Voting System Testing
Voting System Certification
The Help America Vote Act of 2002 established the U.S. Election Assistance Commission (EAC) as the federal agency to certify independent testing laboratories and establish voting system standards. These Voluntary Voting System Standards came into effect in December 2007, and all voting systems submitted for certification after that date are exhaustively tested by the independent testing laboratories to ensure conformity with them. Once this testing is complete, a report is written and submitted by the laboratory to the EAC. The EAC then determines whether to grant certification of the system. This certification is proof that the voting system has met the performance and test standards.
Previously, the National Association of State Election Directors (NASED) approved independent testing authorities (ITAs) to test voting systems against the performance and test standards established by the Federal Election Commission, which were subsequently updated by the EAC.
In addition to federal certifiation, a voting system used in Maryland must complete state certification. This is a state testing process to ensure that the voting system meets all of Maryland’s statutory and other voting system requirements.
To be certified, a voting system must:
- Protect the secrecy of the ballot
- Protect the security of the voting process
- Count and record all votes accurately
- Accommodate any ballot used in Maryland
- Protect all other rights of voters and candidates
- Be capable of creating a paper record of all votes cast in order that an audit trail is available in the event of a recount
Every voting unit used in Maryland has undergone a comprehensive, two-part State acceptance test. The first part involves a diagnostic test to ensure that each voting unit and all of its components are performing to the required specifications. The second part involves casting hundreds of votes on each voting unit. A report showing the vote totals is printed from the unit and compared against the expected results. This test ensures that the voting unit is accurately recording and counting votes.
The results from these voting units are then transferred to the central tabulating computer that totals all the votes from the voting units. This ensures that every vote put onto the voting unit during acceptance testing is counted by this central tabulating computer, and that the correct ballot totals are maintained throughout the counting process.
Logic & Accuracy Testing
Logic and Accuracy (L & A) testing is the process by which voting equipment is configured, tested and certified for accuracy prior to each election. Each component is tested to verify that it is fully functional and free from mechanical problems and that each voting unit contains the appropriate ballot styles for its designated polling place.
L & A testing consists of multiple phases:
- Each voting unit (both touchscreen and optical scan) is prepared and configured for it's designated polling place. The correct ballot styles are downloaded to the voting unit to be used in that election.
- Hundreds of test votes are cast on each voting unit to ensure that it is recording votes accurately. For each touchscreen voting unit, more test votes are cast than the maximum number of voters that will use the voting unit on Election Day.
- These test votes are recorded and counted by each voting unit as well as the central tabulating computer. This ensures complete accuracy throughout the entire voting and vote counting process.
- All test votes are cleared, and the voting units are locked, sealed and secured ready to be used in the election.
Parallel testing is a method of testing an electronic voting unit by producing an independent set of results that can be compared against the results produced by the voting unit and is cited as a best practice by election administration and computer experts. Maryland was one of the first States to adopt this method of testing, in 2004.
During parallel testing, two individuals read aloud the votes cast on a paper ballot. Two people separately record the votes cast on paper, and two people cast the votes on a touchscreen voting unit. The teams periodically check to ensure that the two hand-tallies match and that the number of ballots cast match. At the conclusion of parallel testing, the two sets of hand-tallies are compared to the results generated by the voting unit. The hand-tallies and the voting unit totals should match.
In Maryland, parallel testing is conducted two times for each election. Parallel testing is first conducted in each county during the pre-election public demonstration and also on Election Day from 7:00 a.m. to 8:00 p.m.
In 2008, fifty ballots were cast and counted during the pre-election parallel testing in each Local Board of Elections (LBE). Each pre-election parallel testing demonstrated the accuracy of the voting system – the hand-tally and the results from the voting unit matched. On Election Day, hundreds of votes were cast during parallel testing. The vote totals and hand-tallies for the remaining units matched excatly.
This kind of testing continues to confirm the accuracy of the voting unit in recording and tabulating votes. Given the fact that every voting unit in the State uses the exact same software, voters in Maryland can be confident that their votes are accurately counted.
Post-election Audit and Verification
After each election, each LBE verifies that the vote totals printed from the individual voting units match the reports generated by the central tabulatinf computer. This manual verification of a designated number of precincts validates that the central tabulator correctly totaled the results from each voting unit.
While the voting unit verification is being conducted, each LBE also conducts a post-election audit to confirm the accuracy of the polling place reports. This includes auditing signed voter authority cards, pollbook data, other polling place forms completed by the election judges, and the official election results.
If a discrepancy is found during the verification or the post-election audit, the local board continues its audit until it determines the cause of the discrepancy.