Fuzzy Math (need mathamaticians to verify this)

[Streetdoc270]

Sources for data are
CNN.com election page for total votes cast
Washington Post exit polls for demographics

Bush-58,880,024
Kerry-55,324,556

WP Demographics
Male Voters -46%
Female Voters-54%
Male vote per canidate - Kerry 47% Bush 52%
Female vote per canidate -Kerry 54% Bush 45%

Total Votes Cast- 114,204,580
By Men -52,534,106
By Women -61,670,473

Kerry -24,691,029.82 Males
33,302,055.42 Females
Total 57,993,085.24

Bush 27,317,735.12 Males
27,751,712.85 Females
Total 55,069,447.97

So, I am showing close to 3 Million votes in Kerry's favor. This leads us to 3 conclusions.

1) My math stinks (very possable)
2) People lied to the exit pollsters (why?)
3) Miscounts nationwide...

I would like serious help with this please. I may be on to something here I just need people who are smarter than I am to help me prove it

[Radio-Active]

1. looks right to me!

I didn't break out the calculator though..

I'd say #3

[Rhiannon12866]

7. I'd say #3, as well. It's the only option that makes sense

But what I fail to understand is how they can have miscounted when they still haven't finished counting.:shrug:

[NewYorkerfromMass]

2. Exit poll numbers are rounded and may be leading to error

I am not sure how you come up with these.

[billyskank]

4. OP came up with those numbers

by multiplying the numbers of men and women voting by each candidate's share of each. Then adding each candidate's male and female totals.

The OP's math does not stink. The answers do not agree with the other figures quoted. Kerry gets a higher number because he won amongs women voters, who turned out in higher numbers. Apparently. (What can I say, I'm sitting here in the middle of England).

[JohnDoe1]

3. Conspiracy Theorists Unite

Exit polls do not equal actual results. They are a subsample that is applied to the nation in general. In an election this close, a small discrepancy from sampling error could easily lead to results predicting the other side victorious. In this case, for all we know females may be more likely to respond to exit polls than males increasing the percentage of female voters. As much as I hate it, we need to realize that we lost and get over it.

[Streetdoc270]

8. I WILL NOT GET OVER IT!!!!!

I was told this in 2000 to 'get over it and move on' and move on we did, into a War with no end, Deficits that my G-G Grandkids will be paying for, and the end of the Bill of Rights all in the name of 'patriotism' and 'security'

Well I for one will not get over another stolen election! I will not stand idlly by as GLBT's get shoved back into the closet and women are left to die in back ally abortion clinics! I will fight the fight that Kerry said he would so there!

(sorry for the shouting but these 'get over it' disrupters are driving me nuts!)

[RaleighNCDUer]

9. Look at it. There are separate tallies for men and women, they

are not grouped together. Therefore, the percentages within each sample are true to their own sample. If you talk to 4 men and 10 women and get 50% on each, you don't say it is skewed toward the women because you have 5 positives from women and only 2 for men. 50% is 50%, period.

Exit polls have historically proven to be accurate within 2-3%. That's why we have used them for decades. That's why a 5%+ variance between the exit polls and the vote count is suspect. It's way outside the normal margin of error.

Only since 2000 have they suddenly become suspect.

[Iceburg]

5. Your math stinks because it does not factor in f(x) where x= BBVR

the Bush Box Verification & Redistribution factor. When the chips are down, take a few from the top and let the chips fall where they may ....

[Awsi Dooger]

6. Those percentages by gender are dramatically different than CNN exit polls

http://www.cnn.com/ELECTION/2004/pages/results/states/US/P/00/epolls.0.html

CNN now lists it as Bush 55 Kerry 44 among men. Last night they had it 54-45, which seemed low. CNN now has Kerry only 51-48 among women, even worse than the 52-47 they posted last night.

The total sample now is over 13,000 which should produce almost zero margin for error. But I'm interested to see other exit polls and if they also differ significantly, which wouldn't make sense.