We understand that you have been asked by Representative Joe Neal to review South Carolina's pending purchase of ES&S iVotronic voting machines and determine whether South Carolina should require the use of a voter-verified paper ballot. VotersUnite! is a national organization of citizens concerned about election integrity in the United States.1 At present, we are the only organization dedicated to informing election officials throughout the United States of the facts regarding voting systems. We are writing to call your attention to the following facts.
The use of direct recording electronic (DRE) voting machines does not allow election officials to comply with the South Carolina law requiring recounts in close elections.
Section 7-17-280 of the South Carolina code of laws requires a recount in all elections where the margin of victory is less than one percent of the votes. When a direct recording electronic (DRE) voting machines is used to record votes, a recount of the original votes is not possible. All that is possible is to review the initial count by checking the record of the votes. If the machine made an error in recording votes, that error would be reflected in the electronic record and would impact the review as well as the initial count.
The use of ES&S iVotronics does not comply with the South Carolina requirement for voting systems to meet the minimum standards of the Federal Elections Commission.
Section 7-13-1620(A) states that no voting system may be approved for use in the state unless it meets the minimum requirements of the Federal Election Commission's national voting system standards. HAVA maintains the current 2002 standards to be those minimum standards. ES&S iVotronics do not meet the 2002 standards.
A recent hearing of the U.S. House Science Subcommittee on Environment, Technology, and Standards found the testing and certification process for voting equipment to be inadequate.
There is no federal oversight of the process. Vendors contract independently with the testing labs and supply the test plans. No one, not even election officials, are informed about the testing process. The tests do not include testing for viruses or other malicious code, and even the Independent Testing Authorities (ITAs) agree that the standards are not adequate.
Even if the iVotronics met the current FEC standards (which they do not), experts and lawmakers agree that those standards are insufficient to ensure the adequacy of the systems.
1 Since April 20, 2004, VotersUnite! volunteers have, at their own expense, delivered "Myth Breakers for Election Officials" to 534 local election officials across the country, as well as hundreds of state and federal legislators, journalists, and other interested parties. Supplemental information about voting systems can be found on our web site, www.votersunite.org.
Section 7-13-1620(F) requires a field test of any new voting equipment, held in two or more precincts during a primary, general, or special election. This law states that the test must demonstrate the accuracy of votes reported on the machine.
We are not aware that such a field test was conducted on the iVotronics. If it has been conducted, the only way to demonstrate the accuracy is by using parallel testing techniques described in the Brennan Center report.2 We are not aware that this has been done during the procurement process.
The use of ES&S iVotronic voting machines and the Unity accumulation software does not meet the South Carolina legal requirement for accuracy in recording and counting votes.
Section 7-13-1640(A)(6) requires that any electronic voting system used by the state shall "correctly register or record and accurately count all votes cast for any and all candidates and for or against all questions." A history of recording and counting problems with these systems indicates that they cannot be relied on for accuracy:
Miami-Dade County, Florida. September 2002. The iVotronic voting machines failed to count 8.2% of the votes.3
Dallas County, Texas. October 2002. The iVotronic voting machines registered the incorrect choices on the screen.4
Broward County, Florida. November 2002. The ES&S voting system failed to count 22% of the votes.5 In addition, the iVotronics registered incorrect choices on the screen.6
Wake County, North Carolina. November 2002. The iVotronic voting machines lost 436 ballots.7
Louisiana. November 2003. One-third of the state's 900 new iVotronic voting machines overheated.8
Broward County, Florida. January 2004. The iVotronic voting machines registered 134 blank ballots in a one-race election whose winning margin was 12 votes.9
http://www.votersunite.org/info/letter_to_mcmaster.pdf