Open source political data analysis: 2004 Election Hypotheses of Fraud Remain Credible

Note that I worked with Edison in the last election.

-----Original Message-----
From: "David Farber" <dave@farber.net>

Date: Sunday, May 15, 2005 10:35 am
Subject: [IP] Open source political data analysis: 2004 Election Hypotheses of Fraud Remain Credible

_______________ Forward Header _______________
Subject: Open source political data analysis: 2004 Election Hypotheses of FraudRemain Credible
Author: Peter Jones <peter@redesignresearch.com>
Date: 14th May 2005 12:19:04 pm
Dave - This is what you might call open source political analysis. But what's needed is open access to the public data that we as citizens have paid for! This team - four statisticians and a political scientist -
collaborated remotely much of this year on analyzing the 2004 presidential election exit poll data. It's the strongest study of its kind I've seen.
They have run multiple simulations with sensitivity analysis to assess the
Edison/Mitofsky hypotheses about Democratic vs. Republican exit poll responders, and definitively reject their hypothesis. They statistically support that the most plausible explanation supported by simulation was a significant vote "shift" from Kerry to Bush.

They emphasize that: "Simulations are only used in the absence of detailed source data. Edison/Mitofsky could help us resolve the lingering questions about exit poll discrepancies by releasing the data to enable us to directly measure - rather than infer - precinct level variance between poll results and official vote tallies." Link to report at bottom of post.
Press contact in press release.

Peter Jones http://redesignresearch.com
>From the Abstract:
New evidence from mathematical simulations conclusively shows that any constant mean exit poll response bias hypothesis such as the "reluctant Bush responder" (rBr) hypothesis is not consistent with the pattern shown by the
Edison/Mitofsky exit polling data. Other explanations are required to explain the Edison/Mitofsky pattern of exit poll discrepancies and overall response rates.

US Count Votes' simulations have demonstrated that exit poll patterns in the
November 2004 presidential election could be produced by an exit poll response bias distribution with constant mean if accompanied by shifting of votes cast for Kerry to Bush; or alternatively, the patterns could be caused by a differential pattern of exit poll response bias that would require further explanation.
FROM THE PRESS RELEASE

2004 Presidential Election: Hypotheses of Fraud Remain Credible; New
Scientific Study Released

kathy@uscountvotes.org electionarchive.org

The persistence of credible hypotheses of election fraud, six months after the election, underscores the fragility of the U.S. electoral system. US
Count Votes continues its systematic statistical study of the discrepancy between the Edison-Mitofsky exit polls and November's reported presidential election results.

Miami, FL. - Ron Baiman, Ph.D. of US Count Votes and the Institute of
Government and Public Affairs of Chicago, will release the new results at the meeting of the American Association of Political Opinion Researchers today, Saturday at a 2:15 p.m. Press Conference in the Hotel Fontainebleau Hilton Resort lobby, 4441 Collins Avenue, Miami Beach, FL
33140.

Mitofsky, of the Edison/Mitofsky group who released an analysis on January
19th of their November 2nd exit poll that had predicted a strong win for Kerry, will also be in attendance at the AAPOR conference.

The National Election Data Archive (NEDA) today has released a new report, demonstrating that data from the Edison/Mitofsky analysis is consistent with the hypothesis of a corrupted vote count, and inconsistent with the competing idea that Bush voters were under-sampled in the poll. Using numerical modeling techniques to simulate the effect of polling bias, NEDA scientists are able to reproduce signature patterns in the Edison/Mitofsky data by incorporating a general shift in the official vote tally in the model.

Most telling is the fact that the highest participation rates and the peak disparity between poll and official returns both occurred in precincts where
Bush made his strongest showing. This feature of the data is inconsistent with the Edison/Mitofsky assumption that polling bias was responsible for the gap.

For the complete report, see http://uscountvotes.org/ucvAnalysis/US/exit-polls/USCV_exit_poll_simulations .pdf
This paper follows an earlier study released on March 31, 2005, by a group of statisticians for the National Election Data Archive Project, Analysis of the 2004 Presidential Election Exit Poll Discrepancies.

-------------------------------------