The D.C. elections board said yesterday that a three-hour lapse Tuesday night in releasing results of the District's primary contests stemmed in part from a delay in receiving cartridges from one of the two types of voting machines used in the city.
The first two sets of vote totals were released by the D.C. Board of Elections and Ethics by 9:45 p.m. Tuesday, but they reflected only the paper ballot counts in 88 of the District's 142 precincts. It was not until 12:45 a.m. yesterday that the board released totals for all but two of the city's precincts, including those votes cast on touch-screen machines.
Elections board officials explained the lapse by noting that the District is the only jurisdiction in the country that uses two different voting systems in each precinct.
One allows voters to record their choices for office on paper ballots that are counted by an optical scanner before the information is sent to the election board via modem. The other enables voters to cast their ballots electronically on touch-screen machines that store the information on cartridges, which have to be physically taken to the board to be tallied.
The board's chairman, Wilma A. Lewis, said the agency needs to determine whether the software that it is using is the best for handling a voting system that relies on two kinds of machines. Elections board officials noted, for instance, that until all the data from the optical scanners has been inputted, the touch-screen software creates a report as if touch-screen machines were the only ones being used at each precinct.
Alice P. Miller, the board's executive director, said that her office received 76 cartridges by around 10 p.m. and had all but the two about an hour later. Miller said the board did not make the touch-screen votes public until the end because releasing the information earlier would have made the reports too confusing.
"These are two different systems, and that information has to be consolidated and totaled and it is quite an operation," Lewis said. "If it were all one system, it would certainly be easier."
She added that the three-hour gap in announcing election results was also time the board took to be thorough. "The board's primary objective is to get out accurate results," Lewis said. "If that means taking a little more time, that is what we are going to do."
Lewis said that Tuesday was only the second time the board has used the dual system; the first was in January for the District's presidential primary. She said the board would use the time leading up to the Nov. 2 presidential election -- when turnout almost certainly will be much higher than on Tuesday -- to identify areas for improvement.
"We have the opportunity to not only look at the process in its entirety, but at focused areas that did not work as well as we would have liked them to have worked," Lewis said. "When November comes around, we hope there would be improvements."
Miller said that the District switched to optical scanner machines in 2002 before a lawsuit charged that they were not accessible to the disabled. A subsequent court settlement, she said, led the city to add touch-screen machines.
"They are continuing to work out the kinks," Lewis said. "I think, quite frankly, there is a growing process involved in this . . . in terms of the technological factor and the human factor."
The board explained that the cartridges from touch-screen machines at two precincts were not included in the 12:45 a.m. totals because one had been left at the voting station -- where it was secured in a safe place -- and the other mistakenly had been put in an official, locked elections board bag containing paper ballots.
Referring to the board employees and volunteers who work on election results, Lewis said, "We have to make sure that everybody is conscious of the fact that the first order of business once the polls are closed is getting the cartridge to the appropriate person so it can be taken down to the board of elections."