Not every Utah voted counted, but more did than in 2000
Percentage improves: This time, only 13 of 1,000 votes were thrown out because of errors
By Robert Gehrke
The Salt Lake Tribune 25 November 2004
Utah voters learned the lessons of the 2000 Florida election fiasco, and the result appears to be a considerable in the number of spoiled ballots, undervotes and election mistakes in the most recent election.
An analysis of the 2004 election results shows that only 13 in every 1,000 ballots in the presidential race were considered a residual vote or undervote, meaning the voter either failed to mark the ballot, an attempted vote didn't register, or the voter marked more than one choice, spoiling the ballot.
By comparison, 16 of every 1,000 ballots in 2000 did not register a vote for president. The change may have accounted for more than 3,350 additional votes in the presidential race in 2004.
The most dramatic improvement came in Morgan County, where in the 2000 election nearly 1 in 15 ballots cast had no vote for president. This year, the figure was cut by more than two-thirds, with fewer than one in 50 ballots being an undercount.
Utah Elections Director Amy Naccarato said the improvement is probably due to a number of factors, including the Florida election problems in 2000 and an effort by the state and county clerks to better train election judges.
"I think after what happened in Florida people actually pick up their ballots and look at the back and they pay attention," she said. "Things like undervotes and chads and provisional ballots are now becoming part of our language. That's one of the things I've seen come out of the 2000 election, is much more educated voters."
Juab County had the highest rate of undervotes in 2004, with more than 4 percent of ballots not reporting a presidential vote. Along with Washington, Weber, Summit and Sevier, Juab was one of the few counties that saw an increase in its undervote rate from 2000.
Doug Chapin, director of Electionline.org, an election reform group based in Washington, said the Utah figures track with what has been reported in many states.
"Given the attention that voters were paying that their vote was counted this time around, I'm not surprised that the numbers went down," said Chapin.
Two Utah counties have an entirely different problem on their hands. Rich and Daggett, which both use hand-marked paper ballots, inexplicably recorded more votes than ballots cast.
"We noticed that," said Sara Lamb in the county clerk's office in Daggett County, where 487 ballots were cast, but 499 votes reported. They've gone back over the counts submitted from the precincts and still haven't been able to figure it out, she said. "We haven't given up yet. We're still working on it."
The Rich County clerk could not be reached. Naccarato said she has not researched the issue.
Utah's undervote may be even lower in the next election. The state has received $25 million in federal funding through the Help America Vote Act, which Congress passed in the wake of the Florida debacle. More than $20 million will be spent purchasing a new voting system. The rest is for a statewide voter database, training, handling of provisional ballots and other requirements under the act.
The voting machine contract was initially intended to be awarded this week, but a demonstration of the systems in the running has now been postponed until January, said Naccarato.
Although there was only a small margin of error in the most recent election, the system still needs to be overhauled. Punch-card machines are used in 25 of the 29 counties. Four others use paper ballots and Emery County uses an optical scan machine, one of the new generation voting machines.
Naccarato said the punch-card machines are dated, tallying ballots can be a problem and they are not user-friendly for voters with disabilities. Chapin said punch cards generally have higher margins of residual votes than other technologies.
"They're just old and this is an outdated technology, and I think we can do better," said Naccarato.