PROBABILITY ANALYSIS PROVES: THE RBR HYPOTHESIS IS VIRTUALLY IMPOSSIBLE!

[TruthIsAll]

This is a sensitivity analysis to determine the percentage of
refusers Bush needed to get his recorded 2-party vote of
51.24%.

The analysis indicates that the probability of Bush obtaining
the percentage of refusers required to achieve his 2-party
vote is VIRTUALY ZERO.

Given the following:

1) Responders to the National Exit Poll accounted for 52-54%
and Refusers 46-48%.
2) Kerry received 51.5% (2-party)of the responders.
3) Bush received 51.24% of the 2-party vote

Determine: 
The required percentage of Bush refusers required to achieve
51.24% of the total 2-party vote.

A probability  analysis is listed below for the various
scenarios.

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

This is the calculation assuming a 53% response rate. 
Kerry won 51.5% of the Responders. 
For Bush to get to 51.24%, he needed 54.33% of the Refusers, a
5.83% increase. The MoE was 1%.

Poll	Pct	Kerry	Bush
Resp	53.00%	51.50%	48.50%
rBr	47.00%	45.67%	54.33%
	Total	48.76%	51.24%
			
**********************************************************			

Percentage of Refusers Required for Bush to get 51.24% of the
2-party vote

K2P: Kerry's 2-party percentage of responders. 
The actual K2P% was 51.5% as indicated below. 

Other K2P% are for sensitivity analysis only.		

Percent of Refusers Required to Have Voted for Bush
 for Various Voter Response Rates 
 and Kerry 2-party percentages

		Responders	
K2P%	52%	53%	54%

50.0	52.58	52.63	52.70
50.5	53.12	53.20	53.28
51.0	53.67	53.77	53.88
51.5	54.20	54.33	54.45 <<<< actual
52.0	54.75	54.90	55.05

Bush Pct Change from Responder to Refuser			
			
50.0	2.58	2.63	2.70
50.5	3.62	3.70	3.78
51.0	4.67	4.77	4.88
51.5	5.70	5.83	5.95 <<<
52.0	6.75	6.90	7.05
			
Probability Matrix
50.0	2.14E-07	1.27E-07	6.06E-08
50.5	6.50E-13	2.07E-13	6.43E-14
51.0	0.00E+00	0.00E+00	0.00E+00
51.5	0.00E+00	0.00E+00	0.00E+00 <<<
52.0	0.00E+00	0.00E+00	0.00E+00

[Febble]

1. See my response on the other thread

You've forgotten to allow for the fact that the proportion of Bush votes to Kerry votes in precinct sample was not the same as in the country.

Get your figures from the E-M report. It's all there.

[Sancho]

11. I must be missing something....

What evidence, profile, or clues actually indicate that the imaginary rBr exists? IF there is an rBr, then someone with the DATA could easily describe the age, income, race, etc. of such a creature along with why we've never heard of such before.

There are clues and motivations that are all over the place that computers were hacked and election supervisors locked the doors and voter rolls were manipulated...all in one direction and in key swing states.

There is no homunculus of a rBr voter, but there are 13,000 people who filled out about 25 questions about themselves!!

IF the polls differ from the actual vote, there's only one explanation with evidence of the difference - and all we need to generalize the 13,000 to the actual vote is the stratification and demographics given by the sample - which EM won't give up...why?

I think we have rTp...Reluctant Truthful Pollsters...

[tommcintyre]

16. "such a creature...never heard of...before". rBr was "invented" in 2000

Ck it out, it's really getting even more unbelievable.

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x369999

[TruthIsAll]

2. According to Febble...

According to Febble:

"If the exit poll discrepancy was due to response bias,
the direction of bias was relative undersampling of Bush
voters. About 46% of all voters approached, both Kerry and
Bush voters, refused to be interviewed.

Of these, E-M calculate that to account for the discrepancy
between the exit poll and the vote count, about 47% of the
refusers would have been Kerry voters and about 53% would
have been Bush refusers. Not a great difference, but enough
to swing the poll. You may not accept this hypothesis, but
this is what the "rBr" hypothesis is. That there
was a slight bias (47%:53%) in the ratio of Kerry
non-responders to Bush non-responders"

[Febble]

3. Yes well

get the numbers from the E-M report and do the math yourself.

What you call the answer doesn't matter. You could equally well call it the proportion of votes switched to Bush.

Actually, you are arguing yourself out of your own case. If rBr is impossible so is fraud.

The numbers are the same.

The magnitude of rBr is the same as the magnitude of fraud. It's the same math. The math, as I have said before, does not distinguish between bias due to bias in the count, and bias due to bias in the responses.

So you might as well get it right.

Then you can call it what you like.

[autorank]

4. but I thought...?????

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=365678&mesg_id=368959&page=

[Name removed]

7. Deleted message

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

[autorank]

9. Showing your true colors!

[TruthIsAll]

5. No, I didn't disprove it. I just said it was VIRTUALLY IMPOSSIBLE.

That's good enough for most of us.

And it's good enough to postulate prima facie evidence of fraud.

[Febble]

8. Well if it's virtually impossible

that makes fraud virtually impossible too.

Are you seriously saying it is impossible that votes were switched in those proportions to Bush?

You surprise me.

[TruthIsAll]

10. Febble, in case you are lurking, I shall respond...

"Are you seriously saying it is impossible that votes were switched in those proportions to Bush"?

Yes, I'm saying exactly that.
Do the math.
You surprise me.

Run the normal distribution function as I did to determine the probability of Bush more than reversing the percentages over the approximate 13000 who refused to be interviewed.

To wit: In order to get 51.24% of the 2-party vote, Bush had to
improve from the 49.50% of the 53% who responded to 53.20% of the 47% who refused:

Here's the probability calculation:

P = NORMDIST(0.495,0.532,0.01/1.96,TRUE)
P = 2.07E-13

or 1 in 4,834,782,208,664
One in 4.8 trillion.

Looks impossible to me.
Doesn't it to you?
How do you like THAT Fancy TIA Function?

Oh, yes, I admit I'm making the ridiculous assumption that there was NO rBr effect. I realize that you believe it is highly presumptuous on my part to believe that Kerry reluctants canceled out the Bush reluctants.

Well, Febble, they would do more than just cancel; the Kerry reluctants were probably former moderate-conservative Republicans who did not want to draw attention to themselves.

Those Bushian naysayers who love your work can be quite vocal, you know. Just look at Sawyer.

[cser]

6. I'm glad I'm not Scottish, or have the wrong educational

backround. Because I can respond. Yay!!!
Truthisall is beyond criticism. Yay!!!

[kster]

12. kick.nt

[autorank]

13. kster, You sly dog

I wondered how this highly worthy post got back up there again and...
here you are.
:hi:

[kster]

14. How ya doing, we have

to make sure this one doesn't get buried,for a couple of days.

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x369999

[autorank]

15. Great! KICK for posterity!

[Bill Bored]

17. TIA, I think you should read Febble's question again.

She asked:

"Are you seriously saying it is impossible that votes were switched in those proportions to Bush"?

She is asking you if you think fraud is impossible based on the numbers and you said:

"Yes, I'm saying exactly that."

TIA, I think you are working too hard on this. Relax, take a deep breath, don't ASSUME that Febble is a Freeper, does not believe in the possibility of fraud, etc., and just look at the question she is asking here.

If I may, she has asserted that rBr and Fraud would look the same in the numbers, i.e., a (constant?) mean bias toward Kerry in the polls or toward Bush in the vote count, respectively.

I would argue on your behalf that they need not look the same, but even if they did, they are not mutually exclusive. There could be both Fraud and rBr, or even rKr if the Fraud and/or the rBr were enough to offset rKr! In fact, there could be rBr in Kerry precincts, rKr in Bush precincts, and Fraud anywhere and everywhere, couldn't there?

My last sentence was a bit of a reach on my part, but unless it's disproved, it has to be considered. So show me it ain't so!

I could easily see how there could be rKr in Bush precincts because Bush voters are, in many cases, wackos! Who would want their gun toting, uninformed, war mongering neighbors to know that they voted against their guy, whom they think is on a mission from God? People are getting thrown out of church now based on their party affiliation! So there must have been some reluctant Kerry responders!

So why not think in broader terms instead of just rBr and Fraud? Include rKr too, otherwise you're just towing Mitofsky's line!

OK, now the two of you, get back to work! :)

[TruthIsAll]

18. Yes, I read it wrong. I took her question as...

"Do you really think it is impossible that Bush gained that many votes?"
Perhaps I was too hasty and misinterpreted.

As for Febble, no she is most definitely NOT a freeper.
Whether or not she really wants to prove fraud is questionable.

If I'm not too cool and relaxed, it's because I can't stand busllshit, lies and misrepresentations.

And Febble has clearly misrepresented USCV.

[kiwi_expat]

19. Isn't "rBr" the net reluctant Bush and Kerry responders?

Of course there are reluctant Bush responders and reluctant Kerry responders. We all agree that is true. :-) That is why there is a 50% response rate.

The term "rBr" is usually used as a net, meaning that the reluctant responders included (slightly) more Bush voters than Kerry voters. But the net could be negative ("rKr") or zero.

We really should use a neutral term for the net, such as ("Positive"/"Negative") "Reluctant Respondent Differential".


Cheers.

[TruthIsAll]

20. True, so lets see what the percentage increase was

in the Bush share of refusers as compared to his share of the responders.

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x371968

[Bill Bored]

21. Good point I think. nt