|
<advertisement>
|
 |
 
|
|
media search: |
 |
January/February 2001 | Contents
VOTING TECHNOLOGY: WHO KNEW?
BY STEPHEN J. SIMURDA
Watching
the Florida ballot recount, Philip W. Johnston found himself quoting Yogi
Berra. "It's déjà vu all over again," he said. Johnston is chairman
of the Massachusetts Democratic Party and a man who won, and then lost, a
chance to run for Congress in 1996 because of a legal challenge over the same
type of punch-card ballots that were disputed in Florida.
Johnston seemed to have won the Democratic primary that year in the Bay State's 10th Congressional District by 266 votes. A partial recount left him the winner, but by just 175 votes. Then his opponent challenged 965 punch-card ballots that had not registered a vote. A Superior Court judge spent several hours examining each ballot and concluded that Johnston had actually lost by 108, out of a total of more than 49,000 ballots cast. "It was very clear that she counted anything" on the ballot, Johnston says, including the now-famous pregnant chads. The state's Supreme Judicial Court upheld the ruling.
That election drama got a lot of coverage in the state's press, but none of the region's big newspapers picked up the larger problem of voting technology. The mess did prompt some action: the secretary of state decided to decertify punch-card ballots in Massachusetts, and a revolving loan fund was created to help communities buy new election technology.
But that's Massachusetts. Across the nation more than a third of voters in Election 2000 used punch-card ballots of some type. It remains the most widely used voting technology nationwide, despite a history of serious problems. Which raises some questions for journalists:
Why wasn't that history more widely known before it became so painfully apparent? Did we miss a rather large story? Can the press help push reforms now?
To people who know about how ballots are physically cast, the disputes in Florida were not surprising. "It's been a big problem for a while," says Roy G. Saltman, a computer scientist who conducted two studies on voting technology while working at the federal government's National Institute of Standards and Technology. His most recent study was released in 1988, and it "called for the end of punch-card technology."
Another 1988 study, by ECRI Laboratories, also advised against punch-card technology. Both studies specifically criticized the most popular punch-card system, called Votomatic. "I think there are very few election officials who have lived with Votomatic for any time who would defend it," says Malin VanAntwerp, who was project director for the second study. He estimates that 30 percent of American voters use Votomatic machines, making it the single most popular voting technology. The study went to every elections office in the nation, but it got virtually zero press coverage.
Here's a partial history of Votomatic's troubles:
* In 1968, San Francisco miscounted 13,000 punch-card ballots in the general election. The San Francisco Chronicle called it "astonishing, alarming, preposterous, and confidence-shaking but, unfortunately, not incredible."
* In 1969, International Business Machines, which introduced Votomatic five years earlier, decided to get out of the business because of problems with the technology. A group of IBM managers left the company to manufacture and sell the technology independently.
* In 1980, a programming error in Orange County, California, caused 15,000 votes for Jimmy Carter or Edward M. Kennedy to be counted for Lyndon LaRouche or Edmund G. Brown Jr.
* In Carroll County, Maryland, sloppy testing in 1984 meant that 13,000 votes were missed. When they were counted, at least one race was reversed.
* In Oklahoma County, Oklahoma, in 1986, some machines failed to count up to 10 percent of the ballots cast.
* In 1988, Republican Connie Mack won a close Senate race in Florida. Oddly, there were almost 200,000 punch-card ballots on which people had voted in the presidential race but not in the hard-fought Senate campaign. Mack's opponent sought a manual recount in five counties, but was denied in all but one.
* In a 1993 special election in Wisconsin, 4 percent of the people who came to the polls and used punch-card ballots to fill a vacant congressional seat failed to register a vote, despite the fact that there was only one race on the ballot. The state's election chief later said that it was common for 5 percent to 10 percent of punch-card ballots to be rejected, which the Wisconsin State Journal called "an unacceptable rate of disenfranchisement."
* In 1996, election officials in Utah's Salt Lake County spent fifteen hours examining nearly 300,000 ballots for loose chads before recounting them, after noticing that some ballots were not completely punched through.
* In 1998, Senator Harry Reid of Nevada was reelected by just 428 votes in an election marred by problems, notably a decision by one county to save money by printing its own ballots. The ballots were misprinted slightly and the vote counting machines could not read them.
| BEYOND PREGNANT CHADS |
|
For news organizations that plan to examine the problems of voting technology in their own areas, Stephen J. Simurda asked his sources to suggest stories that could be pursued in the wake of the Florida fiasco. Here are some questions worth asking:
|
These stories got local coverage centered on specific incidents. But the press had trouble connecting the dots and seeing a larger problem.
There were two notable exceptions. One was a page-one piece in The New York Times on July 29, 1985, which noted that lawsuits filed in three states charged that computerized punch-card ballot systems were "very vulnerable to manipulation and fraud." That story caught the attention of the Markle Foundation in New York, which subsequently supported both of the 1988 studies.
The other exception was a thorough three-part series in the Los Angeles Times in July 1989, which was highly critical of punch-card ballots. William Trombley, who now works as an editor for a non-profit educational organization, wrote that series.
"I went to a lot of places doing those stories," Trombley recalls, "and many of the worst examples I found came from Florida. The machines were old and in bad shape, and a lot of the voting officials seemed pretty incompetent."
Despite his highly critical series, Trombley says he was unable to get political leaders worked up about the issue. "I couldn't get either political party to take it seriously," he says. Trombley also noticed that no other news organization has devoted much attention to the issue in the eleven years since his series.
Saltman agrees, and adds, "If the media don't look at it, nobody looks at it." But he admits the story has an inherent flaw. "I think it's just not sexy enough. Nothing is done until there's a crisis."
But even Philip Johnston, who may have lost an election to bad voting-machine technology, doesn't blame the press for overlooking the issue. "It is the responsibility of people in politics to get this out, not necessarily the press," he says.
Still, journalists can perform a public service by informing people about voting problems and prodding state and federal lawmakers to act. The experts we spoke with stressed that journalists need to look at more than just the problem of pregnant chads. In a 1998 study of voting systems, professor Susan King Roth of Ohio State University asked thirty-two subjects to participate in simulated elections using punch-card ballots. The majority said they disliked the system after voting, complaining that the type was too small and that the layout was confusing.
More troubling, even though each person was given instructions on exactly how to vote in each simulated race, only thirteen of the subjects turned in error-free ballots. Several made multiple errors, particularly elderly participants. Roth concluded that some of these voters were being disenfranchised because of the design of the punch-card ballot system.
Then there is the question of the competency of the people overseeing elections. R. Doug Lewis, executive director of The Election Center, a nonprofit and nonpartisan organization that works with election administrators, notes that resources are a big part of that problem. "About twenty-five percent of election offices in the United States are about as well funded as any other part of government," he says. "The other seventy-five percent are among the lowest-funded." And that means not enough money to train and hire adequate staff, which can lead to poor maintenance of machines, programming errors, and mishandling of ballots -- not to mention making the purchase of newer technology out of the question.
Roy Saltman is appalled that the many problems of punch-card balloting haven't drawn more attention. It is inexcusable, he says, that between 1 and 2 percent of voters in some Florida counties did not register a vote for president. "Normally one-tenth of one percent would be more like it," he says. "This is a very low-performance system."
And perhaps one that more reporters and editors should have seen coming. In the first part of Trombley's series in the Los Angeles Times, he concluded by quoting Kimball W. Brace, who was, and still is, president of Election Data Services Inc. in Washington, D.C. The interview was nearly a dozen years ago, but Brace was prescient: "We're waiting for a volcano to erupt, in the form of a major election scandal . . . . We know it's going to happen, but we don't know when or how we're going to handle it."
Stephen J. Simurda teaches journalism at the University of Massachusetts at Amherst, and is a frequent contributor to CJR.
