BOISE, Idaho — A state audit of more than 19,000 ballots cast in Idaho's general election has identified only six variations, according to a report from the Secretary of State's office.
The variations all resulted from sorting errors or faint markings on the ballots that were counted differently by human eyes and automatic tabulation machines, Deputy Secretary of State Jason Hancock said Tuesday.
“The takeaway is that Idaho elections are well run, and the results that the counties are reporting are accurate and reliable,” Hancock said. “We've not found any significant problems or issues with the counts.”
The audit is required under a law passed earlier this year intended to increase public confidence in election results by checking paper ballots. The procedure also provides a check of the equipment and procedures used to count votes across Idaho’s 44 counties.
State officials randomly selected Ada, Bonner, Bonneville, Kootenai, Minidoka, Nez Perce, Oneida and Shoshone counties for the audit last month. Precincts within each of those counties were also randomly selected.
The vote results reported by county officials in Bonner, Kootenai, Minidoka, Nez Perce and Shoshone exactly matched the hand count performed by auditors.
Very faint markings on some ballots in Ada, Bonneville and Oneida counties likely led to most of the variances when they were counted differently by automatic tabulation machines and the auditors, Hancock said. Another variance in Bonneville County was from a sorting issue, he said.
“For instance, you might rest a pen within a 'no' bubble on a ballot while you are deciding how to vote, and then end up filling in the ‘yes’ bubble. The tabulator may count that tiny ink mark in the ‘no' bubble as an overvote," Hancock said.
Auditors also looked at how many ballots were issued to each precinct, and then checked that number against the number of ballots cast, spoiled and remaining in the county's inventory. Those numbers matched in each county checked, Hancock said.
The general election audit results were similar to those from the primary election, when just 6six variances were found from about 20,000 ballots.
