Is MP still up for the Sandy Koufax award?
FRAUD 3, rBr 0
**********************************************************
STRIKE ONE
Kerry State Vote vs. Exit Poll completion rate
There’s a negative correlation between Kerry's vote and the completion rate. They move opposite to each other. This contradicts the rBr hypothesis, which states that Kerry voters were more anxious to speak to the exit pollsters than Bush voters.
Let’s view the relationship between Kerry's state vote and exit poll completion rate. View the graph.
http://www.democraticunderground.com/discuss/duboard.ph... Graph 1: Kerry's % of the state vote, in ascending order, left to right.
Graph 2: Corresponding Exit poll completion rate, left to right.
For both graphs, I have calculated the best-fit straight line.
y = a + bx is the equation of a straight line, where b is the slope.
Both "best-fit" linear equations are clearly stated and shown as dashed lines on the graph.
Kerry's vote percentage increases in states going from left to right (increasing slope). The corresponding state exit poll completion rates decrease (negative slope)..
There’s a negative correlation between Kerry's vote and the completion rate. They move opposite to each other. This contradicts the rBr hypothesis, which states that Kerry voters were more anxious to speak to the exit pollsters than Bush voters.
If rBr was a viable hypothesis, as you moved into Kerry strongholds, you would expect that there would be more enthusiasm on the part of voters to be interviewed, not less.
If rBr was a viable hypothesis, as you moved into Bush strongholds, you would expect that there would be less enthusiasm on the part of voters to be interviewed, not more.
**********************************************************
STRIKE TWO
PARTISAN ALPHA
Kerry strongholds: K/B =1.062
Bush strongholds and competitive precincts: K/B= 1.184
K/B is clearly not uniform across partisanships, so how do you measure rBR?
http://www.democraticunderground.com/discuss/duboard.ph... EXIT POLL RESPONSE OPTIMIZATION MODEL
7/2/05 10:20 AM
OBJECTIVE:
Determine minimum weighted K/B (alpha) required to satisfy
partisan precinct WPE's and response rates
Precinct Variable Input (Min, Max) Constraints:
1- Response rates: aggregate R:.53HB,.55B,.52NP,.55K,.56HK
2- Kerry 2-party vote:avg 10%,30%,50%,70%,90%
3- Alpha (K/B): minimum (1.1545)
4- WPE:-10.0%HB, -6.1%B, -8.5%NP, -5.9%K,0.3%HK
VOTE Mil. Pct (Input)
Kerry 59.027 48.76%
Bush 62.029 51.24%
Total 121.056
POLL Mil. Pct
Kerry 63.127 52.15%
Bush 57.929 47.85%
Bush-recorded 51.24%
EXIT POLL RESPONSE (input)
Aggregate 53.5% R
Kerry/Bush 1.1545 alpha
This is minimum alpha required to satisfy
the WPE/partisanship constraints
PROBABILITY
Bush gains 3.39% from the exit poll to the vote: 1 in 62,953,509,332
Kerry exceeds 50% of the vote:100.0%
*********************************************************
STRIKE THREE
Naysayer Hobson's Choice: Final NEP or rBr? Take your pick.
Of the 122.3mm who voted in 2004, 43% were Bush 2000 voters and 37% Gore voters?
First of all, it’s impossible. But even if it was true, how can they claim Bush voters were reluctant when they outnumbered Gore voters by 6%?
http://www.democraticunderground.com/discuss/duboard.ph... Let's continue our discussion of the 43%/37% Bush/Gore
weighting split in the How Voted in 2000 demographic of the
Final National Exit Poll. Bush won the poll (and the vote) by
51-48%.
To review, according to the poll, 43% of all 2004 voters were
former Bush 2000 voters and 37% of 2004 voters were former
Gore voters.
Gore won the popular vote in 2000 by 50.999mm to 50.456
million for Bush.
Some, but not all of you have stipulated to the mathematical
fact that for 43% of 2004 voters to have been Bush 2000
voters was an impossibility, since 43% of the 122.3 million
who voted in 2004 is 52.59 million. And yet we know that Bush
only got 50.456 million votes in 2000.
So now you have a choice:
1) You can argue, against all mathematical logic, that the
final national exit poll was correct; that 43% is a valid
weighting multiplier for Bush and 37% is a valid weighting
multiplier for Kerry. And therefore the Final Exit Poll
correctly matched to the recorded vote count, as reflected by
the 51-48% Bush exit poll win.
or
2) You can continue to hypothesize the reluctant Bush
responder theory. Of course, by doing so, you must reject the
43%/37% split, since that weighting mix implies that Bush 2000
voters outnumbered Gore 2000 voters by a whopping six percent
margin.
So which is it?
Do you claim that Bush won the final exit poll (13660
respondents) and that the 43%/37% split (although
mathematically impossible) is still a valid weighting mix?
Or do you reject the final exit poll and promote the rBr
hypothesis as a possible explanation for the exit poll
discrepancies?
You cannot have it both ways.
Take your pick.
43/37 vs. rBr.
Impossible vs. implausible.