The Boston Globe
Voting errors tallied nationwide
By Brian C. Mooney, Globe Staff
December 1, 2004
More than 4,000 votes vanished without a trace into a computer's overloaded memory in one North Carolina county, and about a hundred paper ballots were thrown out by mistake in another. In Texas, a county needed help from a laboratory in Canada to unlock the memory of a touch-screen machine and unearth five dozen votes.
In other places, machine undercounting or overcounting of votes was a problem. Several thousand votes were mistakenly double-counted in North Carolina, Ohio, Nebraska, and Washington state. Some votes in other areas were at first credited to the wrong candidates, with one Indiana county, by some quirk, misallocating several hundred votes for Democrats to Libertarians. In Florida, some machines temporarily indicated votes intended for challenger John F. Kerry were for President Bush, and vice versa.
In the month since the election, serious instances of voting machine problems or human errors in ballot counts have been documented in at least a dozen states, each involving from scores of ballots to as many as 12,000 votes, as in a North Carolina county. On Election Day, or in later reconciling tallies of ballots and voters, local officials discovered problems and corrected final counts. In some cases, the changes altered the outcomes of local races. But in North Carolina, the problems were so serious that the state may hold a rare second vote, redoing a contest for state agriculture commissioner decided by fewer votes than the number of ballots lost.
After the disputed vote in Florida four years ago, Congress passed the Help America Vote Act of 2002 and authorized $4 billion so states could create central computerized voter lists and replace outdated voting systems such as punchcards by 2006. But many states have not completed the overhaul, and this year's election unearthed enough problems -- both with older technologies and newer electronic touch-screens -- that two federal agencies plan unprecedented nationwide inquiries. The investigations by the Government Accountability Office and US Election Assistance Commission will begin early next year and be completed by mid-2005, at the earliest.
In addition, minor presidential candidates requested recounts in four states -- a partial one completed yesterday in New Hampshire, and statewide in Ohio, New Mexico, and Nevada.
None of the recounts or inquiries is expected to affect the results of the presidential election, which Bush won by more than 3.3 million votes.
Those who believe that either or both of the past two presidential elections were manipulated by a vague conspiracy to elect Bush have done statistical analyses of voting patterns in Florida and argued that the voting discrepancies were much larger and systemic, but their studies have not stood up to scrutiny from academics and other analysts.
Most of the concerns, which have rocketed through the Internet, center on computerized voting or tabulating machines, including some that do not keep a paper record for audits and recounts. Some computer scientists acknowledge that these systems could be vulnerable to tampering.
''I would hesitate to take seriously the conspiracy theories, but there are certainly gaps and vulnerabilities that have got to be addressed," said DeForest B. Soaries, chairman of the US Election Assistance Commission, which was created by the 2002 law and plans to conduct hearings around the country on the voting.
''We are convinced that while the election went relatively smoothly compared to what many had expected, that does not eliminate the need to study the results and collect data to document machine malfunctions and other administrative matters," Soaries said.
Since 2000, watchdog groups have intensified their monitoring and cataloging of complaints and errors. The nonpartisan Verified Voting Foundation and other groups built a database of more than 30,000 ''election incidents" reported across the country this year. Most were routine, but nearly 900 involved significant e-voting problems, including malfunctions that shut down machines, lengthening waits at the polls. There were 42 reports of total breakdowns of machines in New Orleans and 28 in Philadelphia and ''15 reports of catastrophic machine failure" in Mercer County, Pa.
The most serious problems occurred in North Carolina, where 4,438 e-votes disappeared in Carteret County. In at least five other counties, major double-counting or undercounting was discovered and corrected by North Carolina officials during their tabulations.
Johnnie McLean, deputy director of the State Board of Elections, attributed many mistakes to ''the human element, brought on by fatigue." In Carteret, for example, election workers apparently did not notice the ''Voter Log Full" message on the black box as the UniLect touch-screen failed to record the electronic votes, she said.
''If we had problems in the past, they were not magnified like this," McLean said, referring not only to the closeness of the statewide race, but also the extraordinary scrutiny of voting since 2000.
Examples of other major problems that were reported on and after Nov. 2, then later corrected, include:
Thousands of ballots were mistakenly double-counted in Sandusky County, Ohio; Sarpy County, Neb.; and Grays Harbor County, Wash. Democrats in Washington must decide by Friday whether to seek a second recount in the closest governor's race in state history. One recount has been held, reducing Republican Dino Rossi's 261-vote lead to 42 votes over Democrat Christine Gregoire.
In Gahanna, Ohio, a suburb of Columbus, Bush was temporarily credited with 3,893 more votes than he actually received in a precinct where only 638 voters cast ballots on a Danaher electronic voting machine.
In Franklin County, Ind., a tabulator credited about 600 straight-ticket Democratic votes to Libertarian Party candidates.
In Collin County, Texas, the memory card of a Diebold Election Systems touch-screen machine had to be sent to a laboratory in Canada a week after the election to extract information about 63 votes cast before the machine froze and was taken out of service.
In South Florida's Broward County, ''multiple misrecordings" occurred when votes for Kerry on touch-screens made by Election Systems & Software Inc. appeared as Bush votes, and there was at least one account of a Bush vote going to Kerry, the Verified Voting group reported. Broward voters discovered the problem on screens that allowed them to check their selections before entering them electronically.
The long list of documented problems has fueled the suspicions of conspiracy theorists, activists, and the minor presidential candidates who requested the New Hampshire and Ohio recounts.
Ohio decided the 2004 contest, but since the 2000 election, Florida remains the focus of the doubters and the devotees of various scenarios that suggest skullduggery, in part because early exit polls overstated Kerry's strength.
No group has been more aggressive than Seattle-based Black Box Voting, which bills itself as ''consumer protection for elections." Led by founder Bev Harris, the organization is seeking election records from around the country for audits of the results. The primary focus is Florida, where internal computer records have been requested in all 67 counties, and the results in glitch-plagued Volusia County, in the east-central part of the state, are being contested.
Four years ago, during vote-counting on election night, a faulty memory card initially deducted 16,022 votes from Democrat Al Gore's vote total in Volusia. Despite spending about $300,000 to upgrade equipment and avert a repeat, there were memory card problems this year in tabulators for six Volusia precincts. The optically-scanned paper ballots were re-fed into other counting machines to reach an accurate tally, a county election official said.
''All day long, I get desperate calls from people who are in so much pain," said Harris, the Black Box founder, who said she is convinced fraud occurred in some places Nov. 2. ''They say: Can you fix it? Can you solve it? Can you turn around the presidential election? We're not trying to turn the election around. We're trying to get elections to be more transparent, because with the new machines, it's not transparent."
Deanie Lowe, Volusia's supervisor of elections, said she has complied with Harris's record requests and offered to recount, free of charge, any three of Volusia's 179 precincts selected by Harris.
Harris, however, said records for all precincts were not turned over, and Black Box will seek a 50-precinct recount in the county, which Kerry won but by a smaller margin than Gore did in 2000.
Much of the postelection focus on Florida resulted from a pair of analyses that claimed Bush's vote totals in the state were inflated by two vote-counting technologies.
The first analysis originated on the Web and cited results in rural, overwhelmingly Democratic counties in Florida's panhandle, where Bush crushed Kerry. All use optical scanners. What the analysis failed to note is that Bush routed Gore by nearly equal margins four years ago in the same conservative counties that have been tilting Republican for years in national elections. A team of Miami Herald reporters reviewed 17,000 ballots in three of the counties, basically confirming the election results.
Then, a broadly reported second study by a team at the University of California at Berkeley, using an academic statistical method, asserted that ''Irregularities associated with electronic voting machines may have awarded 130,000 excess votes to President George W. Bush in Florida." In Broward County alone, the study said, Bush ''appears to have received approximately 72,000 excess votes." Palm Beach and Miami-Dade counties, which also use touch-screens, were also cited as anomalies.
But if Bush had actually received 72,000 fewer votes in heavily Democratic Broward, his total this year would have been less than it was in 2000 -- even though nearly 132,000 more ballots were cast. Kerry won all three key counties, Broward by more than 209,000 votes.
Bush carried the state by 380,978 votes, or about 5 percent of 7.6 million cast.
If recounts are the skeptics' best hope to uncover systemic irregularities, they got off to a rocky start in New Hampshire. Completed yesterday at the request of independent candidate Ralph Nader, the Granite State re-tally of 50,600 votes in 11 towns and city wards that use optical scanners increased Kerry's total by 87 votes and Bush's by 62.
Secretary of State William M. Gardner said scanned ballots have worked well in New Hampshire. Indeed, the largest discrepancies found this year were in a legislative race involving hand-counted ballots, he said.
On deck is Ohio, which tipped the Electoral College to Bush. At the earliest, the recount of 88 counties won't begin until Dec. 13, according to the secretary of state's office, the same day the Electoral College is scheduled to formalize Bush's reelection. The recount could take anywhere from a few days to a few weeks. Minor presidential candidates Michael Badnarik and David Cobb of the Libertarian and Green parties, respectively, have said they would formally seek the recount once the state certifies the official results Monday. They will incorporate the review of 155,000 provisional ballots, which were not included in preliminary tallies that showed Bush winning Ohio by 136,483 votes, or about 2.5 percent.
How long the recount takes will depend on whether Badnarik and Cobb ask for a manual inspection of any or all of the 5.5 million ballots, said Carlo LoParo, spokesman for Ohio's secretary of state, J. Kenneth Blackwell.
Of complaints about long lines that discouraged some from voting and allegations that there was a shortage of machines in some urban Democratic areas, LoParo said such decisions in Ohio are made by county boards of elections with two Republicans and two Democrats.
Long waits in Ohio and elsewhere resulted from the system being overwhelmed by a high turnout, said Doug Chapin, director of the nonpartisan electionline.org, which monitors reform efforts.
More attention should be paid to providing an adequate number of machines in polling places, he said, as well as ''finishing the job" mandated by the Help America Vote Act. Forty states, for example, have yet to comply with a mandate to establish a central, statewide database of registered voters. That will reduce questions about voter eligibility at election time, Chapin said.
Whatever the outcome of the recounts and the official inquiries by federal agencies, the impetus for improve voting systems will not fade, he said.
''This is not a fringe issue, because a sizable group is interested in pursuing this as a policy issue going forward," Chapin said. ''There's now a critical mass of people involved who want to address the problems that occurred in 2004. This issue is not going to go away."