Optical Scan Systems Not Substantially Better Than Punch Card in Assuring Accurate Vote Count - Data Underscores Need to Upgrade to Modern Electronic Systems, Cox Says

Secretary Cox: 2000 Election Data Shows Undervote Rates Consistent Among All Voting Equipment

ATLANTA … Georgia Secretary of State Cathy Cox said today that a new statistical analysis of the November 7th, 2000 election results shows consistent rates of “undervotes” regardless of the type of voting equipment used, and none of the three major systems employed in Georgia counties offers a significant advantage in assuring a full and accurate tally of votes.

Almost 95,000 Georgians who went to the polls either chose not to vote for president or, by error, lost the right to have their vote counted. “I find it difficult if not impossible to believe that so many citizens chose not to make a selection in the most important race on the ballot,” said Secretary of State Cox. “One cannot help but conclude, then, that the votes of thousands of Georgians who thought they had made a choice were not counted because of confusion, error, or equipment failure,” Ms. Cox added.

Ms. Cox said the data underscores her view that Georgia should move toward the adoption of a uniform system of state-of-the-art electronic voting equipment to reduce voter error and add convenience and speed to the vote counting process.

The Secretary of State’s study found that undervotes (the difference between the number of actual ballots cast and the number of votes recorded in final certified results) in the presidential race showed a statewide mean average of 4.4 %. The mean averages of counties aggregated by voting equipment are: Punch Card: 4.6%, Optical Scan: 4.5% and Lever Machine: 4.2%.

Undervotes may be generated in three different ways: 1) the voter decided not to make a choice in that particular race, 2) the voter attempted to make a choice but it was not read or computed by machines used in the vote tallying process, or 3) the voter inadvertently chose more than one candidate in a single race, creating an “overvote” which voided their choice in that race.

“This new study is the first time Georgia has taken a close look at undervotes and examined these variations by the type of voting equipment used at the county level,” said Secretary Cox. “We think the data makes very clear that the optical scan system, while enjoying some advantages over even more antiquated methods, is not the answer if we want the most convenient, user-friendly and accurate system for future Georgia elections.” Secretary Cox said that while some optical scan counties like Cobb and Gwinnett showed very low undervote totals, seven counties using that equipment tallied undervote percentages of greater than 7 %, and one county’s undervote in the presidential race was 18.8 % of ballots cast.

Only 13 of 159 Georgia counties report overvotes (where the voter chose more than one candidate in a single race) as a separate category, and all of those counties utilize optical scan ballots. In 12 of those 13 counties, overvotes represent a very substantial majority of the total difference between ballots cast and votes counted in the presidential race. The data would seem to suggest that, at least in counties that use the optical scan system, most undervotes are not the result of voters simply “passing up” the presidential race, but rather are generated by inadvertent voter error or poor ballot or system designs.

In Carroll County, for example, 82.5 % of the variation between ballots cast and votes recorded in the presidential race were overvotes. In Dougherty County, that number was 88.7 %, in Douglas County 85.3 % and in Walton County 84.9 %.

Ms. Cox said that voter confusion probably accounts for a substantial number of the errors made on optical scan ballots. Individual inspection of these ballots shows some with an “x” marked by a candidate, the candidate’s name circled or the candidate’s name written in by hand. None of these marks would be read by counting machines.

There are also indications from county election officials that voters have particular difficulty with one type of optical scan ballot that requires electors to complete an arrow adjacent to a candidate’s name (rather than blackening a circle). “Minimizing the opportunity for voter error and confusion should be a critical goal in any changes we adopt in voting technology,” said Secretary Cox. “While ‘hanging chads’ have received a great deal of attention as a result of the Florida recount experience, it is clear that there are a host of other issues we need to address to make sure that every voter, whatever their age or education, ability or disability, can clearly and easily express their choice in the voting booth. For those who suggest that optical scan ballots are a solution to the obvious deficiencies of the punch card system, I believe the numbers tell a different story.

This study is yet one more reason my office will seek funding from the General Assembly in the upcoming session for a field test of new electronic voting equipment, with the goal of full implementation of a uniform statewide system by 2004,” she added.

Ms. Cox said her office is also nearing completion of a report to be provided to the Governor and members of the General Assembly documenting complaints and problems in the 2000 elections and offering proposed solutions. The report will address voting equipment, staffing, policies and procedures and statutory improvements. 

Editors Note: A spreadsheet with undervote data for 157 Georgia counties is attached. Two Georgia counties that employ paper ballots are not included in this report.

 Press Office ~ 110 State Capitol ~ Atlanta, GA 30334 ~ 404.656.4269 FAX: 404.657.5804