While the official results of a post-election audit for the Aug. 10 primaries have yet to be released, a coalition of citizen organizations has already released its own analysis of the election, which outlines a number of concerns about the integrity of the state’s election audit system.

The report, released on Wednesday, is the sixth such report that the Connecticut Citizen Election Audit Coalition has released since the state adopted an optical scanner and paper ballot system for the statewide election in 2008.

The coalition’s executive director Luther Weeks said that, while there have been improvements since the group’s last audit, the August primaries uncovered a problem with the state’s random selection process that determines which towns are chosen to be audited.

According to state law, the Secretary of State’s office is required to audit 10 percent of the state’s voting districts after an election to determine whether the machines designed to count the votes are working properly. A master list of voting districts, submitted to the state by each municipality, are put into a bin and selected at random. The ballots from the towns selected are then hand-counted and compared with the results generated by the machines.

Weeks said that the master list used to select the towns was inaccurate and contained voting districts that don’t exist.

“Unfortunately, we discovered that the list of polling districts for the random audit drawing was missing some districts and is otherwise inaccurate and ambiguous,” he said in a press release. “The integrity of the audit requires an accurate list of districts that is verifiable by the public.”

Weeks said that an inaccurate list creates holes in the state’s ability to assess its elections and could be exploited if not rectified.

“It gives someone the ability to avoid [auditing] certain districts if they wanted to,” he said in a phone interview Wednesday. “If there were discrepancies it may be impossible to verify whether an error was made.”

Deputy Secretary of State Lesley Mara said that the list of voting precincts is compiled from information provided to the state by each of its 169 municipalities. While it is the responsibility of each town to inform the state of their voting districts, Mara said that there is no law requiring the towns to report changes to their voting districts.

For instance, the town of Enfield went through the legal process of consolidating some of its voting districts, Mara said. However, the town never informed the state so the master list was never updated to reflect less precincts in the town, she said.

Mara said that the problem could be easily rectified by a bill requiring towns to submit an up-to-date list of districts and polling places two or three weeks before each election.

Mara also defended Registrars of Voters from some other complaints outlined in the coalition’s report, including weakness in the ballot chain of custody, an area she said the state has strict standards when compared with other states.

She also said that the official audit, which is compiled by the state and analyzed by a team at the University of Connecticut’s computer science department, has shown continued improvement since the state began using the optical scanning system.

“We expect people to get better at following procedures as they get more familiar with the technology,” she said.

She also noted that University’s audit reports have reflected that the counts generated by the machines have been accurate.

“The consensus is that the machines really are doing what they are programmed to do,” she said.

Mara also noted that the coalition’s report was based off of the reports of its observers, a group of volunteers who go to polling places throughout the state and report on the effectiveness, organization, and methods of the officials running the polls.

But the role of the coalition’s observers are not well defined and not mentioned in any law regarding the voting process, which only indicate that it must be done in public, she said. Also the criteria the districts are judged on are sometimes very subjective and not in the state’s laws, Mara said.

For instance, one criterion observers were asked to assess in this year’s audit was “Do you have any concerns over the way the room was laid out?”

“The coalition has a favored way of getting this done,” she said, “but before we issue a report card, it ought to be in the law.”

Mara said that subjective questions and varying training and experience on the part of the observers makes it difficult to use some of the feedback generated by the report but she was careful not to criticize the group and said they are “a critical partner” in continuing to improve the integrity of the state’s system.

“There is always value when citizens are involved and watching their government,” she said.

The UConn report on this year’s audit of 73 precincts is expected to be released within the next week, Mara said.