Ohio's Diebold Debacle: New machines call election results into question

The Free Press: Speaking Truth to Power          www.freepress.org

by Bob Fitrakis & Harvey Wasserman
November 24, 2005

Massive Election Day irregularities are emerging in reports from all over Ohio after the introduction of Diebold's electronic voting in nearly half of the Buckeye State’s counties. A recently released report by the non-partisan General Accountability Office warned of such problems with electronic voting machines.

E-voting machine disasters

Prior to the 2005 election, electronic voting machines from Diebold and other Republican voting machine manufacturers were newly installed in 41 of Ohio’s 88 counties. The Dayton Daily News reported that in Montgomery County, for example, “Some machines began registering votes for the wrong item when voters touched the screen correctly. Those machines had lost their calibration during shipping or installation and had to be recalibrated. . . .”

Steve Harsman, the Director of the Montgomery County Board of Elections (BOE), told the Daily News that the recalibration could be done on site, but poll workers had never performed the task before.

The city of Carlisle, Ohio announced on November 22 that it is contesting the results of the November 8 general election as a result of Montgomery County vote counting problems. Carlisle Mayor Jerry Ellender told the Middletown Journal that the count on the city’s continuing $3.8 million replacement fire levy is invalid “since they are not sure if Carlisle voters received the right ballots on the new electronic voting machines.”

Harsman, according to the Journal, said, “poll workers incorrectly encoded voter cards that are used to bring up the ballots on the electronic machines in precincts in Germantown and Carlisle.”

At least 225 votes were registered for the fire levy in precincts with only 148 registered voters, according to the Journal. In addition, 187 voting machine memory cards were lost for most of election night in Montgomery County, according to the Dayton Daily News.

In Lucas County, election results appeared more than 13 hours after the close of polls. The Toledo Blade cited “‘frightened’ poll workers,” intimidated by the new “touch-screen voting machines.”

The Blade found that despite an $87,568 federal grant to the Lucas County Board of Elections for “voter education and poll worker training . . .” only $1,718.65 was spent from the grant.

The Blade also reported that ten days after the 2005 election, “Fourteen touch-screen voting machines have sat unattended in the central hallway at the University of Toledo Scott Park Campus.” The GAO report warned that touch-screen machines are easily hacked and should be kept secure at all times.

In Miami County, the Board of Elections fired the Deputy Director, Diane Miley, following a 20-minute closed-door session reviewing the November 8, 2005 general election.

The Free Press had reported that in the 2004 presidential election, Miami County was cited in the seminal Moss v. Bush election challenge case. The county was specifically cited for an early morning influx of 19,000 additional votes, mostly for Bush, after 100% of the vote had been reported.

The AP reported additional irregularities in the 2005 election in Ohio. In Wood County, election results were not posted until 6:23 a.m. after poll workers at four polling places accidentally selected the wrong option on voting machines preventing the machine memory cards from being automatically uploaded, according to the Board of Elections Deputy Director Debbie Hazard.

In five counties – Brown, Crawford, Jackson, Jefferson and Marion – using Diebold machines, there were problems with the counting of absentee ballots as a result of “the width of the ballot,” the AP reported.

In Scioto County, the vote count was not finished until 4:30 a.m.. Board of Elections Director Steve Mowery informed the Portsmouth Daily Times that, as a result of machines undergoing insufficient testing and absentee problems, things went “poorly.”

Many counties used “roving employees” assigned to pick up memory cards from voting machines. In Lucas County, these “rovers” traveled “to multiple locations before delivering the cards to the election office at Governmental Center.” The polls closed at 7:30 p.m. but, “The final memory cards were delivered to the Board of Elections office just before midnight,” according to WTOL Channel 11 News, Toledo.

Toledo’s WTOL Channel 11 News posed the simple question: “Did the delay in returning memory cards to the election office open the door to possible vote fraud?”

Amidst these massive glitches, Ohio Secretary of State J. Kenneth Blackwell, who personally negotiated the deal for the Diebold machines that he called the “best in the nation,” insisted through his spokesperson Carlo LaParo that “The new touch-screen systems went well.”

Odd results for election reform initiatives

The Reform Ohio Now (RON) campaign saw polls throughout the state showing two of its four election reform to be passing easily. Both the Columbus Dispatch and University of Akron Bliss Institute polls predicted victories for Issue 2 and Issue 3, only to see them go down to sudden and statistically unexplainable defeat. Issue 2 allowed for early voting in Ohio and Issue 3 reduced the amount of money an individual can give a candidate from $10,000 to $2,000. Both were predicted to pass with 59% and 61% of the vote, respectively.

The Bliss Institute of Applied Politics’ survey was completed on October 20 at the University of Akron Survey Research Center, and found that Issue 2 seemed likely to win approval with more than three-fifths of likely voters.

The Dispatch mail-in poll was completed on Thursday Nov. 3, just prior to Election Day. The Dispatch poll is so accurate, that at least two academic studies have been published about it in the Public Opinion Quarterly (POQ). The first paper documents that the Dispatch poll between 1980-1984 was far more accurate than telephone polling. The study showed the Dispatch error rate at only 1.6 percentage points versus phone error rates of 5%. A companion study published in POQ in 2000 dealt specifically with the question of statewide referenda. A quote from the study: "The average error for the Dispatch forecast of these referenda was 5.4 percentage points, compared to 7.2 percentage points for the telephone surveys."

The academic study concluded that the Dispatch's mail survey outperformed telephone surveys for both referenda and candidate's races.

The fact that the Dispatch was nearly 30 points off in predicting the "YES" vote on Issue 3 should raise concerns.

Dispatch Associate Publisher Mike Curtin shrugged off the worst polling performance since the infamous Literary Digest predicted that Alf Landon would beat FDR in 1936. In an email obtained by the Free Press, Curtin told California voting rights activist Sheri Myers, “There is no evidence of any irregularities in Ohio’s 2005 voting results.” Curtin, according to election attorney Cliff Arnebeck, had also dismissed anyone who raises issues about Ohio’s 2004 presidential election results as “conspiracy theorists.”

Curtin co-authored the scholarly papers on the Dispatch’s legendary polling accuracy. Editorially, the Dispatch has not endorsed a Democratic presidential candidate since Woodrow Wilson in 1916.

Curtin pleaded with the voting rights activists, “Please don’t buy into the conspiracy theories without any shred of evidence.” Curtin did not deal with the specifics about how the polling, which he was so proud of, was up to 40 points off on certain issues for the first time ever. In another email explaining the unprecedented Dispatch polling debacle, Dispatch Editor Darrel Rowland told a Tribune Media Services columnist that, “I also can’t imagine voting technology is to blame, when both Democrats and Republicans are involved in every crucial step of the way.”

Under oath testimony at public hearings sponsored by the Free Press after the 2004 presidential election revealed that election workers admit that they have little or no knowledge of how e-voting technology works and are totally reliant on private vendors for vote counting inside the “black box.” Ohio’s other major newspapers routinely suggest what Rowland “can’t imagine.”

Rowland did note that despite the Dispatch’s recent embracing of its unprecedented incompetence at polling that, “Over the years we have found that the people who return our mail poll are likely voters – the holy grail in political polls. Our track record in gauging public opinion in this state regarded as a national political bellwether is unparalleled.

Don McTigue, the attorney for RON, told the Free Press that Blackwell had issued a ruling barring RON volunteers from the county vote counting rooms on election eve. McTigue and the RON volunteers had filled out a request form to view the counting eleven days prior to Election Day, but Blackwell had added a new form to verify which group was representing the issues. This new form was not filled out, McTigue admits.

Matt Damschroder, the Franklin County Board of Elections Director, allowed the RON observers in anyway, despite their being barred from the vote counting rooms in other counties.

This is the second straight election in which the polling organizations were spectacularly wrong in Ohio. In the 2004 election, the media consortium exit polls, as well as the Harris and Zogby polls, all declared Kerry the winner on Election Day.

Democracy in jeopardy

One of the first times electronic voting machines were used, in the 1988 New Hampshire presidential primary, former CIA director George Herbert Walker Bush pulled off a stunning and unpredicted upset. The last poll before that primary showed Senator Bob Dole winning with 8 percentage points. Bush won by 9 points, a startling 17-point shift. Bush’s e-voting victory allowed him to claim the White House and paved the way for his son to become the United States’ chief executive.

Diebold electronic voting machines use non-transparent, proprietary software to count the votes. Diebold’s CEO Wally O’Dell is one of President Bush’s major donors and fundraisers.

Election Day news coverage from the 41 counties that adopted Diebold touch-screen machines makes it clear that poll worker ignorance about how to use the high-tech equipment and machine glitches were widespread problems in 2005. Diebold technicians in many areas were key in producing the final vote results.

Use of e-voting machines has resulted in two elections with improbable results in Ohio, with potentially catastrophic outcomes for American democracy – especially if they are ignored.

--
Bob Fitrakis & Harvey Wasserman are co-authors of HOW THE GOP STOLE AMERICA’S 2004 ELECTION & IS RIGGING 2008, available at http://www.freepress.org/ and at http://www.harveywasserman.com/, and, with Steve Rosenfeld, of WHAT HAPPENED IN OHIO?, to be published by the New Press in 2006.