Watchdogs Spot E-Vote Glitches

[Guest guest]

http://www.wired.com/news/evote/0,2645,65579,00.html?tw=wn_tophead_1

 

Story location: http://www.wired.com/news/evote/0,2645,65579,00.html

 

06:21 PM Nov. 02, 2004 PT

 

Watchdogs Spot E-Vote Glitches

By Kim Zetter

 

The jury is still out on e-voting machines used in the election but

reports collected by late Tuesday evening by election watchdogs seem

to contradict assurances by voting company representatives that the

election should " put to rest the unreasonable suspicion " about

e-voting machines.

 

The National Protection Coalition, composed of several nonpartisan

groups that include the Electronic Frontier Foundation and Verified

Voting, reported Tuesday afternoon it had received more than 600 calls

from voters complaining about problems with e-voting machines around

the country.

 

A separate group, Common Cause, reported receiving 50,000 calls,

though not all of them were related to voting technology. Both groups

had established toll-free phone lines for voters to report problems.

 

The National Protection Coalition received 80 reports of problems in

New Orleans where machines made by Sequoia Voting Systems failed to

start on election morning, resulting in voters being turned away from

polls because election officials didn't have a back-up plan. By late

afternoon some machines still had not booted up.

 

Lawyers for the National Association for the Advancement of Colored

People and EFF filed a complaint in Civil District Court for the

Parish of New Orleans to force election officials to keep the parish

polls open longer to accommodate voters disenfranchised by the faulty

machines. Sequoia did not return a call for comment by press time.

 

In Florida, where George W. Bush won the 2000 presidential election by

only 537 votes, 10 touch-screen voting machines failed at precincts in

Broward County.

 

Voters in Florida and Texas complained about calibration problems with

touch-screen machines. Problems occurred when voters touched the

screen next to one candidate's name and an " X " appeared in a box next

to another candidate's name. The Election Protection Coalition also

received more than 32 reports from various states that spread across

all the top e-voting brands made by Diebold Election Systems, Election

Systems & Software, Hart InterCivic and Sequoia.

 

These problems involved e-voting machines that appeared to record

votes correctly when voters touched the screen, but indicated a

different selection on the review screen before voters cast their

ballot. In some cases voters had to redo their ballot five or six

times before the correct votes took.

 

" If we end up with a race as close as predicted, small changes could

mean the difference in who wins the presidential election, " said Cindy

Cohn, EFF's legal director. " We don't have any margin of error by

voting machines in a close race. That's particularly troubling. "

 

Voters in Palm Beach County, Florida, reported that when they went to

vote on Sequoia machines some races on their electronic ballots were

already pre-marked before they started voting. They had to ask poll

workers to assist them in removing the selections from the ballot so

they could start with a clean ballot. In some cases they weren't

successful in doing this.

 

In Texas, voters casting straight-party tickets reported that machines

cast ballots for candidates outside of their chosen party. For

example, if a voter chose to vote straight Republican, rather than

automatically marking all Republican choices on the ballot, the

machine marked some Democratic choices.

 

And in Pennsylvania voters in at least six precincts that used an

older variety of e-voting machines made by Danaher were prevented from

voting because of failing machines. Election officials claimed in news

reports that they had never had problems with the machines in the

past. But Cohn expressed doubts about this.

 

" Is it true they never had problems in the past or was no one looking

(at problems)? " she said. " I suspect if there had been folks looking

at elections past (we would have seen problems with the machines). But

election officials were able to pretend there were no problems with

their machines with no one watching. "

 

Michelle Shafer, spokeswoman for Hart InterCivic, said the problem

that occurred in Texas with her company's machines were caused by

voters rather than by the machines. The Hart machines are not

touch-screen machines but instead use a wheel that voters turn to make

their selections. Shafer said after choosing the straight-party

option, many voters turned the wheel to manually go through the races

and click their choices individually to emphasize them, not realizing

that in doing so they de-selected their choices. Shafer said they

probably then mistakenly moved the wheel to select a candidate from

another party.

 

" It's not a machine issue, " Shafer said. " It's voters not properly

following the instructions. "

 

David Beirne, spokesman for Harris County, where some of the problems

occurred, said voters had made the same mistake two years ago when

political parties instructed voters to go back through the ballot and

emphasize their choices.

 

" I think that often times the voter information passed out to voters

is incorrect, " he said. " We encourage voters to take their time and

ask questions and watch the videotape demonstration that's provided. "

 

Representatives for other voting machine vendors couldn't immediately

be reached for comment.

 

Doug Chapin, director of the Election Reform Information Project, a

nonpartisan research group, characterized the e-voting problems

reported so far as " lots of littles " that didn't add up to major

complications. " We know of no major meltdowns anywhere along the lines

some people were worried about, " he told the Associated Press.

 

But David Dill, a founder of Verified Voting and Stanford computer

scientist, said although the reports might not appear to be

statistically significant -- given that more than 100,000 touch-screen

machines are being used in 29 states this year -- they raise questions

about the number of problems that are not being caught or reported.

 

" We're only receiving a small percentage of (reports) on the problems

that are actually out there, " Dill said. " Most voters wouldn't be

motivated to call in and complain and may not know about the number

for calling. "

 

Cohn said most voters, if they report problems at all, tend to report

them to election officials. She said election officials have been

" stingy " in the past about sharing that information with lawyers and

watchdog groups.

 

The Information Technology Association of America, which recently

began representing the e-voting machine vendors, called e-voting for

early voters a " success. "

 

" Returns suggest nothing but the accurate and secure operation of

electronic voting machines, " ITAA President Harris Miller, said in a

statement.

 

But Dill noted that the data that voting machine vendors and academics

generally use to evaluate the integrity of e-voting machines doesn't

include the kinds of problems that voters have been reporting.

Generally, the number of undervotes and overvotes on a machine are

used to measure their effectiveness.

 

Undervotes occur when a machine records no choice for a particular

race -- either because the machine failed to record it or the voter

chose to skip the race. Overvotes occur when voters choose more

options or candidates than the race allows. E-voting machines are

supposed to make it impossible for voters to overvote.

 

But " recording a vote for a wrong candidate is not something that

shows up in the statistics, " Dill said. It doesn't show up in the

statistics because officials have no way to know whether a machine

incorrectly recorded votes. Without a paper trail or some other way to

independently verify that the votes on the machines are the votes

voters intended, there's no way to truly measure the accuracy of the

machines.

 

The group said it would continue to monitor reports in the days after

the election. But it would not be able to monitor problems that might

occur with counting software in election offices after the polls

close. Cohn said only party representatives are allowed to observe

this function in most states.

 

The group also would not be able to monitor whether the number of

votes cast on machines in each precinct match the number of voters who

sign in at the polls. These figures, if they don't add up, can point

to significant problems with machines dropping votes. The numbers can

also help determine if votes were lost or changed from the time the

polls closed to the time they arrived to the central election office

for counting.

 

Will Doherty, executive director of Verified Voting, said the group

wanted to do this, but didn't have the resources to focus on that this

time around.

 

" Despite trying, we unfortunately won't have the kind of widespread

nationwide statistics that we had hoped to have. "

 

End of story