A TIA CHALLENGE: REFUTE THESE ODDS!

[TruthIsAll]

This is for the Mystery Pollster and other interested parties.
Go through the analysis of probabilities in this post based on state exit poll Z-Scores.

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x361736

Go here for state Z-scores and corresponding probabilities:
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x361617


In a nutshell, the data shows that the Z-score =1 standard deviation was exceeded by Bush in 35 states. For Kerry, Z>1 in FIVE states.

For Bush, the Z-score of 2.20 was exceeded in 16 states.
For Kerry: NONE

I have computed probabilities for both Z-score thresholds.

The standard deviations are based on the formula:

StDev = sqrt (KP*BP/n)
KP = Kerry exit poll percentage.
BP = 1-KP = Bush exit poll percentage
n = state exit poll sample size

Calculating the probabilities for each occurrence:
Prob (Bush Z >1 in 35 states): 1/4500 trillion
Prob (Kerry Z >1 in 5 states): 91%

To calculate the probability of the Z-score, we use the Excel
function:
Prob P(Z) = 1-NORMSDIST(Z),
where Z = the number of standard deviations, as above.

To calculate the probability of the Z-score exceeding 1 in at
least N states, we use the Excel function:
Prob PB(N>Z) =1-BINOMDIST(N-1,50,P,TRUE)

Favor N; Z; P(Z); PB(N)
Bush 35; 1.0; 1/6; 1 in 4500 trillion
Kerry 5; 1.0; 1/6; 90.9%

Probabilities for other Z-score levels:
Favor N; Z; P(Z); PB(N)
Bush 16; 2.20; 1 in 72; 1 in 160,000 trillion
Bush 10; 2.60; 1 in 215; 1 in 23 trillion
Bush 4; 3.35; 1 in 2475; 1 in 165 million
Bush 3; 3.75; 1 in 11307; 1 in 74 million
Bush 2; 4.10; 1 in 48382; 1 in 1.9 million


If you can refute these results, please do so.
I look forward to all replies, whether at DU or elsewhere.

[vince3]

1. Kick for TIA

[LatePeriduct]

2. I'll try my best....

(Puts on James Baker special hat)

The fact is they are making it up as they go along, these exit poll takers are oversampling every area in the united states....

Its because the cencus takers of the government are really incompetent....Nobody purged any voters...Its just that the exit polls were way off....Sure there was plenty of voter fraud and thats why Bush won.....But voters are to blame for that..

[Botany]

3. Odds are the election was stolen ......

..... exit polls right when people are voting

..... exit polls wrong after the people stopped voting

Got that?

I watched on secured computers @ a Kerry HQ in downtown Columbus, Ohio
on election night and saw the * vote come in. It did not matter what state
or what part of the country or what group was voting the #s were 70 to 90%
for * in the returns.

In Iowa w/ 30% of the vote in before 11/2 Kerry had a 13 point lead but he lost Iowa?
So better than 3 out of the 4 remaining 70% of the voters who voted went to *?

Iowa has one of the highest % of College degrees per population in the U.S.
Having a College degree was one of the factors in people voting for Kerry.

Keep it up TIA!

[bleever]

4. Kick!

[DELUSIONAL]

5. Kick

[TruthIsAll]

6. Still waiting...

.

[kiwi_expat]

7. OK. I'll bite.

There is no way to refute the numbers. There is no way that the sampling errors could be random.

Either there was fraud and/or there really is a Reluctant Bush Responder phenomenon. My guess is that both are true, at least in states like Ohio, where the weather was atrocious and there was a 100ft. limit imposed on the NEP interviewers for at least part of the day.

The Cincinnati 4-M observers said that a lot of "repub types" ignored the NEP interviewer. I can easily imagine Republican businessmen ignoring what looks like a campaigner (hard to read the NEP badge after 4pm) who is standing with actual campaigners, in the pouring rain. I think the repub businessman would be focused on getting home to the nice dinner prepared by the little lady.

[TruthIsAll]

8. Keep drinking the Kool-aid. You forget uscountvotes.org showed that

according to Mitofsky's own data, it was just the opposite. And you forget the National Exit Poll How voted in 2000 Demographic. You know Gore got more votes than Bush in 2000, right? The mpossible 43%/Bush/37% Gore weights soil the Final 13660 completely - and you still don't see it?

Quite frankly, Kiwi, you ignore everything plausible and accept the impossibilities. You just can't get across the river of Lies and Deception. All this analysis and all your good work - to no avail.

What's the point, at this point? If you still want to believe the BS, go right ahead.

Very sad.

[kiwi_expat]

11. How would you like to help us find out what actually happened in Ohio?

So far we have been focusing on NEP precinct Cincinnati 4-M,
because there is such a large discrepancy between the NEP raw
numbers and the canvass results:
               Bush Kerry %Kerry
  Cincy4M vote  312    211   40%
  polling place 946    808   46%
  NEP Raw        10     21   68%
  NEP weighted   14.4   22.6 61%


Liam_laddie is going to do a manual recount of Cincy4M.  We
certainly hope he will find evidence of fraud.  However we
have been discouraged by the reports from the observers at 4M.
 It is possible that the Cincy4M NEP data is worthless.   The
recount will tell us the answer.

If we can't rely on the NEP data to point us to precincts most
likely to have fraud, where do we go next?  I'm thinking that
a spreadsheet of all the 1000+ Hamilton County precincts,
sorted by %Bush win, might point us in the right direction.  
Do you have any other suggestions?

How would you like to make us a "%Bush win"
spreadsheet based on the official Hamilton County results?  
It would be  very helpful. 
http://www.hamilton-co.org/boe/inputdata/Electionsresults/Archived/G04OFFCANVASS.xls


p.s.  It is not very nice to solicit comments and then accuse
someone, who responds honestly, to be "drinking the
koolaid".   I just want to find out what actually
happened in Ohio.   As we all do.

[TruthIsAll]

12. Sorry, but your comment about RBR was very disappointing.

After RBR has been thoroughly refuted by analyzing Mitofsky's own data, and the LACK OF ANY EVIDENCE WHATSOEVER TO JUSTIFY IT, did you really expect that I could just accept that RBR excuse of yours.

To me, that RBR comment is very telling. It's a total mystery how someone as knowledgable as you are can actually float RBR as a serious response to my post.

You want to find out what happened in Ohio? Great.

But first take a look at that Reluctant Bush Responder Blackwell.

[MontageOfFreedom]

13. Or, there is one alternative explanation which makes a bit of sense.

And I only challenge you with it, because you asked and because I would like it refuted.

Somehow, in the state exit polls also, the exit poll interviewers interviewed the wrong voting demographics in every single area, in almost all of the areas of the cities.

This was because of some unforseen phenomenon, that all of the many thousands of extra special Bush voters came from evangelical churches. Therefore none of them had any time to consider taking the exit poll interview, when in this context religion is priority.

[TruthIsAll]

15. You mean they go to church services on Tuesday?

Oh, yes, that's it.

Sunday is golf day.

[MontageOfFreedom]

16. Point taken. It borders on the extremely insane.

There is absolutely no conceivable way, that many voters appeared out of a magician's magical hat. Not even Jesus has that kind of power.

Although, a slight of hand and partisan BOEs rigging the election sure does.

[kiwi_expat]

14. What is your guess re. Cincy4M NEP/canvass difference?

What do you guess happened at Cincy4M?
................ Bush..Kerry %Kerry
Cincy4M vote... 312 211 40%
polling place..... 946 808 46%
NEP Raw............. 10 21 68%
NEP weighted. 14.4 22.6 61%

I certainly hope there was enough fraud to account for the entire raw to polling-place 44% spread. If so, we will surely find it in the recount.

Do you not entertain the possibility that the Cincy4M NEP data is worthless? And, given the weather and distance factors, that much of the Ohio NEP data might be suspect?

My RBR PARTIAL-explanation was based on the Cincy4M observers comments. The observers were not Republicans.

I am still hopeful that we will discover fraud by Ohio recounts. I'm just not confident that the NEP data will point us to it.

I am serious about the Hamilton County "%Bush win" spreadsheet. Could you help us, please?

Cheers.

[mgr]

20. 4M does not mirror all precincts present

You will have to correct the weightings to match the demographic of all the precincts at the polling place. 4M appears to be more republican in composition than the others, and would exacerbate the weighting given by NEP (Don't assume NEP adjusted, its only one precinct!)

Send me the Hamilton County spreadsheet, I believe Bill has my e-mail, let me see what is up.

Thank you, Mike

[kiwi_expat]

22. Thanks, Mike.

I have PMed you some data and the Hamilton County spreadsheet link.

In addition to sampling error and fraud, several other factors might help explain the difference between the Ohio NEP data and the canvass results: under/over votes (some of which may be fraud), rejected provisional ballots (Hamilton County rejected provisional ballots cast at the wrong table at a polling place), and absentee ballots.

NEP did not run telephone surveys in Ohio, so the absentee voters were not interviewed. There were 100 absentee voters in Cincy4M. That is 19% of the vote.


Thanks, again, Mike.

[kiwi_expat]

17. "Reluctant Bush Responder Blackwell" is responsible for 100ft. limit.

If the Ohio NEP raw data is suspect, it is due in great part to Blackwell.

He is also responsible for much of the Ohio voter suppression.

Maybe Blackwell is responsible for Ohio tabulator fraud, as well. I sure hope so - and I hope we can prove his involvement.

[kiwi_expat]

23. TIA, can we please discuss this calmly?

I am very interested in discussing this further.   First of
all let me state very clearly: I DO NOT THINK BUSH WON OHIO.

I think we also both agree that there is legitimate sample
error in precinct samples.   Mitofsky assigns a 6% MoE to very
small samples.   THIS DOES NOT MEAN THAT FRAUD DID NOT OCCURR
ALSO!!

****The question is, is it possible that the legitimate sample
error could tend to favour one candidate?****

Based on the Cincy4M observers comments, and based on my own
past experience as a part-time market researcher interviewer,
I think there are certain types of people who are more
receptive to being approached by a stranger and being asked to
answer questions.

We have very few NEP:canvass comparisons to examine.  One set
we do have is for Hamilton County, Ohio  (please see below).  
Notice that the largest discrepancies are at the polling
places that had the smallest samples.  I think it is possible
that the interviewers at those polling places were less
experienced/capable and that they were easily brushed off -
except by certain types of people (i.e., nice helpful people).
  Does that not seem plausable to you?


Thanks for considering this.


           Kerry raw-NEP|Kpolling-place|Kerry loss
(NEP#52)CINCINNATI 4-M  68%     46%        22%
     sample = 31

(NEP#57)HARRISON C       43%     29%      14%
     sample = 28

(NEP#55)FAIRFAX B        50%      37%     13%
     sample = 16

(NEP#49)ANDERSON JJ    32%       24%      8%
     sample = 34

(NEP#51)CINCINNATI 22-E 88%       82%     6%
     sample = 53

(NEP#63)SHARONVILLE 4-C 30%      30%      0%
     sample = 48

(NEP#54)EVENDALE D       25%     33%      -8%
     sample = 52

[kiwi_expat]

24. "Reluctant Kerry Responders" in Cuyahoga County. :-)

The Cuyahoga County NEP precinct that had the greatest deviation was precinct 11, which had a raw-to-actual Kerry GAIN of 11.82%.

So far, the most consistent factor among the OHIO NEP precincts having really large deviations is that they are NATIONAL precincts. See:
http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=329749#364923


p.s. TIA, I beg you NOT to start a new thread called Now 'They' Are Saying 'Reluctant Kerry Responders'.

"Reluctant Kerry Responders" is a joke!!! :-) :-) :-)

[berniew1]

9. What are the odds Bush really won all precincts in Clermont County, Ohio

even the minority precincts where an underfunded African American candidate for state judge won the precinct?

**********************************************************

• Kerry did not win a single precinct of Clermont’s 191 precincts.
• Judge Connally won 5 Clermont precincts even though she had a low-budget campaign and her opponent, Moyer, was the sitting Ohio Chief Justice and had a well-financed campaign..
• Connally had more actual votes than Kerry in 161 of Clermont’s 191 precincts.
• There were more than 72,000 total votes in the Chief Justice race and more than 87,000 total votes in the presidential race.
• Countywide, Connally had 4,146 more votes than Kerry, even though Connally’s chief justice race had 15,000 fewer total votes than the presidential race.
• Kerry’s percentage of the precinct presidential votes has a moderate negative correlation (-0.536) with the precinct turnout percentage. In other words, in precincts where Kerry had a higher percentage of the presidential vote, he tended to have a lower percentage turnout
****************
White Stickers on Kerry ballots

http://rawstory.com/news/2005/index.php?p=7

[TruthIsAll]

10. Well, since blacks went 90% for Kerry, I'd say..ZERO.

And that's being conservative.

[kster]

18. kick.nt

[davidgmills]

19. They don't want to do math

They just want to do bias, that way you don't have to do math.

[TruthIsAll]

21. Still waiting...5 days and counting

.