[an error occurred while processing this directive]

REVISITING THE U.S. VOTING SYSTEM: A RESEARCH INVENTORY

November 27-28, 2006

Convened by the American Association for the Advancement of Science

Lillie Coney

ELECTRONIC PRIVACY INFORMATION CENTER

In 2002, with the federal enactment of the Help America Vote Act (HAVA), Congress for the first time created a role for the federal government in the administration of local election when federal offices are on the ballot. Many changes made by HAVA will impact all elections not just those that are federal. The law among other things created a new federal government agency to provide guidance to states, and made obsolete punch card and lever voting machines by mandating disability and language minority access requirements.

The result of the HAVA has been a historic shift from lever, paper, and punch card voting systems to Optical Scan and direct recording electronic (DRE) systems. The numbers of voters who lived in counties served by different types of voting systems have seen a dramatic change in their voting experience. According to Election Data Services, a for profit political consulting firm specializing in redistricting, election administration, and the analysis and presentation of census and political data with geographic information system the transformation from punch card, lever, and paper systems is nearly complete. The numbers of registered voters who live in counties using Optical Scan voting systems has increased from 46.7 million or 29.5% to 84.0 million or 48.9%. While those registered voters living in counties served by direct recording electronic (DRE) voting systems have increased from 19.7 million or 12.4% to 65.9 million or 38.4% within 2 federal election cycles.

Technology that facilitates the right of citizens to participate in public elections may threaten privacy, especially when it is associated with the administration of elections and, under certain conditions, the very act of voting. The use of technology in the online and offline voting process is growing in popularity around the world. The Charter of Fundamental Rights of the European Union and the United Nations Universal Declaration of Human Rights' support the right of citizens to both privacy and self-governance. Democracies are universally defined as the most efficient means of supporting self-governance through citizen participation in the form of voting. The secret ballot has long been considered an integral requirement of democratic governance.

The balance between a state’s right to ensure that intimidation and election fraud are not present in public elections, and the voter’s right to privacy has resulted in the development of the secret ballot and restricted zones around voting compartments. Because of the documented history of voter intimidation, coercion, and fraud associated with third-party knowledge of how individual voters cast their ballots, it is important not to underestimate the importance of voter privacy.

Transparency is also an important value in the administration of public elections. It is not good enough to assure voters, media, and contestants in public elections that the process was fair. Transparency is a challenge in an environment where fully electronic voting systems are employed. Electronic voting systems are not new to the public election process they have been used for ballot tabulation for decades. The occasional loss of physical ballots due to the use of electronic vote tabulation systems was also a well-documented problem. However, what is new is the adoption of voting systems that record and tabulate votes on a single device. In this environment transparency is a serious challenge for election administrators to overcome. Manufacturers of voting systems claim proprietary rights that block public scrutiny of their products. The added twist to the new electronic voting landscape are the consequences of the occasional lost of ballots or voter selections. Instead of these incidents remaining hidden from public knowledge they are more readily made known by media reports. In addition, the demands for greater transparency post-2000 are placing election administrators and elections under unprecedented scrutiny.

In the November 7, 2006 election several instances of electronic voting machine failures marred the voting experience for voters in the states of Pennsylvania, Arkansas, Florida, Maryland and Virginia. The problems ranged from electronic poll-book failures to insufficient numbers of voting machines to serve polling locations. The most notable problem was the failure of the ES&S iVotronic DRE voting system, which resulted a 13% undervote in the race to fill the position for the 13^th Congressional District in Florida. What made this particular race a major challenge for electronic voting system is the 377-vote margin separating the top to candidates. It is estimated that 18,000 voters did not have their preference registered in that race due to a failure of the touch screen voting system.

This election also signaled greater interest in how changes in voting technology are impacting voters and election administrators. To learn more about the adoption of new voting systems Rice University and EPIC’s voting project conducted a survey of voters in Jefferson County Texas. This research project is thought to be the first of its kind to be conducted to determine the impact of recent changes in voting systems. The research effort looked at an election environment that used the Optical Scan and DRE voting systems. The research involved data on how long it took voters to vote on the two types of voting systems, and collection of voter opinions about their choice of voting system on Election Day. The analysis of the results of the surveys will take several weeks.

The introduction of new voting systems into the public election environment has had other unintended consequences. One of the issues receiving little attention in the move to modernize the US election process is the reliance of the system on an army of volunteer poll workers. Poll workers do receive some financial compensation for their time, but it is minimal in most cases. The addition of new voting systems does impact the necessary skill set that poll workers will need to function in the 21^st Century poll location. Compounding the issue of poll worker training is the average age of poll workers, which in 2004 was estimated to be 70 years of age. As a consequence, EPIC and Rice University are also involved in a follow up survey of Jefferson County, Texas poll workers who worked in the November 7, 2006 election to gather information on their experiences. The goal of this research will be to better inform county election administrators on the retention, and recruitment of Election Day workers. Like many areas faced with mass upgrades in technology they have seen a large number of poll workers indicating that they will not be returning to poll work in the coming year.

To learn more about EPIC’s voting work visit: http://www.epic.org/privacy/voting or http://votingintegrity.org/