Edited on Wed Mar-30-05 01:44 PM by TruthIsAll
First, you say the ABSOLUTE MAXIMUM percentage in the "MIX" (of how voted in 2000) for Bush is 41.26% based on the TOTAL number of Bush 2000 voters/Total 2004 votes.
That is true.
But this is only the "absolute maximum" if the exit poll was a poll of all 122M people (and the 17% and 3% were right, Assumption I). But it is not. It is an exit poll of 13000 people. I guess this applies to Assumption I. In a poll of 13000 out of 122 Million, I don't know why you would assume that the 17% mix is an exact reflection of the actual numbers. Where were the samples taken, were they representative, were the people truthful, ...
You fail to see that the 41.26% is not dependent on how many were polled. It is just the fraction of the total number of Bush 2000 voters (50.456) divided by total votes (122.26) in 2004.
So it is AN ABSOLUTE MAXIMUM.
IT IS IRREFUTABLE ARITHMETIC.
AND THAT IS THE BASIS FOR THE ANALYSIS.
You show a complete lack of understanding regarding polling in general and exit polling in particular.
Must I resort to basics?
No one polls 122 million people.
They poll a representative sample.
The accuracy of a poll has nothing to do with how many are in the universe (here 122 million) to be polled. I'm sure you will disagree, but you would be wrong.
The accuracy of any poll is based on the NUMBER polled, assuming a random sample.
So for 10 million, or 50 million or 100 million - it makes no difference.
If 1,000 are polled. the margin of error (MOE) is 3.1%
If 10,000 are polled, the MOE is 1.0%
If 13,047 are polled, the MOE is 0.875%
How do I know? Pick up a statistics text.
MOE = 1/sqrt(N), where N is the sample size.
Edison/Mitofsky have been doing this for years.
They state that the National Exit Poll MOE is 1.0%
This agrees with the formula.
FOLLOW THIS LOGIC VERY CLOSELY, BECAUSE YOU HAVE FAILED TO PICK IT UP IN MY POST.
The BUSH 43% WAS IMPOSSIBLE.
His actual number had to be LOWER than 43%, because it is limited to the 50.456 million who voted for him (or 50.456/122.26).
So his Maximum is 41.26%
Therefore, AT LEAST ONE of the other groups (New voters, Gore Voter, Other Voters) MUST HAVE HAD A HIGHER PERCENTAGE THEN the number(s) presented in the poll. And the ORIGINAL numbers were 37%, 17 and 3%, respectively.
OK, let's continue:
The BUSH 41.26 WAS ALSO IMPOSSIBLE.
Some of his 2000 voters died.
The annual death rate is 8.7 per thousand.
That is 3.5% over a four year period.
So his MAXIMUM vote is REDUCED by 3.5%
And his NEW MAXIMUM becomes 39.82%
Therefore, AT LEAST ONE one of the other groups (New voters, Gore Voter, Other Voters) had to be have been HIGHER than the number presented.
And those numbers were 37%, 17% and 3%, respectively.
OK, let's continue:
The BUSH 39.82% WAS IMPOSSIBLE.
Some of his 2000 voters stayed home and did not vote.
We don't know how many.
Again, this means that AT LEAST ONE of the other groups (New voters, Gore Voter, Other Voters) had to be HIGHER than the number presented.
THAT is MY point.
Now, understand this: I assume that the 17% and 3% are correct to be CONSERVATIVE. We know that 3% is a MAXIMUM, because only 3% VOTED FOR NADER et al in 2000.
So the only way that the 17% (WHO DID NOT VOTE IN 2000) is incorrect is if the Gore vote WAS HIGHER THAN 39%. But let's assume that the 39% is correct.
THEN THE 17% WHO DID NOT VOTE MUST BE A MINIMUM NUMBER.
BUT 54 TO 57% OF THOSE WHO DID NOT VOTE IN 2000 VOTED FOR KERRY, DEPENDING ON WHICH POLL YOU CHOOSE TO BELIEVE. SO THAT MAKES IT EVEN LESS LIKELY THAT BUSH COULD HAVE WON.
I guess here is my point. You assume the 17% and 3% are correct, and that the 43/37% are wrong. Why would you trust one over the other? Only because they did not change? Your 41.26% calculation does not justify this, because again, the exit poll %s are not based on all 122M voters... it is based on 13000. It could have feasibly yielded 100% for Bush if the right (wrong) voters were polled. Admitedly impossible without severe tampering, but a divergence of a percentage point or two is feasible. You are talking about a sampling... not the absolute total numbers. There is no "absolute" maximum possible in an exit poll. And part of what you are basing your extrapolations (to the full results of the election) on are the results of an exit poll.
HERE YOU TOTALLY MISS THE MOST BASIC POINT OF THE ARGUMENT:
THE 41.26% HAS NOTHING WHATSOEVER TO DO WITH THE RESULT OF THE ELECTION, IT IS BASED ON TWO FACTUAL NUMBERS.
1) HOW MANY VOTED FOR BUSH IN 2000: 50.456 MILLION
2) HOW MANY VOTED IN TOTAL IN 2004: 122.6 MILLION.
THESE ARE HARD NUMBERS.
ONLY A HARD-HEAD WOULD DISPUTE THAT 50.456 MILLION IS PRECISELY 41.26% OF 122.6 MILLION.
Also, you close by saying:
A) "THEREFORE, THE FINAL 13660 EXIT POLL, WHICH IS WEIGHTED TO
MATCH THE RECORDED VOTE, IS WRONG."
B) "THEREFORE, SO MUST THE RECORDED VOTE BE WRONG."
But why might not the exit poll be wrong? Or why might there not be merely evidence that their "weighting" is messed up. To me, Statement "B" does not logically follow from statement "A".
You fail once again to see the obvious.
I will repeat.
THE FINAL EXIT POLL IS WRONG - WE JUST PROVED IT.
THE FINAL EXIT POLL MATCHED TO THE RECORDED VOTE - WHICH BUSH WON.
IF THE FINAL EXIT POLL IS WRONG, AND IT WAS MATCHED TO ANOTHER NUMBER (THE RECORDED VOTE) THEN....
THE RECORDED VOTE MUST ALSO BE WRONG.
Please do not assume I am a Freeper because I am not convinced. I have seen your posts and I am trying to find rock solid evidence of fraud, but I am sorry, i just do not see it yet. Believe me, I wish I did. I see serious problems with the fact that the percentages changed so drastically with only 600 more voters polled. but the Exit polling is not just based on these raw numbers, apparently (or so they say) and always (even the first numbers) includes "weighting". This gives them all the room in the world to wiggle out of actually having to explain anything. I think it's highly suspicious, but it also makes it highly unlikely we'll ever prove fraud based on exit polling
That's OK if you are not yet convinced.
I am sure that as you come to understand the logic, you will become convinced.