't add up... And with every other tidbit of evidence collected that the election was rigged we should all say I bet your right TIA. I don't need dead on balls accurate numbers. What I am saying is that TIA has proven beyond personal reasonable doubt that one of the results were not accurate.

If you want numbers that prove beyond DNA that the election was controlled here you go... If a glitch were just a glitch then I would expect them to be distributed evenly. What I don't expect is an uneven distribution of glitches that favored *. Even more I wouldn't expect those glitches to above 90% benefit one candidate over the other. I don't expect this because a glitch should be random occurring error, but at a +90% benefit to * random is barely even a possibility.

Now why no investigation?

A question beyond the marketing what are the “Exit Polls” designed to do? Perception of reality folks, since we can’t know what the results of the election are perception wins.

[mgr]

41. Do some research and learn patience

There are lawsuits and there are FBI investigations. What more do you want?

Mike

[Chi]

45. What more do you want?

Heya Mike
Here's a couple of novel thoughts....

Investigations by the Attorney Generals from each state with excessive reports of election anomalies and/ or suppression.

Investigations by congress, congressional committee, or independent council.

How bout ANY official governmental investigation at all, state or federal.

It appears the official government position is to ignore it, or do nothing. I hope you don't agree with that position.
(if you know any official federal government investigations into this last election, feel free to post it. I'm good at admitting I'm wrong)

[mgr]

49. What do we have so far? (sorry no links)

I'm a skeptic but also an optimist.

Two suits in Ohio. Some articles indicating FBI looking into matters in a couple of counties, again Ohio. Conyers of course.

Hugh mess in North Carolina, apologies to Skids but I have not kept up. Maryland. Florida. New Mexico. Any one of these blowing up puts the matter back on the national radar.

There are places that I suspect of problems that are not mentioned here, Arizona in particular (Colorado's also off the radar). Any place we shine a light is a start, and you must keep faith, and patience, and try to avoid flabby reasoning. It will come out in the long term. It is not easy, which is why I have such a problem with exit poll analysis.

Think of this as a science problem, the official position is the null hypothesis that has to be disproved. Citizens do not want the government acting precipitously (like Iraq), and inertia is the characteristic of all bureaucracies (I should know, I work in one)

BTB, I have no problem with being wrong, this is a community, what happens to me individually that helps the community, I'm cool with.

Mike

[Chi]

51. What we seem to have is all citizen initiated.

In the country I once knew, If rights were so obviously denied (massive repression) the government was expected to police itself.
Separation of powers, checks and balances, the way our government was designed.
I don't find it acceptable to see every investigation initiated by the citizens who were denied rights.
The burden should not fall on the individual.

There is a big difference between precipitous and indifference.
I only expect investigation, not immediate action.
But I guess that is too much to ask since all we get from D.C. is an attitude that 'nothing happened..why look into it'.

How much patience would you propose for investigation into Florida 2000 election, or is that a different story?

BTW, I was NOT insinuating you would not admit being wrong if you were, I was just saying I could be wrong.

[mgr]

55. This is a democracy

The will is with the people, freedom requires taking responsibility--the protection of our rights rests with us, we can delegate it, but we cannot expect our proxies to know our concerns. What you describe would be more in line with the benevolent dictatorship of Plato's Republic and the like. I think you are overstating the role of big government, and if you were to examine popular movements from the past that address issues of rights, they all have their origins with the people. This is the position one takes when one is tired of the vigilance, sleeps, only to wake up in chains.

As to being wrong, I did not take your comment as an insinuation, but have wanted to state for some time that when addressing any issue, one should not fear being wrong--it is no big deal. Establishing that someone is wrong, for our purposes should not be looked upon as counting coup. Being wrong is OK, it is making errors that are not (e.g. persisting with a wrong assumption, after it is shown to be wrong).

I have seen individuals make a big to do over a slight mistake, as proof that the individual's other positions were subject to the same type of error. Veracity has its place in a court room, but each argument must be weighed as to its merits, separate from where or who the source was.

Mike

[Chi]

57. I think we might see our government differently.

Actually I disagree that the will of our country rests with the people, it used to, and it should, but it doesn't anymore.

I also disagree that the protection of rights belongs to the individual. I believe the President, and Congress, all have to swear an oath to defend the Constitution (our rights).

I also glanced over the amendments that protect our voting rights. They all appear to attach this...
"Section 2. The Congress shall have power to enforce this article by appropriate legislation."
Which to me says, they are responsible to enforce those rights, and have the authority to.

As far as our proxies knowing our concerns...I'm confident they got that message loud and clear when the electors from Ohio were challenged.

I think the term 'big government' wouldn't apply based solely on the subject of defending our rights.
I would actually think that would be one of our government's highest priorities.

I agree, rights movements in the past were grass roots based.
But the enforcement of such rights were not, at least to my limited memory and knowledge.

[mgr]

61. Its the follow through that counts

I doubt that we are too far apart, but when I recommend patience, what I am saying is that the process to rectify what has happened is underway. It may appear glacially slow right now, but we cannot lose sight of the prize. Our reward may only be reform of how elections are conducted, or a descent into anarchy that we nearly experienced with Nixon's resignation (I would like to envisage an impeachment, but after the crap the country went through with the attempt to impeach Clinton, the popular will may not be steeled to address true high crimes and misdemeanors)

You may want to ask yourself if you have reified our government into an entity of its own volition. It is only an instrument that is meant to reflect our collective will. If our will is weak or fragmented, then government will not address our needs. I think that the nation is fragmented by partisanship to where the collective interests of all are overlooked, and addressing the fraud issue may bring that to an end.

Mike

[Chi]

66. I also favor patience

I also favor patience, but I am not as optimistic about it as you.
To me, the only real chance we have left is the recount suit. If they can get forensics in to examine the tabulators, have supervised hand recounts of a number of counties (chosen by the Dems/Glibs experts), and full examination of the poll books...I would be satisfied.
Actually...throw in an investigation to the Warren county debacle for good measure. 8)
Then there's reform...

I didn't know about any Nixon resignation chaos, guess I should read up.

I haven't reified our government, I only expect it to be what the constitution says it should be. Since the Constitution is a living/breathing document, so must our government match it's essence.
But the basics I expect of it is to defend 'the people', whether it be from invasion, or it be a covert attack on our rights.

I suspect this is the smallest common denominator of our disagreement.
I think our Government should protect our rights, you think it should be done by individuals using the system.

[libertypirate]

50. What I want.....

Due process, I don't think I am overstating what should be expected.

You tell me to do some research, how about taking your own advice.

There were over 1000 whistle blower cases reported last year I think 10 saw any action. Do you even know what you are talking about? Only 1 percent of whistle blowers are being found telling the truth. Bull Shit!

Doesn't matter how many investigations, wagon circling lawsuits, these people aren't playing by the rules; they make them up as the go along.

[Chi]

53. Due process....Well said

Heya libertypirate (waves)
Sorry if I butt into your conversation inappropriately, I'm sketchy on forum etiquette.
I meant no disrespect.
(bow)


If anyone wants a link to the ignored whistleblowers story...here ya go
http://www.peer.org/news/news_id.php?row_id=483

[mgr]

62. Federal whistleblower protection program

I should remember a site I visit every two to six months (slaps forehead). However, these are folks who have already made the issue known, and seeking protection through federal law. As I alluded to before there are many ways to blow a whistle without endangering yourself or your employment.

Mike

[Chi]

69. Chi backs away after hitting 'post message'.

** Mike, I know you said you don't want to talk about this, so feel free to take a pass.**


I don't believe this part has anything to do with protecting the whistle blowers employment.

=========================Copied & Pasted===========================
According to the figures released by Bloch, in the past year the Office of Special Counsel—

* Dismissed or otherwise disposed of 600 whistleblower disclosures where civil servants have reported waste, fraud, threats to public safety and violations of law. Bloch has yet to announce a single case where he has ordered an investigation into the employee’s charges. Bloch says that 100 disclosures are still pending; and

[mgr]

56. Link? Source?

I am a fucking hell of a lot closer to whistle blowers than you may think. Your statistic describes what area of work--government, private? Are you describing the whistle blower protection program? Why should doing something one thinks is right be easy or safe? Look at how you speak to me when I am saying the emperor might not have any clothes, would an employer behave any different from you?

Civil servants that are whistle blowers are often idealists that assume the agency they work for will do what is right, and will do them no harm. Those civil servants who are a little more worldly, or better connected, leak to the press and public advocacy groups to address the grievance, but they are never accounted for because they are protected by their anonymity. Local press, unlike MSM, takes a very dim view of local government malfeasance, and will protect a whistle blower's identity. I would assume the same for private.

Vetting of complaints and whistle blowing occurs, the intent when engaged in a lawsuit is to make a case winnable, this may be one possible reason there are so few verifiable perfect whistle blowers. But, when one gets to this point in the whistle blower 'process', there is a strong likelihood of informal vetting finding the facts did not stand scrutiny.

Get your back down, and use your reason. Due process takes time, Hotspur. If you don't mind, I do not wish to discuss whistle blowing any further. If you wish to continue, PM me.

Mike

[autorank]

63. Tawdry Rhetorical Devices Discredit mgr's Argument...

"mgr", I read this thread with some interest since I have an open mind on the entire election fiasco. I also have a strong belief that TIA's analysis along with the other statisticians is excellent.

In reading your post's three things popped out:

1) In criticizing what you call the "strong fraud" argument based on Neaps you state: "The strong fraud argument would have it that the Dukakis/Bush contest was won by Dukakis as shown in the raw exit poll results, but went to Bush due to election fraud" This is lame beyond words. TIA has pointed to the 1996 and 2000 NEPs and as a measure of their precision. Yet you refer back to 1988. Why would you do that? To obscure the power of the TIA, et al, analysis using NEPs and the real digging Truth and the other s did. This is deliberately misleading and makes your entire analysis suspect.

2) You ignore the elephant in the living room, the Ukrainian elections. Are you arguing that the exit polling there was of the magnitude required to catch fraud? Are you arguing that the polling design and methodology were superior to the NEPs? Ironically, the Ukraine results validate the TIA, et al, analysis.

So with 1996 and 2000 NEPs showing real precision, the Ukraine experience showing NEPs are actually able to catch fraud, what is the final rhetorical device?

3) Argument by veiled authority. If you only knew who I was you would really, really believe me. You state: "I am a fucking hell of a lot closer to whistle blowers than you may think." I'm not asking you to reveal your identity, far from it. I am objecting to the veiled reference to some special knowledge based on not revealing it.

Here is my indisputable series of proofs regarding election fraud. If there were no tricks with the NEPs, the data would be revealed for total examination. If there were no fraudulent acts in Ohio, the entire process would have been made transparent from pre election planning, election procedures, and post election analysis. If there were no problems with voting machines, tabulators, etc., the software and procedures, the personnel involved, and the entire process would be open to all for examination. And, if a real recount could clear our suspicions, then the odious Gov. Richardson of the small state of New Mexico would not have squashed the recount there by trying to extort a $1.5 million fee from those requesting the recount. This is all you need to know that the election was stolen through fraud.

The movement for election reform, like all movements, relies on the axiom: lead, follow, or get out of the way. I'm tire seeing TIA and others spend their time responding to these types of posts.

[mgr]

64. Big boys use rifles, not shotguns.

Your post reminds me of a story about a drunk duck hunter that shot himself in the foot with his shotgun, because it was moving.

You do not address the gist of the thread, but question my veracity. Nor do you do my argument justice. An open mind that relies upon belief and good will is not a skeptic's, but a partisan's. Nice rhetorical trick. Let's see, you selectively fail to note that I explicitly don't associate TIA with the strong fraud position; you argue that the exit polls in the Ukraine are right on, when it has been shown over and over again that it depends on which exit poll and which election you select; and that I may not have credentials to address how whistle blowing occurs. What do any of these points have to do with the fact that NEP could discern strong bias at the precinct level without resorting to the election results for confirmation?

With your incisive review and analysis of what is going on with the site, let me inform you that I am the one trained in statistics and experimental design, TIA is a mathematician (more likely a software engineer) playing at being a statistician. Telling some one to get out of the way because they are undermining your movement for election reform because they state that you need to get your facts and arguments straight is just about the most ignorant position one can take in a party grounded in reality.

Mike

[FogerRox]

65. Autorank--you hit the nail on the head.

[autorank]

68. Thanks!

:hi:

[autorank]

67. haha, Funny analogy...

...and a little hostile too (hmmm..). I don't hunt and I don't shoot myself in the foot.

You used the argument by authority again, "trained in statistics and experimental design" versus "TIA is a mathematician," which, of course, is just that, an argument by authority. It fails to prove anything, particularly in a largely anonymous forum.

How about commenting on the core of my message:
-------------------------------------------------------------------
(From my original message)

Here is my indisputable series of proofs regarding election fraud. If there were no tricks with the NEPs, the data would be revealed for total examination.

Comment:

If there were no fraudulent acts in Ohio, the entire process would have been made transparent from pre election planning, election procedures, and post election analysis.

Comment:

If there were no problems with voting machines, tabulators, etc., the software and procedures, the personnel involved, and the entire process would be open to all for examination.

Comment:

And, if a real recount could clear our suspicions, then the odious Gov. Richardson of the small state of New Mexico would not have squashed the recount there by trying to extort a $1.5 million fee from those requesting the recount.

Comment:

This is all you need to know that the election was stolen through fraud.
----------------------------------------------------------------

[FogerRox]

70. Yeah I think the transparency Issue really sums it up.

Locking down the WArren COunty center because of a TERROR threat, which no=one--FBI_homeland SEc CIA Who ever----
has never been able to confirm.

[autorank]

71. Exactly, and has anyone offered an explanation for that anomaly.

Our side demanded one but nothing forthcoming. Of all the outrages, that one is my favorite purely for its absurdity. Yikes!!!

[FogerRox]

73. I used to do Magic shows--birthday partys and such

Ever heard the term misdirection?

LOL--

"Theres terorrists here in WArren county -------
----------RUN FOR YOUR LIFE"

----OK down load that file in the GEms --------

[mgr]

80. And Warren County was repeated in every BOE in every state.

Give me a break, we know that this is under close scrutiny, and is part of a legal action in Ohio. And it is singular.

Mike

[mgr]

72. But you took a shotgun approach...

The mark is the argument, you hit at everything but, kind of like a spray of buckshot. You shot yourself in the foot with 'the open mind but I believe', argument, being clearly decided in your opinion, but maintaining a pretense of your not having made a choice.

You are making the same confused analysis that everyone else engages in. Why do you need the exit poll argument, when there is already compelling and sufficient evidence for abuse in Ohio or Florida, or New Mexico for that matter.

Why should NEP release the data? It is, until determined otherwise, proprietary and intellectual property. You are asking someone to give up a livelihood in return for what?

How do points two and three relate to the national exit poll, or to the results in the state exit poll. If someone tells you a margin of error is 5% of 100,000,000 how many people is that. How many votes do they need to steal to take the election? Last I heard its about 3,000,000 swing. Let's see which is larger, 5,000,000 or 3,000,000?

Let me see how argument and refutation work here. I make a reasonable argument based upon my experience and work, that functions to negate a logical inconsistency, no one seems to refute the argument on its merits, but resorts to a variant on the ad hominen type of argument, that I cannot speak from authority, though the argument stands or falls regardless of it. But, in your mind, TIA is the authority, because his arguments align with your prejudices, and ignorance--you make the assertion that he is a statistician, when he himself has stated otherwise.

The problem is that you have it back wards, if the argument I make stands, and is verified, then the authority I state that I have is supported. Thus it may give additional weight to later arguments I may make, but it does not make them consistent, valid, or correct.

Mike

[Bill Bored]

74. I see this thread is still kicking around, so tell me mgr,

if the exit polls were off, why weren't they off equally in all states? Why were they off more in states with Republican Governors for example? And why in states with straight party voting? And why the most in states with Republican Govs AND Straight Party voting? Coincidence?

If nothing fishy went on with this election, why wasn't the within-precinct error of the exit polls roughly the same in all states?

Let's say the polls were off. Say by 5% before the adjustments. Why were they not off ~5% everywhere and why were they most inaccurate in areas where the ruling party is running things, or where a certain votin' machine configuration (straight party ticket) can be manipulated to favor a particular party?

I think it's time we stopped looking to these polls for ABSOLUTE accuracy and think in terms of RELATIVE accuracy. Then we might be able to infer some things about how the fraud was committed and by whom.

What do you think of that?

[mgr]

78. You are asking a lot

Sorry for the late response.

1.) I don't think your analysis of a distinction between governor's party affiliation and exit poll outcome holds much water. How many of these states have separately elected secretaries of state, and how many are appointees of the governor? Here in California, we have a republican governor with a democrat running the SOS (at the time of the election). It needs further refinement, and I would bet that a nested ANOVA would show that if would not pass an F test for significance.

2.) If you have an association of SP with a republican governor appointed SOS, you may have something. But how many states is this? Does it accord with reports of election day discrepancies?

3.) Variability in the states polls is a different matter from the national election. It is something I am trying address in a different thread as to how one can use these results to possibly detect fraud. Your argument pretty much stands statistics on its head, and calls into question why there is any regional variability in people at all. I suspect that is not what you wished to argue, but....

4.) Paragraph three puts us back into the position we were at the start of the election, your position is somewhat nihilistic.

5.) I think that the exit polls at the state level can be used for our purpose of detecting fraud.

6.) What you may need to define within your mind is why this issue matters. The accuracy of the exit polls and their precision is of extreme importance if computer tabulating and the internet are part of the vote counting apparatus.

Mike

[farmbo]

75. Gawd TIA...what were YOU THINKING!?

Mgr tells us that because Jimmy Carter conceded early in 1980, based on early Exit Polls, that skewers your argument on the fallacy of the final 2004 Exit poll.

Who could argue with THAT logic? :smoke:

And...Oh yeah...Because some of the early 1988 Bush/Dukakis exit polls (compiled by a different pollster) were changed, that means your research on 1992, 1996, 2000 and 2004 exit polls is patently flawed. Whatever. :bounce:

He ignores the Ukraine fact pattern all together...too inconvenient. :crazy:

And as for as the unwavering efforts of each and every Ohio GOP elections official to keep investigators from actually viewing ballots, examining poll books or testing software...well they're just good-hearted people trying to save the Ohio taxpayers from slimy conspiracy theorists. :silly:

Even though his theory requires several great leaps of faith--and makes my brain hurt with its circuitousness-- well, we just have to get over it, move on, and quit being sore losers. :headbang:

So, once and for all, quit bothering these folks with the truth and go back to sleep. :boring:

[berniew1]

76. Exit poll only tells likely outcome; have to look at real cases to determi

determine what caused the shift; and the fraud, manipulation, suppression that caused the shift is well documented
http://www.flcv.com/ussumall.html
http://www.flcv.com/summary.html

I don't know why this kind of thread is still going on??

The E P s are useful to indicate the likely outcome and where to look for problems; but you have to look at what happened in the counties and precincts(using things like the EIRS and county SOE irregularity reports) to assess what really happened and why the swing. And there were hundreds of thousands of such reports or fraud and irregularities on election day; a really useful and large sample of what occurred. There really was a paper trail; and it paints a clear picture.

[autorank]

77. Great links but the TIA work is an important parallel path, one that

reinforces all the other paths. My great hope is that we'll get a smoking gun that is so compelling and comprehensive, we will get the attention we need. Your links are extraordinary and reflect amazing work. The TIA work is the reductio ad absurdumusing the "election official's" own data. It is compelling and needs to continue as part of the overall information gathering and history of this fraud. I have total respect for ALL OF YOU working on election fraud. It is the most important movement in this country since the Civil Rights movement.

[mgr]

79. And, just where were you?

When I stated that in Ohio one needed to document the actions by obtaining affidavits and inspecting poll books?

For one who made a claim to a stats background you are pretty cavalier in rejecting the purpose of an exit poll to address fraud, such as ballot stuffing that the average Joe cannot detect except by ballot comparison.

Mike

[davidgmills]

81. I have posted on this board many times

Why I think, as an attorney, the Courts are totally incapable of handling a national election controversy.

State laws govern election contests and are designed to redress state election controversies only. Their timetables make all national election contests moot before any real redress can occur. So to ask for affidavits makes no sense to me.

What does bother me about your argument is this. First, we now know, as a fact, that early in the morning hours of November 3, the exit polls showing a Kerry lead of three points was changed to show an exit poll matching the actual tabulations. The question is why the change? Why was this was done at such an hour with so little forethought? Why couldn't Mytofsky let the numbers be?

What would have happened if on the morning of November 3, all the national media outlets had informed everyone that there was this exit poll / actual poll descrepancy and the pollsters were sticking to the numbers they got? Would we have had another Ukraine? Would Kerry have conceded if people had taken to the streets? Well, we will never know.

I think it goes against most scientific ethics, of any kind, to change numbers of this importance, under such circumstances.

Secondly, why does it take until the day before the inauguration for Mytofsky to produce a report attempting to justify the early morning changes of November 3? This is the second highly suspicious fact. If you can make changes in an hour, why does it take two and one half months to justify those changes?

Thirdly, why when Mytofsky released his report, did he not release all the underlying data which could justify his early morning November 3 changes? This is also a fact.

It is a combination of these facts and the exit poll numbers which greatly increase the suggestion of fraud. It is a little like the murderer leaving town right after the crime has been committed.

[mgr]

82. Let me expand a little

The statement I made was during the time of the Ohio election suits being prepared, and were within that timeframe before or while the vote was certified. I do not disagree with your opinion based upon your practice as an attorney, but would ask what you mean by real redress?

We know from the poll taking community that the national election results were not where the exit polls emphasis was, and that the methodology was understood to allow a priori and a posteriori weighing of results. The change reported at 3 a.m appears to reflect when the a posteriori weighing was completed, or TIA's contamination of the pristine sample occurred. It has the appearance of scientific inappropriateness only if one were changing the numbers of those originally sampled, if one changes how they are weighed based upon a known and published methodology would not be so.

This is where your epistemology and mine veer away from each other is with what constitutes proof. I am only evaluating patterns from a scientific perspective which requires slightly different standards as to argumentation and cause, and one area that science does not address is accumulation of circumstance, which to a scientist constitutes speculation.

If I may, I was surprised that NEP was able to get the report out so quickly. I thought March likely. Consider the levels of post election verification they had to go through to determine it they had the right talleys, weighted the numbers properly, run the appropriate multiple regressions to address what appears to be sample bias. One could equally argue that the timing was meant to embarrass Bush on Inauguaration Day--the interference by the Ohio SOS is quite damning.

The fact that they took another month and half to make their data available, is from a scientific perspective, is moving along quite quickly. Usually, one does not see one move from paper presentation, to article, to depositing one specimens in the herbarium in a period of months, but in years. The evidence of strong bias that they attribute to poorly trained pollsters undersampling bush supporter may reflect a rush to publish.

You have to consider one other factor in your rush to prosecute NEP, why is there no hue and cry over the LA Times exit poll whose results are consonant with NEP's? Is it possible that both were to deposit their raw results simultaneously? There is no explicit methodology statement for the LA Times polling as there is for NEP, there is no election follow up report, just raw results deposited with Pew. I would suggest that NEP's behavior so far has been far more transparent and open (of course, I am just underway addressing the Times poll, and I may be wrong about all of this, but the fact that I am talking about an exit poll in April, and know of no one on this site to address the Times poll in depth, begs the question).

Mike

[davidgmills]

83. Real Redress

I will only respond to the legal issues as the stats are not my field.

There is no such thing as a federal election law designed for a presidential, senatorial or congressional candidate to challenge election results. One of my major issues is that if we are going to have election reform, we badly need federal statutes for federal electoral controversies. Historically, both senators and congressmen were not elected directly by the people; they were elected by the state legislatures. That changed about 100 years ago. As you know, we still don't elect presidents directly; we do that through the electoral college. Probably because we have never gone to directly electing the President, there have never been any federal laws enacted to challenge elections for our federal officials. Our federal officials are elected on our state ballots and all election contests are also done with state election laws.

No one cares if in a state senator election, it takes nine months to get discovery completed and an election contest trial completed. That is about the average timetable for the average state election contest. State election contests are usually only county wide and not state wide. People criticised Gore for only challenging Bush in a handful of counties; but that is the way it had to be done. There was no such remedy as a statewide electoral contest. That was not an option for Gore.

State laws simply don't work on the national level, especially for President. In Bush v. Gore, the Supreme Court never really said Bush won, it just sent the case back to the Florida courts for further work. But by then the clock had run out. It was very apparent to me that the same thing was going to happen in Ohio from day one. The state law timetables for electoral contests are not at all conducive for producing a recount and concluding a lawsuit in the timely manner needed to reslove the contest prior to electoral college certification.

We need two changes. First a federal ballot by itself, separate and apart from any state ballot. Such a ballot would not involve a recount of "state ballots." This would expedite recounts of elections of federal officials and also cut down on voting errors for federal officials often caused by long state ballots. It would produce a very uniform ballot all over the US.

Secondly, we need a federal election law providing for statewide recounts, with federal court jurisdiction, with rocket recount procedures, a rocket docket, with rocket discovery and a trial all prior to any electoral college certification date. Without it, there is no meaningful redress.

[mgr]

84. Damn straight

I fully agree with you.

Mike

[davidgmills]

85. And the Federal ballot ---

Paper, pen, and hand counted after being placed face down in a transparent plastic box.

[Peace Patriot]

88. "Without it, there is no meaningful redress."--davidgmills

"Without it (federal ballot; federal verification process), there is no meaningful redress."

-----

I agree that something must be done to provide citizens with avenues of redress in a stolen federal election--and that other national measures are needed (for instance, elimination of the Electoral College, instant runoff voting, and mandated auditing recounts); however, I don't agree that there is no other meaningful redress.

When the Ohio Secretary of State began inventing rules for voter registration--such as having to submit the registration on 80 lb. paper--a decent federal government would have immediately gone into action to insure that the Voting Rights Act was being enforced, and thus could have prevented the massive abuse against black voters and other Democrats that occured there on election day.

Instead, the Bush Cartel sat back and let it happen. They WANTED it to happen.

When those abuses were exposed, a decent federal Attorney General and a decent Congress would have investigated them. We do not have a decent AG nor a decent Congress. They don't care about justice and fairness.

When the federal election reform bill (HAVA) was designed, supposedly in answer to the election problems in 2000 in Florida, the President and Congress should have gone out of their way to, a) insure transparent elections, with essential provisions such as a paper trail in electronic voting; b) prevent conflicts of interest such as Bush partisans owning and controlling the electronic voting machinery; and c) prevent corruption at the local/state level, with 4 billion dollars at issue.

Instead, they went out of their way to create an egregiously non-transparent election system, with no controls on conflicts of interest, and permitted obscene lobbying by private electronic voting companies (buddies of theirs) to state election officials.

When the exit polls showed a Kerry win, this strong evidence of fraud should have been available to the American public on election day, not six months later when it's old news. Instead, the news monopolies CHANGED the exit poll data, on everybody's TV screens on election day--"adjusted" this data to fit the official result (that Bush won)--and not only did they doctor this information, they then discounted and hid the true information for many months. In a free country, with a working 1st amendment, the press would have been the first line of redress. Instead, the news monopolies acted as propagandists for the Bush Cartel.

We already have laws and principles of government, and mechanisms of redress, that should have prevented this election from being stolen. But none of them is working properly, due to the inordinate influence, power and financial resources of the corporate cabal that is running our country.

To try to fix this with a federal ballot, etc., ignores the deeper wounds that have been inflicted on our democracy. I understand this myopia. It's truly difficult to grasp the magnitude of the crisis we are facing. But we need to overcome that myopia, and keep the full magnitude of the problem in front of us. The Bush Republicans DIDN'T WANT a fair election. They acted in every way possible to PREVENT a transparent election. They don't believe in democracy and good government. Their purpose is to DESTROY democracy and good government. And they are in a position of supreme power right now, and bent on taking one fascist measure after another, to entrench themselves, and make it impossible to cast them out.

No mechanical repair--such as a federal ballot--is going to fix this problem. A fascist coup can manipulate a federal ballot as well as a state one. And if your avenue of redress is blocked by Bush Cartel appointed judges, what have you gained?

TRANSPARENT elections are our only hope. Recountable, auditable elections--which can still be achieved with grass roots citizen efforts at the state/local level.

I'm also something of a states' rightest on this matter. If we separate federal and state ballots, and give more power to the federal government on federal elections, we will have lost the "balance of powers" check on federal fascists. The variety in state governments and even in voting methods can be a strength.

What if the Bush Cartel uses power over federal elections, not to adopt transparency measures, and to create redress for citizens, but rather to require paperless electronic voting run by BushCon companies on a national basis? This would be their most likely use of such powers, if recent history teaches us anything.

In fact, we face this very peril right now. This Bush Cartel majority in Congress could RIGHT NOW create a uniform federal election system that would end any hope of fair elections for president or Congress, ever again. Look what they're doing on the filibuster in the Senate--assaulting a time-honored tradition that is designed to protect dissent and insure some balance against executive tyranny! I would just as soon they keep their hands off election reform!

[LatePeriduct]

86. Your analysis requires several factors.

Every precinct in every state must be that far off, and the wrong voter areas had to have been sampled.

O.k so assuming its true, how I understand it is that the odds of that happening are almost smaller than water dropping in the same place twice.

But if I adjust your analysis, and just look at the common factor from what I have heard....it tells me that in all those different voter areas, a whole huge number of votes were purged.

Looking at the evidence of the EIRS automated reporting system, I have to put 2 and 2 together. Mickey Mouse didn't rig the exit polls. It was the result of purged voters all over the place which has the highest probable truth to it, so I'll just use occums razor.

Okay, so the states purged the voters. So what else is new?

[mgr]

87. I'm not making much sense of this post.

Your first paragraph captures a concern that I have, that in looking at the overall picture, national, we may have missed patterns at state exit polls that may allow for discriminating voting discrepancies. But I don't think that is where you are going.

As to water striking the same place twice, (are you sure you did not mean lightning?), it depends upon how precise your measurement is, and how accurate.

What common factor have you heard?

Looking at the EIRS database we should be able to evaluate which states are likeliest to have irregularities, and are where we should focus, recalculating the exit poll using precincts that do not exceed the WPE to where they may have mis sampled the voting population. There is a very common statistical method to apply this, and could be employed to combine precincts sampled by NEP and by LA Times. I think that many are awaiting the USC Votes to do it, and not taking it up on their own.

Purging of election rolls are not characteristic of all states. Those that are purged are unlikely to be sampled as having voted, if they never received the voting information in the first place.

I'm sorry but you are vague to the point where I don't know what you are applying Ockham's Razor to. You need to realize that Ockham's Razor comes into play with two reasonable arguments, you select the one with fewer syllogisms. It has nothing to do with consistency, validity, or truth. In case you were not aware, Ockham's Razor was specifically applied to prove that God created everything, as it is the simplest argument with the least number of syllogisms.

Mike

[LatePeriduct]

89. On the contrary. You make perfect sense of it.

"Purging of election rolls are not characteristic of all states. Those that are purged are unlikely to be sampled as having voted, if they never received the voting information in the first place."

Purging of the voters from the rolls does not mean they are not allowed to register, it means they are disqualified.

I'm guessing you figured out that each state has a central BOE right? That is where all their information is stored. Florida has already come out and admitted voters were purged out of the system.

It means quite alot of them were probably sampled, or else they wouldn't be recorded in the EIRS. Your analysis is showing skewed results, a straw analysis which is quickly debunked when you take all factors to consideration.

If they really did exit-poll the wrong voter areas everywhere, there would be a front page story already documenting everything wouldn't there. Instead there is a whole network of stories proving vote supression and vote purging, and touch-screen mishandling. So where is all the statistician backed reports of the exit polls being totally wrong?

Sorry to say, but your idea of ockum's razor is exactly that. You agree with your own analysis, therefore you have to agree with TIA, USCountVotes, and all the other experts including Steven Freeman. The exit poll results were not mishandled enough to produce thousands of mistaken voters.