Federal office to probe vote procedures

By Rick Klein, Boston Globe Staff  |  November 24, 2004

WASHINGTON The US Government Accountability Office is launching an investigation in response to allegations of voting irregularities that emerged in the aftermath of the Nov. 2 election, marking the first response by the federal government to concerns about the vote that have swirled around the Internet over the past three weeks.
 

The investigation, which was requested by 13 Democratic members of Congress, will not be a comprehensive look at all of the tens of thousands of allegations of irregularities. Instead, it will look broadly at vote-counting procedures around the country in an attempt to address concerns about the accuracy of the count.

Among the issues to be covered are potential tampering with computerized voting machines, barriers to voter registration, and varying standards for counting ''provisional ballots," which are given to voters who say they are eligible but whose names do not appear on the rolls.

''This is something of broad national interest," said Ralph Dawn, the office's assistant director for congressional relations. ''This will not be a catalog that addresses every allegation. But we will look at things from a systemic nature is it an anecdote or is it something we can generalize toward suggesting a problem and a potential solution?"

The investigation was announced on the same day that members of a federal panel said they will hold hearings around the country on complaints of irregularities in the presidential vote. DeForest B. Soaries, chairman of the US Election Assistance Commission, said he knows of ''gaps and vulnerabilities that have got to be addressed."

''We in no way think that the lack of a crisis stemming from Nov. 2 eliminates the need for action," said Soaries, whose agency was created in response to the Florida recount four years ago. ''We know some problems existed everywhere. We just don't know the extent to which there were problems."

The hearings will focus on the quality of voting machines, prevention of tampering, and ways to preserve records for recounts; statewide voter databases, which 40 states still lack, to assure that local officials do not bar eligible voters; and procedures for casting and counting provisional ballots.

Even proponents of those hearings and the Government Accountability Office's investigation said they do not believe they will alter the result of the election. The House Democrats who requested the investigation say voting problems were not widespread enough to have influenced the election's outcome.

In any event, the probe's expected six-month time frame means results will not be known until after President Bush is inaugurated for a second term. But the lawmakers said a thorough investigation is necessary to preserve the integrity of the voting system, regardless of who the winner was.

''What is far more important than the ultimate result is the public's confidence in it," said Representative John Conyers Jr., a Michigan Democrat who led efforts to get the Government Accountability Office involved. ''The purpose of this investigation is to figure out what systemwide deficiencies exist in voting machines and voting procedures so that those flaws can be fixed by legislation prior to the next election."

Allegations continue to arise citing differences between exit polls and results in some states, combined with individual cases in which voters reported irregularities at their precincts. Election problems were reported across the country, with most of the attention focused on Ohio and Florida, two large states that Bush won by relatively narrow margins. Kerry would have won the election if he had carried either.
The Green and Libertarian party candidates for president are pursuing statewide recounts in Ohio, where one precinct awarded nearly 4,000 extra votes to Bush and where frustrated voters waited up to eight hours at some urban polling places. The state's Democratic Party, meanwhile, is suing the state over standards used to count the 155,000 provisional ballots cast on Election Day. The unofficial election-night tally from Ohio, before the provisional ballots were counted, had Bush winning by about 136,000 votes.

A team of researchers from the University of California at Berkeley has suggested that electronic voting machines in some parts of Florida may have awarded Bush between 30,000 and 260,000 more votes than he was projected to get, based on past results and comparisons with this year's returns elsewhere in the state.

In addition, in some of the 15 Florida counties that used touch-screen computers, some voters said they pushed the button for Kerry but saw the screen mark the ballot for Bush. Because there is no paper backup for machines in those counties, it is impossible to determine whether those votes were properly counted.

Dawn said the investigators will begin by identifying which allegations could be more than isolated incidents, since purely local issues are best handled by state authorities. ''We will listen to individual issues, synthesize the information to try to look for systemic issues that can be generalized and solutions that can be found for them," he said.

No investigation will be able to address all of the theories that have emerged in the past three weeks, said Richard Hasen, an election law specialist at Loyola Law School in Los Angeles. But an investigation could dispel some myths, Hasen said, allowing lawmakers to separate spurious complaints from built-in problems with the nation's voting system.

He suggested backup systems for electronic voting and having nonpartisan state officials in charge of voting.

''I still haven't seen anything that convinces me that somehow the election results are in question that's good news but it's quite obvious that there were serious problems both with the technology and the rules," Hasen said. ''There's a lack of faith on the part of many people in the new voting technology, and it would have been better if you had trust over people who run the elections."