Efforts to verify Dane County’s voting-machine output were still in their childhood for the 2015 elections. The Wisconsin Election Integrity Action Team conducted efficient, effective and routine citizens’ audits that met nationally accepted standards for transparency, but because we hadn’t yet found a professional statistician willing to work for free, they didn’t meet validity standards. And Dane County Clerk Scott McDonell wasn’t even trying to conceive an official process — he was on record that verification was “unnecessary and possibly contrary to statutes.” Since then, the citizens’ audit process has grown to adolescence. A civic-minded statistician volunteered, and our March 12 public audit of the February election verified with 99 percent confidence that voting machines identified the correct Supreme Court primary winner. About 30 public observers were satisfied they could see every vote; they even participated in randomly selecting nine precincts at the start of the event. We also examined a suspicious result in one Madison precinct, where the voting machine saw no votes on 1.26 percent of the ballots, compared to only 0.14 percent among other machines. The public count satisfied everyone present that the machine total was accurate. An observer who knows registration requirements explained that a large elderly housing complex may explain the blank ballots, because homebound “permanent absentee” voters can maintain that status only as long as they return a ballot in every election. As for official audits, McDonell’s office may just have given birth to a county audit process! If you dig into the Dane County website, you can find a recent report of his close-to-the-vest efforts, beginning in December 2015, to devise his own system for verifying voting-machine output.
According to the report, he hired an experienced chief inspector (head poll worker) to develop a procedure for verifying two randomly selected voting machines. The process doesn’t seem intended to achieve validity (no statistical justification for the sample size), transparency (no mention of citizen observers), or timeliness (it was done only after results had already been declared final). However, McDonell might be intending to make it routine after every election, so that thieves will know there’s never a time when the figurative security cameras are turned off.
It seems to have been a difficult delivery. Because McDonell did not consult with an IT professional, the chief inspector experienced massive problems with the zipped digital-image files. The poor man stayed up all night manually extracting the images — a process that wasted time, was unpleasant, introduced error, and impaired the ability of the public to know that no one had tampered with the images. Lost ballot images threw off the results, so the audit itself had to be checked with a hand count of paper ballots.