In a story similar to the 2000 presidential election, Florida's 2002 primary election was dogged by problems. This time, however, touchscreen machines were at the center of the controversy.
September 19, 2002
Two years after the world was transfixed by hanging chads and a presidential race which never seemed to end, election officials in Florida were hoping for a nice, peaceful primary on Sept. 10.
But the state that was left awash in controversy and calls for election reform after the Bush-Gore presidential election of 2000 was in the spotlight again after it took a week to reach a resolution to its 2002 gubernatorial primary.
A different election brought new kinds of upheaval. This time, however, the blame was not aimed at paper ballots with confusing directions and tiny, detachable chads. Instead, much of the fracas focused on touchscreen ballots.
When the polls were supposed to open the morning of Sept. 10, many remained dark. Others might have well remained so, because a lack of training or failure to collect the proper activation equipment left the electronic balloting equipment shuttered.
"It's shameful," Florida Gov. Jeb Bush told The Associated Presson election morning. "The state put up money -- significant sums of money -- for training, for machines. There's no excuse for not having precinct workers in a precinct for voting, no excuse for not turning on the machines."
It took seven days before former U.S. Attorney General Janet Reno conceded the Democratic nomination for governor to opponent Bill McBride. Much of that stretch was spent counting ballots in the South Florida counties of Miami-Dade and Broward, which were also at the center of the 2000 election controversy.
The election's aftermath is awash in conjecture over what went wrong. To some, touchscreen balloting is a bad idea, fraught with potential abuses and inaccuracies. Others pin the blame on a lack of training and a fear factor that enveloped too many election workers.
About the only aspect of the election most people can agree on is that further refinements are needed to prevent what happened in Florida from happening anywhere else in the U.S.
"We're in a bad situation because a lot of states have invested tens of millions in these machines and they don't meet the necessary standards," said voting consultant Rebecca Mercuri. "Hopefully people will look at Florida as the wake-up call."
Fixing a problem, but Â…
Florida reportedly spent more than $30 million on touchscreen ballots following the 2000 presidential election. The goal was to eliminate confusing ballots, such as the butterfly ballots that confused many South Florida residents who could not tell if they were voting for Democrat Al Gore or Reform Party candidate Pat Buchanan.
![]() |
Election Systems & Software's iVotronic touchscreen ballot was used in 32 Florida counties during the state's primary election on Sept. 10, including Miami-Dade and Broward. |
One company, Omaha, Neb.-based Election Systems & Software Inc. (ES&S) supplied 22,440 of its iVotronic machines for use in 32 of Florida's 67 counties, including Miami-Dade and Broward.
The company, whose slogan is "Better Elections Every Day," touts the iVotronic's size (9 ¾ pounds and one square foot), audio capabilities, and ease of use as the machine's strengths.
"Opening and closing the polls requires only two steps by the poll workers," according to ES&S's iVotronic promotional material. "Voting is accomplished simply by the touch of the voter's finger on the ballot screen."
But the machines proved problematic from the start in Broward and Miami-Dade counties. Inadequate training and a lack of familiarity with the equipment appeared to be at fault. Lori Nance Parrish, chairwoman of the Broward County Board of Commissioners, told the Washington Post that poll workers at 50 polling stations failed to pick up the devices needed to turn on the machines the night before the election.
It became apparent at the beginning of the day that there would be problems. Reno waited more than 20 minutes to vote as she and reporters watched precinct workers struggle to activate the machines.
Problems continued throughout the day, as poll workers discovered they did not know how to change the machine's batteries or download votes after the polls closed. Various reports said that poll workers, frustrated over the new machines, failed to show up on election day or left before the balloting was completed. At least 600 voters left the polls without casting ballots, according to the AP.
The age factor
Stephen Ansolabehere, a political science professor at the Massachusetts Institute of Technology, studied the 2000 Florida balloting for the Caltech-MIT Voting Technology Project. After witnessing what happened in Florida this year, he concluded that a lack of training frustrated elderly poll workers. A similar situation occurred in New Mexico two years ago, he told the AP.
"A lot of poll workers quit," Ansolabehere said. "They couldn't learn how to use these new machines. They were intimidated and frustrated by them."
But some election workers said blame the technology, not the workers. Bill McKamy, an 80-year-old poll worker in Miami-Dade County, gave the AP a different slant on the day's activities. McKamy said he was forced to hang on to his precinct's ballots for 16 hours after the election because his younger co-workers got fed up and quit.
"If you think it's loused up with elderly people, try young kids," McKamy said.
A retired World War II veteran, McKamy said a sense of duty compelled him to remain at the polls, which were ordered to stay open an extra two hours by Gov. Bush because of the voting problems.
"I'm a military man," he said. "If you're put on a post, and you say you're going to do it, do it."
Training camp
ES&'s spokesman Todd Urosevich said the company has taken a hard look at what happened in Florida and has come up with one conclusion: training must be emphasized.
The company was not contracted to train poll workers, according to Urosevich, though that service is available. Instead, ES&'s trained county staff members, who then led poll worker training.
"Not everybody learns at the same pace. You can't get a cross-section of America, throw them in the same room, and expect them to all come out at the same level of competency. You've got to train them." Todd Urosevich |
"The lesson we've learned coming out of this is that when we're not actually contracted to conduct poll worker training, we'll need to motivate county administrators to do all that they can to carry out full, complete poll worker training," Urosevich said.
"A touchscreen system has new responsibilities for poll workers, as opposed to the punch-card systems," he added. "I don't know that they're increased responsibilities; they're just different."
The training needs to be detailed, but instructors need to be cognizant of the learning abilities of each poll worker, Urosevich added.
"Not everybody learns at the same pace," he said. "You can't get a cross-section of America, throw them in the same room, and expect them to all come out at the same level of competency. You've got to train them."
Defending or debunking
According to a National Association of State Election Directors chart dated July 8, 33 different touchscreen balloting systems are approved for use by that organization. That includes four machines manufactured by Diebold Election Systems Inc., a subsidiary of kiosk and ATM manufacturer Diebold Inc., and six ES&'s models, including the iVotronic.
To Rebecca Mercuri, the founder of Philadelphia-based computer consulting firm Notable Software Inc., a glut of machines is part of the problem.
"What I've seen in the last year is a lot of products rushed to the market without security controls, safeguards, and usability testing," she said. "The products, particularly in the voting arena, are being designed to 1990 standards that are hopelessly out-of-date to modern technical standards."
ES&'s has been in business for 36 years, and the company said more than 100 million ballots in each of the last four presidential and congressional elections were cast on ES&'s equipment. Michael Limas, ES&'s executive vice president and chief operating officer, was quick to defend the iVotronic the day after the Florida election.
"The iVotronic equipment manufactured by ES&'s used by the Boards of Election in Miami-Dade and Broward counties accurately captured 100 percent of the votes which were cast," Limas said in a prepared statement. "No votes were lost or not counted."
But to Mercuri, the problem is not whether the touchscreen ballots worked or not, but the entire concept of touchscreens as an electoral tool. She is strongly opposed to electronic balloting for a variety of factors, including:
-- Failure to verify that the ballot is being properly cast;
-- No individual print-outs for voters to scan, making it difficult to perform an independent audit;
-- A lack of certification to international computer security standards; and
-- Encryption problems that could allow hackers to check ballots, particularly on Internet-based balloting.
She believes that one way to prevent problems in the future is to provide voters with a paper ballot when they are finished voting -- the electoral equivalent of an ATM receipt.
"The paper can be inspected by the voter to make sure their vote counted," Mercuri said, "and it then can be turned in and used in case of a recount. This is recognition that any computer is unstable and can lose data; this is a problem. Voters need to be able to see whether their vote counted or not.
"It's really important to have this paper audit trail," she added. "In banking, you get a receipt, there's a camera, and it's traceable to an individual. There's a number of ways to audit it and trace it back to the people that are involved."
What's next?
For the touchscreen balloting industry, the next proving ground is Georgia. Florida's neighbor to the north spent even more money, $54 million, to install 19,015 AccuVote-TS touchscreen ballots manufactured by Diebold Election Systems. Unlike Florida, the machines have been installed statewide. And unlike Florida, Georgia is preparing for a general election without giving the machines a test run in the primaries.
"It's really important to have this paper audit trail. In banking, you get a receipt, there's a camera, and it's traceable to an individual. There's a number of ways to audit it and trace it back to the people that are involved." Rebecca Mercuri |
"I think Georgia is making a mistake in rolling out new equipment on the general election day without having had it in place for the primary to test it and for people to get used to it," Republican Eric Johnson, Georgia state senate minority leader, told the AP.
But Chris Riggall, a spokesman for Georgia Secretary of State Cathy Cox, told the AP that the new system would work and gain acceptance.
"When you've got one system in all the counties and media markets in all regions of the state, you are able to speak with one voice at one time to everybody: This is how you are going to vote," Riggall said.
Whether Georgia's voters respond with plaudits or pot shots remains to be seen.