SARASOTA - Recounts won't explain why 18,000-plus votes in the congressional race weren't cast or recorded on Sarasota County's touch-screen voting machines, which have a history of similar problems elsewhere, voter advocacy groups and a voting-machine expert said Friday.

The machines can't produce a paper trail that would show what caused the large number of "undervotes" in the Vern Buchanan-Christine Jennings race, making two anticipated recounts useless, the groups said.

"There is no meaningful recount because the ballot the voter saw on the screen on Election Day is not there any more," said Stephanie Frank Singer, co-director of the ElectionIntegrity.org advocacy group.

Another group went further, demanding elections officials scrap Tuesday's election results altogether.

"Election officials should conduct a revote in this county," said Ben Wilcox, Common Cause Florida's executive director. "To certify this election would make a mockery of democracy."

Supervisor of Elections Kathy Dent, who has said the machines are not at fault, did not return a telephone message left with her office Friday. Although the office was officially closed for Veterans Day, several employees were working in anticipation of a machine recount starting at 10 a.m. Monday.

Unofficial results show Buchanan beating Jennings by 373 votes out of more than 237,000 cast. The 0.12-percent difference is small enough to trigger both a machine and a manual recount by state law.

But 18,382 is the number that has Jennings refusing to concede and considering a legal challenge. That's the difference between how many people voted in Sarasota County and how many votes were counted in the Jennings-Buchanan race.

Dent, Buchanan's campaign officials and other Republican leaders contend those voters either accidentally overlooked the high-profile and acrimonious race, or opted not to vote in it.

"It's a bizarre election, but not a crooked one," said state Rep. Nancy Detert, R-Venice, who finished second to Buchanan in the GOP primary in September. "Lots of people were sending a message, incensed at the phone calls and negative campaigning. Instead of staying at home, they decided to vote in the gubernatorial race and to skip the congressional race."

But Jennings, Democrats and numerous voters maintain the machines didn't record votes, especially those for Jennings. They cite lower undervote rates in other races on the same ballot, as well as the congressional race in the district's other counties.

"It's so much evidence that there's a problem with how the machines were programmed," said Susan Greenhalgh, spokeswoman for VoteTrustUSA, a network of non-partisan groups advocating accurate and transparent elections. "They don't have a verified paper record, so there's no way to go through and recount it and know what the voter truly intended."

On the ballot, the Buchanan-Jennings race appeared on the second screen beneath a heading with the words "U.S. Representative in Congress, 13th Congressional District" in capital letters. Above that, the word "Congressional" appeared but is not highlighted - a format different from other races, in which the top heading was highlighted.

It's not the first time the iVotronic machines, made by Election Systems & Software Inc., have been accused of not registering votes.

According to VerifiedVoting.org, another voter advocacy group, past problems with iVotronic machines include:

• Recording 134 undervotes in a special election for a Florida House seat in January 2004, even though the race was the only one on the ballot and the margin of victory was 12 votes.

• Failing to record 436 ballots at two early-voting locations in North Carolina in November 2002.

• Recording one or no votes by hundreds of voters at several Miami-area polling places in September 2002.

In Sarasota County, the congressional undervote rate for those who used iVotronic machines was more than 16 percent. The overall rate, which includes those who cast paper absentee ballots, was 12.9 percent.

The Orlando Sentinel reported Friday that it found even higher undervote rates in the Florida attorney general's race in three other counties that also use iVotronic machines: Charlotte, Lee and Sumter.

A company spokeswoman defended the machines' performance in Sarasota County, although it was the elections office - not the company - that prepared them for the election.

"We've been in touch with Dent and she indicated that her very strong belief is that the equipment functioned well and as it should," Jill Friedman-Wilson said. "Undervotes are attributed to one factor and one factor only: the lack of selection in this particular race."

That's because there's no way to tell otherwise, said Douglas Jones, a University of Iowa computer science professor and expert on electronic voting machines. As a result, Florida requires elections officials to assume blank votes were intentional.

The only way to tell if the machines malfunctioned is to pull them out of storage and conduct a thorough, and public, examination, he said.

"Federal courts probably are going to have to supervise this as well as the state courts," Jones said. "I think that's where this is headed."

http://www.bradenton.com/mld/bradenton/news/local/15985968.htm