383,000 Missing Votes in California Recall

383,000 Missing Votes in California Recall
Punch Card Ballots Blamed
By RACHEL KONRAD, AP
10/10/03 04:45 EDT

SAN FRANCISCO (Oct. 10) - More than 380,000 ballots cast in the recall election did not have a valid vote on whether to recall Gov. Gray Davis, and most of them were made on punch card systems, according to two independent studies.

Even if the 4.6 percent of Californians whose ballots did not answer the recall question had voted against it, Davis would have lost. The recall passed by a margin of 10.8 percent, and Republican actor Arnold Schwarzenegger enjoyed a comfortable victory.

But California's anomalies could resonate nationwide, as counties scramble to modernize election equipment to qualify for federal funding in the 2002 Help America Vote Act.

In Los Angeles County, nearly 9 percent of people who cast ballots on punch card voting machines - more than 175,000 ballots - did not register a vote on whether to recall Davis, researchers said.

Voters either abstained from the recall question or disqualified their selection by voting both "yes" and "no."

"It's inconceivable that one in 11 people in Los Angeles went to the polls and did not cast a vote on the recall," said Henry E. Brady, professor of political science and public policy at the University of California, Berkeley, who conducted one study.

By contrast, almost every response to the recall counted in Alameda County, which uses an electronic touch-screen system. The 0.7 percent of countywide responses without an answer to the recall question were likely cast by absentee ballot using the optical scan method, said Alameda County assistant registrar Elaine Ginnold.

Harvard University research fellow Dr. Rebecca Mercuri, who conducted the other study, concluded that many of the 383,000 ballots that didn't answer the recall question had their selections erased by malfunctioning machines.

Alfie Charles, vice president of business development at Oakland-based Sequoia Voting Systems, which prints the punch cards for Votomatic machines, said the suggestion the machines were broken was "so far off base it has no credibility whatsoever."

"Some people clearly want to abstain to express their opinion," Charles said. "It's dangerous territory to analyze those numbers."

Brady and Mercuri each compared differences in the total number of ballots cast, based on data published by California's secretary of state, with the total numbers of "yes" and "no" votes on the recall question. The difference is known as "dropoff" or "residual vote."

The number of residual votes on punch card machines totaled 297,775, or 6.3 percent of the votes; the total on optical scan machines was 72,190, or 2.7 percent; and the total for touch screen machines was 13,181, or 1.5 percent.

The state's punch cards, including some machines installed more than 20 years ago, were the subject of a lawsuit by the American Civil Liberties Union.

The ACLU had estimated that 40,000 votes could be lost on punch card systems and argued the recall election should be put on hold. But an 11-member panel of the 9th U.S. Circuit Court of Appeals said the election would go forward.

"They were playing with fire in this election, and it's a good thing the margins weren't close," said Mark Rosenbaum, legal director of the ACLU of Southern California. "I hope this puts to rest claims that these (punch card) machines have any place in a democracy."