States change policies after SHNS points out problems

By THOMAS HARGROVE
Scripps Howard News Service
July 01, 2004

- Top election officials in Delaware, Oklahoma, Tennessee and Texas have vowed they will - for the first time - report the number of ballots cast for every county in their states during the upcoming presidential election.

Counting ballots is critical to determining if inaccurate voting machines or improper counting procedures are causing Florida-like tabulating problems.

"Everybody in the world is going to want to know that this year," Tennessee Elections Coordinator Brook Thompson said. "We will get a reading from our counties as to exactly how many ballots were issued."

Officials in these four states first announced their new policies when contacted by Scripps Howard News Service and asked to defend their past practice of reporting only the number of votes counted for each candidate and not how many ballots were actually given to voters.

"We are aware of this now and I am trying to do something about it," Texas Secretary of State Geoffry Connor said. He said his staff began "talking about this internally" after a Scripps Howard reporter called his office in April to ask why Texas doesn't count ballots.

"I don't think there will be any problem in getting this information," Connor said.

Florida's vote-counting problems drew worldwide attention four years ago when 178,145 ballots in the state did not register a vote for president, sometimes because cardboard chads had not been properly dislodged from punch-card ballots. A study by Scripps Howard News Service of official election returns in the 38 states that reported ballots cast found that at least 1,605,263 ballots did not register a vote for president four years ago.

There were at least 544 counties throughout the nation that had a rate of so-called "undervoting" - the difference between the number of ballots counted and the number of votes cast - that exceeded Florida's 2.9 percent. The study found the single largest error in the nation occurred in Cook County, Ill., where officials concede improperly manufactured punch-card equipment was responsible for most of the county's 122,289 undervotes.

"You want to look at (the number of) ballots cast and then compare that to votes cast just to see what the spread is," Connor said. "If the percentage of undervoting goes very high, then that might be an indication that you need to retest the equipment to make sure it is counting properly."

Oklahoma State Elections Board chief Michael Clingman said "no one's ever asked" for a county-by-county breakdown before. "I certainly agree that this should be done. I think it's probably a good idea."

Clingman said Oklahoma officials need to modify the computer software used to tabulate results on election night to calculate and retain the number of ballots cast. County officials in Texas and Tennessee will be asked to voluntarily provide data they don't normally report.

"We do occasionally get the counties to report information to us that is not mandated by legislation. And that's what we will do," said Thompson.

Officials in Delaware already have sufficient information from county reports to issue ballots-cast statistics in upcoming elections, according to Delaware Commissioner of Elections Frank Calio.

"This information is something we've always had internally. But we've never made this into a public record before," Calio said. "I don't see why it shouldn't be released. It should be released."