Nor was voter turnout significantly different from turnout in the 2016 election at nursing homes in four of the counties for which the Wisconsin State Journal obtained voting data.

The newspaper’s findings again throw cold water on claims by conservative former state Supreme Court justice and GOP special counsel Michael Gableman and supporters that widespread voter fraud was responsible for President Joe Biden’s 21,000-vote victory in Wisconsin. A report on the election by the state’s nonpartisan Legislative Audit Bureau, recounts in the state’s two largest counties, and multiple courts have found no evidence to support their claims.

Nevertheless, during a presentation March 1 before the state Assembly’s elections committee to discuss his most recent findings, Gableman showed videos of purported nursing home residents who appeared to be incapable of understanding the voting process but had, in fact, voted in the 2020 election.

He then pointed to figures in his 136-page “second interim investigative report” claiming 100% of the registered voters in nursing homes in heavily Democratic Dane and Milwaukee counties, and in Racine County, home to the Democratic-leaning city of Racine, cast ballots in 2020. In Kenosha and Brown counties, home to the Democratic-leaning cities of Kenosha and Green Bay, the figures were 97% and 95%, respectively, he said.

Earlier that year, the Wisconsin Elections Commission told municipal clerks they were not required to send special election workers into nursing homes to assist with the election because of the pandemic. Gableman’s clear insinuation was that someone not qualified to conduct an election improperly influenced these vulnerable voters or, worse, cast ballots on their behalf.

A State Journal review of municipal poll books last month showed a far different story in Dane County. Poll books, sometimes in electronic form and sometimes on thousands of pages of paper, allow the public to see who was registered on the date of an election and which of those voters cast ballots in that election.

In only one of Dane County’s 18 state-licensed nursing homes was turnout 100%: Nazareth Health and Rehab Center in Stoughton, where all 12 people listed as registered in the poll book had their ballots tallied. Turnout among all the others ranged from 42% to 91%.

Gableman’s turnout numbers for nursing homes in the other four counties are proving equally false.

Using data requested through the Wisconsin Election Commission’s Badger Voters service — which allows political campaigns and the public to obtain lists of registered voters and voting activity by address — the State Journal calculated voter turnout at all 52 state-licensed nursing homes in Brown, Kenosha, Milwaukee and Racine counties in which people were recorded as having voted in the Nov. 3, 2020, election.

It could find only one where turnout in 2020 was 100%: the Brown County Community Treatment Center — Bayshore Village in Green Bay, where eight of eight registered voters voted.

Otherwise, turnout at the nursing homes ranged from 20% to 94%. Average turnout for Milwaukee County nursing homes was 80%, as it was for Brown. Turnout in Kenosha facilities was 72% and in Racine it was 73%.

Such levels are not out of line with what the state as a whole saw in 2020, when turnout was 72%, according to the bipartisan Wisconsin Elections Commission, and nearly 10% more people cast ballots than in 2016.

What’s more, the nursing home figures overstate turnout in the facilities because they reflect the number of votes cast divided by the number of registered voters, not the larger voting-age population, which the Elections Commission uses as the denominator in its turnout calculations. It’s also long been true that senior citizens are more likely to vote than the population at large.

Nor was the 2020 turnout at nursing homes in Brown, Kenosha, Milwaukee and Racine counties much different from that in 2016. In fact, three homes in those areas saw 100% turnout, with 41 of 41 registered voters casting ballots in an election won by former President Donald Trump, who praised Gableman at the former president’s Mar-a-Lago resort on Tuesday and whose baseless claims of a stolen election Gableman has also taken up.

‘Incorrect and deceptive’

Gableman and his team did not respond to requests for comment for this story. Nor was the newspaper able to obtain comment from Assembly Speaker Robin Vos, R-Rochester, who appointed Gableman, or the Assembly elections committee chair Janel Brandtjen, R-Menomonee Falls, who has called for a “full, cyber-forensic audit” of the election.

Wisconsin Voter Alliance president Ron Heuer, who has said he compiled Gableman’s turnout numbers when he worked for him from about Oct. 1 to mid-December, also didn’t respond to a request for comment. The Kewaunee County-based Wisconsin Voter Alliance touts itself as “promoting and protecting the integrity” of Wisconsin’s voting system and has unsuccessfully sued to overturn Wisconsin’s 2020 presidential election results.

In a statement, the ranking Democrat on the Assembly elections committee, Rep. Mark Spreitzer, D-Beloit, called Gableman’s report “misleading” and said “the only way that they didn’t choose to intentionally mislead the public is if they have no idea how this works.”

“Mike Gableman should not have made an assertion about nursing home turnout based on partial data, and should have reached out to the actual experts — like the Wisconsin Elections Commission — that could have explained how this works,” he said. “Instead, he released a report that was incorrect and deceptive.”

Heuer said last month that he used the state’s voter database as of August 2021 to look at whether every registered voter at every nursing home in the five counties voted in the November 2020 election, although he erroneously included other kinds of group living arrangements, such as independent living for elderly people, in his calculations.

A spreadsheet of his findings provided to Spreitzer’s office, which provided it to the State Journal, shows columns for the number of active registered voters and number of votes cast in the 2020 election, often resulting in turnout figures that are inexplicably well above 100%. Spot checks of the data on registered voters and votes cast don’t match what the State Journal found in poll books.

Heuer has declined requests to go into detail about how he arrived at his turnout figures.