National Election Data Archive Project
31 March 2005
Although the election is over, serious questions remain unanswered.
On March 31st, our National Election Data Archive Project released a new scientific paper and executive summary report on the discrepancy between exit polls and official vote tallies in the 2004 presidential election.
More than 27,000 reports of irregularities in the 2004 election were submitted to the independent "Election Incident Reporting System"; thousands more were reported to other organizations.
Unprecedented discrepancies between exit poll results and final tallies in several key states occurred that still have never been explained. It has only recently been officially confirmed (by the exit pollsters themselves) that on election night the final set of exit polls showed John Kerry defeating George Bush by 3% of the popular vote and a clear majority of 316 electoral votes. Our statisticians analyzed Edison/Mitofsky's own explanation of their exit poll discrepancies, and found serious flaws in their argument. Exit polls have been used for years to detect corruption of official vote tallies - most recently in Ukraine.
A General Accounting Office (GAO) investigation is already underway into the security and accuracy of voting technologies, the distribution and allocation of voting machines, and counting of provisional ballots. The FBI, the Department of Justice and the Congressional Research Service have also been officially asked to investigate various aspects of the 2004 election.
We believe that the new National Election Data Archive Project can apply statistical techniques to identify the best places to focus the investigation.
We have formed a volunteer scientific research project to create and analyze - for the first time ever - a database containing precinct-level election results for the entire United States.
This rich mine of data will be made publicly available and analyzed by our project's affiliated mathematicians, pollsters and statisticians, as well as by an independent peer-review board. Our goal is to use this data to develop and test techniques to reliably detect precinct-level vote counting errors worthy of conventional investigation. We will start with an in-depth statistical analysis of the 2004 election.
By the national election in November 2006, for the first time in American history, it could be possible for candidates to be reliably warned of indications of machine or human-caused vote count errors in time to challenge the results. With a sound scientific approach and methodology, it may be possible for our project staff to develop statistical evidence in support of legal filings and serve as expert witnesses for candidates, regardless of party affiliation.
Please help us now to defend the integrity of the American ballot.