COUNTY OF SAN MATEO

Inter-Departmental Correspondence

BOARD OF SUPERVISORS

 

DATE:

December 4, 2006

BOARD MEETING DATE:

December 12, 2006

SPECIAL NOTICE/HEARING:

NONE

VOTE REQUIRED:

Majority

 

TO:

Honorable Board of Supervisors

FROM:

Supervisor Mark Church and Supervisor Rose Jacobs Gibson

SUBJECT:

Preliminary Report on Performance of Electronic Voting Machines in the November 2006 Elections.

 

RECOMMENDATION:

Accept the Preliminary Report on the Performance of Electronic Voting Machines in the November 2006 Elections.

 

VISION ALIGNMENT:

Commitment: Realize the potential of our diverse population; and responsive, effective and collaborative government.

Goal(s) 2 and 20: Civic engagement through uniformly high voting among diverse populations; and government decisions based on careful consideration of future impact, rather than temporary relief or immediate gain.

 

BACKGROUND:

On August 15, 2006, the Board authorized the execution of an agreement with Hart InterCivic, Inc. (Hart) for new voting system hardware, software, and support services and for an election management system. The Board also authorized the County’s Chief Election Officer or his designee to execute necessary applications, agreement, and claims to secure grant funding, and to execute subsequent amendments to make minor changes in the Hart agreement.

Electronic Voting Machines were introduced by the Chief Elections Officer in order to comply with the Help America Vote Act of 2002 and California law, which requires that voting systems used in elections after January 1, 2006 be certified compliant with federal and state standards intended to help all voters cast ballots accurately, to assist non-English-speaking voters, and to allow voters with disabilities to access polling places and to vote privately and independently.

At the August 15, 2006 hearings, the Board of Supervisors formed a subcommittee comprised of Supervisor Mark Church and Supervisor Rose Jacobs Gibson in response to public comment expressing concern over the reliability and accuracy of the new voting machines.

 

DISCUSSION:

Preliminary reports of Electronic Voting Machine performance during the November 2006 election have been positive, and a more thorough audit of these machines will be performed after the December 5, 2006 certification of election results. At this time, however, reliability and accuracy do not seem to be issues.

Each of the 472 voting precincts in the County had one Hart voting machine available to voters. Malfunctions were reported but most of these were attributed to the first time usage of the machines by poll workers. There were, for example, instances where the voting machine was not plugged in properly causing the units to run on battery power which eventually ran out and there was some difficulties with the initial setup of the machines first thing in the morning (paper ballots were available at all polling places so that voting was not impeded). This incremental roll out of the new voting machines seemed to work well considering the experiences of jurisdictions across the country that rushed to complete a full deployment of new machines and new technology.

The incremental deployment of the eSlate voting devices provided the elections office, the pollworkers and a small number of actual voters the opportunity to become familiar with the devices in a very controlled fashion.

To address concerns regarding the security and integrity of the election process, the election office implemented most of the security recommendations contained in a recently published election security study completed by the Brennan Center for Law and Justice at NY University School of Law. Physical building security enhancements were made including the installation of additional alarms and cameras. The chain of custody of all machines was electronically tracked through a system of scanners and bar codes. Machines, ballot boxes and other relevant items were protected by serialized locks and/or tamper evident seals. In addition, the elections office ran two “computer hash” programs, one prior to the election and one following the election to ensure that the County used the certified version of the vote counting software to conduct the vote count and to prove that the software used to conduct the election had not been altered.

The office also worked with Professor David Dill of Stanford University and other members of ACCURATE (A Center for Correct, Usable, Reliable, Auditable and Transparent Elections) to perfect new and improved procedures for the audit process which immediately follows the election – called the Canvass of the Vote. ACCURATE is a multi-institution, interdisciplinary, academic research project and eventually some or all of these new processes could be made a part of state law which would greatly strengthen this important aspect of elections, not just for San Mateo County, but for California and the nation as a whole.

Some of the new and improved procedures such as the method to randomly select the precincts for the 1% manual recount and the inclusion of absentee ballots were implemented as a result of working with this group. The recount is still in progress as of this writing but preliminary results have shown the manual count to match the electronic count, showing the new voting system to be accurate.

Volunteer support played a major role in the successful use of electronic voting machines by the public. Approximately 500 student poll workers (mostly from high school) assisted at the polling places and helped in the set-up of voting machines, answered questions from the public regarding voter machine use and generally assisted in the work of the precinct boards.

Volunteer support was also evident in the dedicated efforts made by our veteran poll workers who visited the drop-in computer lab at Tower Road, on their own time, to practice the set-up, tear-down and trouble-shoot techniques for the new voting equipment – after receiving approximately seven hours of training. More than 1,000 members of our poll worker ranks made these visits — which reveals their commitment to making Election Day a success for our County.

There was also some involvement from voter activists in the County before Election Day, on Election Day and during the canvass of the vote. Activists were present at these different junctures to observe operations. Election Office officials reported that they were respectful and orderly. Activists were also present during the mandatory 1% manual tally process. The time devoted to the activists and observers - explaining local procedures, state laws and answering questions - did divert staff from managing and coordinating vital election operations. As such, a system to address public observation and questions might need to be developed so that it does not interfere with ordinary operations within the Elections Office in the future.

Following the certification of the election on December 5th, the County Elections Office will conduct a thorough audit of machine performance. This audit will take approximately 30 to 60 days. At the completion of this review by the County Elections Office, a more thorough report of machine performance will be presented to the Board in January or February of 2007.

 

FISCAL IMPACT:

NONE