919
Member
- Joined
- Jan 15, 2005
- Messages
- 9,361
- Reaction score
- 73
NEW REPORT says PRES 04 vote outcome 1 in a million
Thu Mar 31st, 2005 at 10:31:43 PST
A new report just released confirms what many of us here have believed for a long time. The likelyhood of bush winning by 2.5% when exit polls showed Kerry winning by 3% just doesn't add up.
The full report can be found here
http://electionarchive.org/ucvAnalysis/US/Exit_Polls_2004_Edison-Mitofsky.pdf
I for one wouldn't drive in a car designed by priests instead of engineers.
I believe tha math, and I don't believe that 1 in a million is very good odds that this thing wasn't rigged.
Table 7. (p. 18): Median Within Precinct Error (WPE) by voting equipment
Type of equipment used at polling place Median WPE Overall Paper ballot -0.9
Mechanical voting machine -10.3
Touch screen -7.0
Punch cards -7.3
Optical scan -5.5
Methods: First, take all of the precincts that were exit-polled on election day calculate the "within precinct error" (WPE, the discrepancy between the exit poll for each precinct and the actual tabulated vote total for that precinct). Second, divide the precincts into groups by voting method used. Third, within groups sort the precincts by the value of the WPE. Fourth, take the middle one (the median) of the sorted group, which represents a "typical" precinct error. The "typical" precinct error was minimally against Kerry for paper ballots (-0.9), and was huge in the other categories (ranging from -5.5 in optical scans to -10.3 for mechanical voting machines).
Thu Mar 31st, 2005 at 10:31:43 PST
A new report just released confirms what many of us here have believed for a long time. The likelyhood of bush winning by 2.5% when exit polls showed Kerry winning by 3% just doesn't add up.
The full report can be found here
http://electionarchive.org/ucvAnalysis/US/Exit_Polls_2004_Edison-Mitofsky.pdf
I for one wouldn't drive in a car designed by priests instead of engineers.
I believe tha math, and I don't believe that 1 in a million is very good odds that this thing wasn't rigged.
Table 7. (p. 18): Median Within Precinct Error (WPE) by voting equipment
Type of equipment used at polling place Median WPE Overall Paper ballot -0.9
Mechanical voting machine -10.3
Touch screen -7.0
Punch cards -7.3
Optical scan -5.5
Methods: First, take all of the precincts that were exit-polled on election day calculate the "within precinct error" (WPE, the discrepancy between the exit poll for each precinct and the actual tabulated vote total for that precinct). Second, divide the precincts into groups by voting method used. Third, within groups sort the precincts by the value of the WPE. Fourth, take the middle one (the median) of the sorted group, which represents a "typical" precinct error. The "typical" precinct error was minimally against Kerry for paper ballots (-0.9), and was huge in the other categories (ranging from -5.5 in optical scans to -10.3 for mechanical voting machines).