IF YOU BELIEVE THESE 5 IRREFUTABLE FACTS, THEN YOU MUST ALSO BELIEVE...

[TruthIsAll]

THAT KERRY WON THE ELECTION BY AT LEAST 3 MILLION VOTES...

It comes down to these simple facts:

1. The Final Exit Poll (13660) promulgates an impossibility in
using the 43% Bush/37% Gore "Voted in 2000"
demographic weighting.

2. Even if ALL 50.456 million Bush 2000 voters were still
alive and returned to vote, his maximum 2000 voter share is
just 41%, as stated in the "pristine", unadjusted
Preliminary Exit Poll of 13047. 

3. Approximately 3.5% voters have died since 2000, so the Bush
2000 voter share is now reduced to a MAXIMUM OF 39.8%,
assuming 100% turnout. Of course, it MUST BE EVEN LESS THAN
39.8%, because an unknown number of Bush 2000 voters stayed
home this time. But we will assume 100% turned out - to be
conservative.

4. We KNOW that at least 17% were NEW voters, because we KNOW
the change in total voter turnout from 2000 (104.777) to 2004
(122.26).

5. We KNOW that 3.17% voted for Nader/Other.

Therefore, it follows that if the Bush 2000 share is 40%, we
can calculate the Gore 2000 share of 40%: 

    Gore = 100% - Bush - NEW - Nader/other
     40% = 100% - 40% - 17% - 3%

CONCLUSION:

I.
Using the Final Election Poll (13660) stats and the 40/40
split, Kerry is an easy winner by 3.5 million votes. The Final
Exit Poll weights ADJUSTED the Preliminary Exit Poll weights
in order to MATCH the Recorded Vote.

HERE'S THE CALCULATION, WITH THE IMPOSSIBLE BUSH/GORE 43/37
MIX CHANGED TO 39.8/40.3%. 

All other Final Exit Poll numbers are unchanged:

VOTED
2000	Mix	Bush	Kerry	Nader
No	17.3%	45%	54%	1%
Gore	40.3%	10%	90%	0%
Bush	39.8%	90%	9%	1%
Other	2.6%	21%	71%	8%

	100%	48.18%	51.04%	0.78%
	122.26	58.91	62.40	0.95


II.
Using the Preliminary ("pristine") Election Poll
(13047) stats and the 40/40 split, Kerry is a landslide winner
by 7 million votes.


VOTED
2000	Mix	Bush	Kerry	Nader
No	17.3%	41%	57%	2%
Gore	40.3%	8%	92%	0%
Bush	39.8%	90%	9%	1%
Other	2.6%	21%	71%	8%
	100%	46.68%	52.37%	0.95%
	122.26	57.07	64.02	1.16


The detailed mathematical analysis is explained here:
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x351674

 
 

[Wickerman]

1. I hope you are putting all this stat info together in a package

like a book or something. Tie it all together. Your posts are always enlightening and would be a fantastic read.

[Disfronted]

2. I don't follow

I still don't understand why the statement:

"Both the final and unadjusted exit polls were wrong"

Is not a possible explanation.

If the preliminary exit poll is wrong, and it's numbers are used to extrapolate the final (adjusted) exit poll, then the fact the final poll is wrong does not mean the unadjusted poll is right. It could be the unadjusted poll is wrong and that is why the final poll is wrong.

[TruthIsAll]

3. You are making several errors.

The Final poll was wrong. Period.
It changed a possible 41%/39% split to an impossible 43%/37%.

The Preliminary 13047 poll result is a feasible one, although it is still inflated for Bush.

The Final 13660 is totally impossible.

1) The 41%/39% split in the Preliminary poll is extremely unlikely, but has a very small probability, if one assumes that only a handful of Bush voters died - and that all those still alive turned out to vote.

THAT'S WHY IT'S CLOSER TO BEING "RIGHT".

2) The 43%/37% split in the Final poll is impossible.

THAT'S WHY IT'S ABSOLUTELY WRONG.

[Disfronted]

5. You miss the point

Of course the final poll is wrong in breaking down the votes. I am not disagreeing.

That does not mean that the preliminary poll was right. In fact, given that the final poll is extrapolated from the preliminary poll, one conclusion you can draw is that the final poll is wrong (as you have proven) because the prelimary poll is also wrong.

[TruthIsAll]

6. YOU MISS THE POINT. FOLLOW CLOSE.

THE FINAL EXIT POLL WAS MATCHED TO THE RECORDED VOTE.
IT WAS NOT EXTRAPOLATED FROM THE PRELIMINARY POLL AT ALL.
TO MATCH THE RECORDED VOTE THE WEIGHTS HAD TO BE ADJUSTED.
THE WEIGHTS HAD TO BE ADJUSTED FROM A POSSIBLE 41%/39% TO AN IMPOSSIBLE RESULT (43%/37%).

THAT IS WHY THE FINAL EXIT POLL IS FLAT OUT WRONG.
THAT IS WHY THE PRELIMINARY EXIT POLL IS CLOSER TO THE TRUTH.
NO ONE SAID IT WAS EXACTLY RIGHT.

BUT NO ONE CAN SAY IT WAS WRONG.
YET WE CAN SAY THAT THE FINAL POLL IS ABSOLUTELY WRONG.

YOU NEED TO READ THE FULL POST AGAIN.

SIT BACK, HAVE A COLD DRINK AND THINK.

[Disfronted]

7. I'm having a drink, but...

Where did I say the final poll was right? Answer that.

I'm missing where the unadjusted poll has to be right because the final is wrong.

They can both be wrong.

Edit - To clarify, unless you can prove the unadjusted poll was right, your analysis is just speculation - maybe even probable, but not PROOF.

[TruthIsAll]

8. Let's try it again.

We agree.
The Final Exit Poll is wrong, because 43/37 is impossible.
The Final Exit Poll was forced to an impossible 43/37, in order to match the recorded vote.

What does that say to you about the recorded vote?

In other words, if the preliminary poll was possibly correct (41/39) why change it to an impossible 43/37?

The question answers itself.
IT HAD TO BE CHANGED FROM 41% TO 43% TO MATCH THE RECORDED VOTE.

JUST ANSWER THAT QUESTION, PLEASE.

[Brotherjohn]

10. No no no no.... It says absolutely NOTHING about the recorded vote.

Sorry, but it doesn't.

Unless, that is, you know for a fact that every other number in the exit poll is a 100% accurate reflection of the actual vote count, voter preference, etc. And you don't. No one does.

There are an infinite number of ways the numbers could have been tinkered with to match the recorded vote. They simply chose a way which made no sense based on the facts (maximum number of possible 2000 Bush voters voting).

They could have chosen any number of possible combination of percentages (both within the realm of possibility and without) and still ended up with the same net totals for Bush and for Kerry.

Let's try this:

4 reds + 4 blues + 4 greens = 12 total

1 red + 8 blues + 3 greens also = 12 total

Proving "4 reds" is impossible doesn't prove 12 total is also impossible.

[Disfronted]

11. I am still working on the drink...

Who you voted for in 2000 and who you voted for in 2004 are two seperate statistics. It is plausible that the unadjusted exit polls understated support for Bush while accurately reporting the breakdown of the voters by "who they voted for in 2000" or WTVFI2. In this scenario, since the WTVFI2 information was already correct, adjusting the exit poll to match the actual vote (and correct the understatement of Bush support) results in the creation of an error in the WTVFI2.

I'm not saying that is what happened (I really do not know) but it is plausible. Either way, what you posted is not PROOF.

[Brotherjohn]

9. Hate to tangle again, Truth, but I still see major problems.

You convinced me yesterday, and you are inarguably right mathematically, that the 43% is an impossibly high number of 2000 Bush voters.

That also showed me that there are BIG problems with this exit poll.

It does not, however, follow that Bush's election by the numbers presented in the final "official" tally is impossible.

You are picking and choosing numbers in the exit poll to believe as gospel while simultaneously proving that others cannot be true.

You state that "Using the Final Election Poll (13660) stats and the 40/40 split, Kerry is an easy winner by 3.5 million votes." and "Using the Preliminary ("pristine") Election Poll (13047) stats and the 40/40 split, Kerry is a landslide winner by 7 million votes."

Yet ALL of those election poll stats you are using (peliminary and final) may be innacurate. In fact, your analysis casts doubt upon them.You prove that some of the numbers can't be true and recalculate maximum possible numbers, yet you don't prove that the other numbers you trust from the exit poll which feed into your calculations are true. So therefore you cannot say your final calculations are true.

While you can be fairly sure that around 17% more voters voted this time, what makes you so sure they voted 41-57 Bush-Kerry? They could have voted in any possible breakdown? 50-50 Bush-Kerry. Or any of the other breakdowns could be wrong. The only evidence we have of these breakdowns is the exit polling, which you have shown is faulty.

I personally think it's likely Kerry actually got more votes, but the exit polls were tinkered with to math the "official" tally. I'm just saying your analysis doesn't prove this (using any realistic definition of the word I can see). I don't think it proves anything at al about the votecount, just that the exit polling cannot possibly be an accurate representation of the vote count.

If you have proven (and you have) that the total voting electorate could NOT have consisted of 43% Bush 2000 voters, why are you so willing to believe ANY other aspects of this exit poll. It's clear from your analysis that there are major faults with this exit poll, and that convinces me that they probably tinkered with it to get their final results to match the official tally (and did a piss-poor job of it, at that).

But that says nothing whatsoever about the actual election results. It merely proves the exit-polling was faulty. Yet you still choose to believe every aspect of that led into your calculation.

[TruthIsAll]

12. Ok, brother, here we go again. Let's talk probabilities.

The Margin of Error is 1%.

Even assuming a 50/50 split in New Voters, Kerry still wins. That would mean a deviation of 4%, from 54% to 50%.

Now we know the earlier exit polls had Kerry a 57-41% and 59-39% winner of new voters. And we know that the 41/39% split was possible, even if extremely unlikely.

But let's use the IMPOSSIBLE Exit Poll of 13660 and make it even more so by reducing Kerry to 50% of new voters.

The odds of this 4% deviation:
= 1 - NORMDIST(0.54,0.5,0.01/1.96,TRUE)
= 2.22045E-15
or 1 in 450,359,962,737,050

Would a prosecutor call that PROOF?
Would a jury call that PROOF?

Would a jury convict given those odds in a civil case?

In a prior post, you said that I have proved ONE number was wrong, the Bush 43%.
It had to be lower.

You were wrong.
TWO numbers were wrong.
Don't forget the Kerry 37%.
It had to be higher.

It is obvious, beyond a reasonable doubt, that you and I have a different view on just what constitutes proof.

[Disfronted]

14. That is assuming a random sample

If the sample was not truly random, it is plausible that the exit poll would be off to that degree.

Examination of past exit polls suggest the sample very well may not be nearly as random as you are assuming.

[TruthIsAll]

15. You are barely breathing now. It was a randomly-selected sample.

Read the Mitofsky report.
1% MOE
13047 respondents.

You are left to parroting desperate Naysaer talking points which were refuted a long time ago.

[Brotherjohn]

17. You are arguing that they changed their numbers to perpetuate...

... a fraudulent Bush victory, yet you are simultaneously citing their methodology as trustworthy?!?

[TruthIsAll]

20. You are reaching.

Your valid issues from the other day have degenerated into circular arguments. I do not question their methodology. No one does. I question the results.

The Preliminary Exit Poll was pristine. I do not question that.

I question the change from 41 to 43% in the Final. We agree it is impossible, yet they had to force a match to the recorded vote.

But as uscountvotes.org has stated, Mitofsky-Edison has never considered the possibility that the vote count was fraudulent, only that Bush Voters were more reluctant to speak to exit pollsters. That has been disproved.

My analysis speaks for itself. No amount of spin on your part can cloud the FACTS. The only question is your definition of proof as compared to mine.

Proof beyond a reasonable doubt (99%) is not required in a civil case. You only need to believe in the preponderance of evidence (50%). Here probabilities are MUCH LOWER than 1 in a million.

Have you reached a verdict?
We have, your honor.

[Brotherjohn]

24. "I do not question their methodology. No one does." NO ONE DOES?!?

I think many many people do, including myself, Disfronted, and yes, uscountvotes.org. Your problem is precisely that you believe (emphasize "believe") that the preliminary exit polling is "pristine". You have still not answered why, and your very analysis opens up their trustworthiness to question.

Remember, YOU are the one who put a supposed mathematical proof out there. All you are getting back is contructive criticism on that proof. Some of us simply feel you are assuming many things with little basis. Your answers have not satisfied us.

[Disfronted]

18. Of course Mitofsky attempted to create a random sample

That doesn't PROVE it was truly random.

Be a little more specific.

[TruthIsAll]

22. Ok

Only that this is my final reply to you.

You are a classic naysayer.
Pure circularity.
There is no point in continuing to spar with you.

Bye.

[Disfronted]

23. Glad you see it my way

So I guess you agree with me then - we really have no proof that the unadjusted exit poll was accurate.

[Brotherjohn]

16. The MOE is 1% only if the sample was actually a random, representative...

... sample. What you have in fact done is actually to cast doubt on THAT.

Yes, based on that assumption, you're talking astronomically high odds for the numbers to work.

The US Count Votes study released yesterday said as much. Which is why they came to the following conclusion: "Two alternatives remain. Either something was wrong with the exit polling, or something was wrong with the vote count."

That something is wrong with one does not necessarily mean something was wrong with the other, EVEN if they yield the same total votes for Bush and Kerry.

Although the proven exit-polling problems alone merit an investigation of the election.

[Brotherjohn]

19. You are PROVING that the exit poll is faulty.

You are not PROVING anything about the reported results (by any definition whatsoever).

Proving the Bush 43% is wrong is one and the same as proving the Kerry 37% is wrong. The two are not independent of each other. You proved one number is wrong.

[TruthIsAll]

21. You are getting careless, Brother John.

For Bush 43 to be too high, Gore 37 had to be too low.
Do we agree?

Nader = 3%
Do we agree?

New voters = 17%
Do we agree?

Then
Gore = 100 - Bush - 3 - 17.
Do we agree?

Then
If Bush = 43, Gore = 37 (impossible).
If Bush = 41, Gore = 39 (very unlikely).
If Bush = 40, Gore = 40 (possible).
If Bush = 39, Gore = 41 (possible).
Do we agree?

But, now you will surely say, maybe NEW = 18%
Okay, but..

NEW = 122.26 - 104.77 + 3.67 = 21.16
(2000 voter deaths =.035*104.77 = 3.67)
Do we agree?

And 21.16/122.26 = 17.3%
Do we agree?

That's a pretty HARD number
Do we agree?



But we all know who won the majority of NEW voters, don't we?
Do we agree?

Brother, come back to the real world.
Return back from your sojourn into a bizzaro universe.
Where red is blue and blue is red.

[Brotherjohn]

25. You are pretending to use hard numbers to make a mathematically...

... proven case, yet you are saying things like:

"But we all know who won the majority of NEW voters, don't we?"

Well, THERE's a mathematical proof for you!

You question completely the breakdown of Bush 2000 voters, and rightfully so, based on the possibilities; yet you refuse to question for a millisecond the breakdown of voter preference within each group.

Besides, again, if their sampling was NOT random (although you religiously believe they somehow used perfectly representative sampling), they could have had any number in the world for Bush 2000 voters. It wouldn't be possible for the whole sample set of 122M, but it certainly would be possible for a poll. Perhaps improbable, but less and less so if their sampling was less representative.

Look, we're just butting heads here to no effect. I too think Kerry won a majority of new voters, and I too think the preliminary Edison-Mitofsky results were probably much nearer to the truth than the finals. I also think the numbers were changed to match the "official" tally, whether that tally was accurate or not. I think it poses dire questions about the election as a whole, given that exit-polling has traditionally been reliable.

But I will maintain until my grave that nothing you have shown here PROVES Kerry won the election. And YOU were the one who put several posts out there pretending to MATHEMATICALLY PROVE that Bush lost. Sorry, you just haven't done that. I wish you had.

If you had, don't you think you'd be getting a LOT more replies? We're a smart bunch here at DU. We can all crunch numbers.

[TruthIsAll]

29. A non-denial denial. Face it, John. YOU BELIEVE. ALL THE REST IS BS.

You have just said you agree with everything I said.
Except its not proof.
But I never said that, did I John?
I said: If you believe these five facts... then you also must believe...

Do you believe Bush stole it, John?
Apparently you do.

Let's see what you have just said you believe:

1.All my facts regarding the "How voted in 2000" demographic are correct.

2.The Final Election poll was clearly bogus.

3.The earlier polls were close to the truth.

4.The final poll numbers were matched to the "official tally", whether or not they made sense.

5.Kerry won the popular vote.

6.Kerry won the EV

7.The discrepancies pose dire questions about the election.

etc etc

John, do you for one moment think about why you believe all this?

Here's one possible reason:
Because I crunched the numbers for you.
You never did them yourself.

Did you read Freeman or Baiman and/or the rest?
Maybe that's another reason.

John, welcome to DU.

Oh, don't worry. Other DUers will be back on their computers tonight responding (dare I say favorably?) to what I have presented here.

[NationalEnquirer]

4. Slam Dunk my friend! n/t

[keepthemhonest]

13. TruthIsAll,

are you going to the Nashville conference and showing everyone all of your statistical analyses you have done since the nov 02 election?
Hope to see you there.

[Brotherjohn]

26. Back to your original post... Yes, we KNOW the 5 original facts.

They are for the most part irrefutable (some assumptions about number of deaths and such, but reasonable ones).

You then make a big leap, and that big leap is as we discussed above here in other posts.

We DON'T "KNOW" that 57% new voters voted for Kerry. We DON'T "KNOW" that 71% of the "other" voted for Kerry". The preference of the 2000 voters and new voters in the 2004 elections is not KNOWN. You simply choose to believe the breakdowns as in the exit polls (final for 3.5M win, prelim for landslide of 7M). In short, we DON'T "KNOW" that the rest of the exit poll data is reliable.

Your think your logic is this:
KNOWN FACT (X5)... IRREFUTABLE CONCLUSION.

In reality, it is this:
KNOWN FACT (X5)... (many unknown facts which need to be trusted to reach the...) IRREFUTABLE CONCLUSION.

The "many unknown facts" are how each group voted in 2004. You assume this is accurate to 1%MOE, and even trusting the final poll results, this yields a Kerry win.

First, of course there's the issue of the sample not being random (which I won't be-labor here again).

But ALSO, as you pointed out yesterday, 1% MOE for a sample of around 10,000.

But the smaller samples within the 13000 (3% "other"; 17% new, etc) all have larger MOEs AS TO THE QUESTION of who THEY voted for in 2004. These are smaller polls within the bigger poll.

All told, these sources of additional error can certainly result in deviations from your assumed irrefutable conclusion. But even if they couldn't, all it would prove would be that E-M have a strong vested interest in making their poll look accurate (to the point where they'd fudge numbers)... whether that was to avoid embarrassment or to be yield validity to fraudulent election results.

Again, YOU were the one who "threw down the gauntlet", as it were... saying that "if we believe these 5 facts... then we MUST believe". Sorry, I don't. I don't think it's the smoking gun you hoped it would be (I wish it were). That's not to say it isn't TRUE. But the logic does not follow as you say it does. And you have set the terms of it as a logical argument.

[TruthIsAll]

27. Sorry, I believe. Too bad you don't . What DO you believe?

I laid down the gauntlet? Very dramatic prose, I must say.
Especially after my dry recitation of the facts. Kind of makes
it sound sinister.

Well, thanks for kicking it up, anyway.

No, I'm not laying down the gauntlet. Just laying out the
facts and replacing an impossible result with a very plausible
scenario.

Now all THREE exit polls show that Kerry won, since we have
removed the 43%/37% cancerous mix as a possibility. This
raises many new questions. First and foremost: Why?

I'm sure that you and many others have in the past few days
learned some interesting new facts which you were not aware
of. I've never seen the IMPOSSIBLE 43%/37% demographic
questioned by anyone. Have you? Don't you find that almost as
odd as the numbers themselves?

IT IS THE ONLY DEMOGRAPHIC WEIGHT WHICH CAN BE FACTUALLY
VERIFIED.
AND IT TURNS OUT TO BE AN IMPOSSIBILITY.

BUT YOU CHOOSE TO FOCUS ON THE ONES WHICH CANNOT BE FACTUALLY
VERIFIED, WELL AT LEAST UNTIL EDISON-MITOFSKY RELEASE THE RAW
DATA. WONDER WHAT WE WILL FIND UNDER THAT ROCK IF THEY DO?

It's been five months since the election. All the research,
all the PhDs, all the probabilities, all the anomalies, all
the naysayers - and the 43% never came up. Until now. And it
seems to have ruffled some new feathers.

To me it's worthy of the "smoking gun" tag. To you,
it's just another bad tooth which must be extracted.

So, what are you left to question:

1. Surely not the mix percentages. I will run other
combimations for you, if you wish. There is not much wiggle
room there. 
But let's keep it realistic.

2. The Kerry/Bush new voter split: The polls, from start to
finish, give Kerry the majority from 59-39 to 54-45. Not even
close. To take it down to 50/50 is really stretching
credulity, don't you think? Do you believe the original exit
polls are off by 7-9%? But, OK, I will run some sensitivities
there as well, if you wish. 
But let's keep it realistic.

3. The Bush share of the Dem vote. It was 8%, they raised it
to 10% in the final. But apparently, it's still not enough.
How much do you need, John? Fourteen percent? I'll run the
scenarios. 
But let's keep it realistic.

Meanwhile, we will stick with this:

Final Exit Poll (13660 respondents)
Changed from an impossible 43/37 to a very possible 40/40.

VOTED
2000	Mix	Bush	Kerry	Nader
No	17.3%	45%	54%	1%
Gore	40.3%	10%	90%	0%
Bush	39.8%	90%	9%	1%
Other	2.6%	21%	71%	8%

	100%	48.18%	51.04%	0.78%
	122.26	58.91	62.40	0.95

[Brotherjohn]

30. Look, let's make up. I appreciate your work. We agree on much.

Look, let's make up. I appreciate your work. We agree on much.

I used the term "laid down the gauntlet" b/c I felt you were taking very personally what was honest constuctive criticism. I don't think it was kicking things up... I think it was accurate. I was merely trying to make the point that when you begin posts (in all CAPS mind you) "IF YOU BELIEVE THESE 5 IRREFUTABLE FACTS, THEN YOU MUST ALSO BELIEVE...", that simply DOES come off as "throwing down the gauntlet", and you shouldn't take replies such as those you've gotten so personally. THAT KIND OF INTRO begs for a give-and-take, constructive criticism of whether or not your thesis does in fact prove what you say. They are merely attempts at picking apart the logic, a scientific exercise... Believe me, I'd like to fail at that and be convinced you have smoking-gun proof of a Kerry win.

Your initial language, and that throughout most of your posts, was much more unequivocal than your last one: "Just laying out the facts and replacing an impossible result with a very plausible scenario."

That is a much more accurate description of what you are doing. It is more "plausible". I agree on that.

You say the 43% "IS THE ONLY DEMOGRAPHIC WEIGHT WHICH CAN BE FACTUALLY VERIFIED. AND IT TURNS OUT TO BE AN IMPOSSIBILITY."

I agree it is problematic. But the thing is, it is only an impossibility in the entire sample universe. It is not an impossibility in a poll of the universe, especially with error, and "weighting", etc.

"BUT YOU CHOOSE TO FOCUS ON THE ONES WHICH CANNOT BE FACTUALLY VERIFIED".
I don't focus on those perse, I'm merely pointing out that you are assuming they are true in making your proof (even though you just said "they cannot factually be verified").

As far as "what I believe", I've alluded to as much in prior posts. I believe Kerry won. I believe the initial exit-poll results were more in line with actual numbers, and that the "weighting" was bogus (why the hell is it even tolerated?). I tend to believe E-M did their best to rationalize their weighting how their initial numbers were so different, for whatever reason (probably to avoid embarrassemnt to themselves and a constitutional crisis). They probably believe their results are accurate (and why not, they MADE them so). They probably DIDN'T ever consider the possibility of fraud on the part of Bush.

I believe the vote total accounts for probably millions of votes lost or improperly recorded by election officials who, lets face it, are in principle the same people who were protesting outside the hospice against the "murder" of Terry Schiavo. I wouldn't put anything beyong them... least of all to drop a few ballots in the trash or inadvertently flip a switch or two. I also believe the total vote reflects millions lost through voter intimidation which, despite their pot calling the kettle black rhetoric, was MUCH worse on the part of Republicans. Stationing (unqualified and biased) "observers" at polling places to "check their credentials"... yeah right! That this was ever allowed should have told us right there the election was lost.

I believe work like what you are doing and what USVoteCounts are doing will hopefully cast more light on problems with exit-polling and the lack of a paper trail in voting in general.

[TruthIsAll]

31. I agree with the spirit of your post. It's really just semantics, emphasis

I am a passionate believer in the truth.
I will never say anything which I cannot back up with facts.
My thread titles are meant to attract.
Yet they are factual.

I never exaggerate the probabilities.
If I claim it's 1 in 300 trillion, it's because that's what the function returned.

I never use data or calculations which cannot be verified. I clearly display the assumptions, formulas and methodology.

Finally I projected Kerry the 2-party vote winner by 51.8%-48.2% in my election model (337 EV) using published pre-election state and national polls with a 99.5% probability.

I was damn close to being right on the money.
http://www.geocities.com/electionmodel/

[Nederland]

28. Brotherjohn and Disfronted

Best of luck arguing with TIA, and I mean that seriously. Many of us here have been trying ever since the election to convince TIA that you cannot simply assume that the exit poll sample is a perfectly random representation of the total voting population. Despite months of trying and hundreds of posts, we failed. Best of luck though, because it seems to me that if we could just get TIA to acknowledge a few fairly simple statistical principles s/he would be a powerful and passionate ally in determining what really happened in this election. Cheers.

FYI, there probably won't be a response to this because I'm on TIA's ignore list.

[kster]

35. Kick ...................n/t

[merh]

32. TruthisAll - have you seen this before?

Have you seen this explanation before?

Associated Press

The exit poll of Mississippi voters was conducted for the National Election Pool - The Associated Press, ABC, CBS, CNN, Fox and NBC - by Edison Media Research and Mitofsky International.

The survey was conducted at a randomly selected probability sample of 15 precincts around Mississippi.

As people left the voting booths, Edison/Mitofsky interviewers asked them to fill out a confidential paper questionnaire prepared by NEP representatives. The interviewers selected voters at a set interval - such as every fifth person - so that each participant had an equal chance of being picked.

The results were adjusted to reflect the different probabilities of selecting a sample precinct and people attending each, as well as by the observed sex, race and estimated age of voters who refused to participate.

(snip)

http://www.blackmississippi.com/indexarticle.html?page=main&id=-1&grid=&fid=941

[TruthIsAll]

33. Edison/Mitofsky say Randomly Selected, but the naysays say...

No way.

Dittoheads.

[merh]

34. Did they publish this explanation of the Mississippi polls because

the exit polls showed Kerry received morevotes than he was "awarded"? Is that your take on this?

:shrug:

[TruthIsAll]

36. Well...if you are referring to adjustments made to the preliminary

13047 poll, then I refer you to the "How Voted in 2000" demographic.

As I have demonstrated, the 43%/37% mix is physically impossible, since it is in conflict with 2000 vote and death rate statistics, so...draw your own conclusions.

Since the weights are verifiable (and completely bogus, not even close), what does this hint about the unverifiable "adjustments" made to the other preliminary exit poll demographic weightings?

Can you have confidence in any of them? In fact, since ALL matched to the vote, the answer is obvious.

The preliminary poll was closer to the truth - and Kerry won it.
The truth must be that Kerry did better than the preliminary poll (41%/39% 2000 voter mix), once factual stats (actual 2000 votes, death rate, turnout ) are used to determine a plausible mix.

My best estimate is 40% Gore/ 39% Bush - which makes it a Kerry blowout.

[merh]

37. Definitely a blowout!

Thanks for the info. :hi: