The districts are Cleveland, Columbus, Toledo, Marion and Campbell in Mahoning County. Scrubbing is the practice of removing students from enrollment without lawful reason.
State Auditor Dave Yost released the preliminary findings Thursday as part of his investigation into potentially irregular attendance and enrollment practices around the state. He anticipates the last pre-election update by Oct. 23, two weeks ahead of fall levy votes in many districts.
"My sense is most schools are not doing this," Yost said. "My sense is also that as we continue this work, we are going to find other schools."
Yost cautioned that his review has not addressed districts' motives for scrubbing the data. He said it could be entirely innocuous.
"There's always been the possibility of criminal referrals. Ohio has a records tampering statute. It's a serious matter," he said. "That said, it does have a 'mens rea,' or an intent element, to it. And so the kind of work we've done here is necessary to support a criminal prosecution but it's not sufficient."
Thursday's results were drawn from a review of an initial 100 school buildings — or about 3 percent of Ohio's 3,688 public schools. The schools, housed in 47 districts, were selected for initial review because they had the highest number of student assessment test takers whose scores were "rolled up" to the state level and removed from district averages.
Yost said some schools in the five districts withdrew students based on a pattern of absences, which could have been influenced by lower test scores, without proper documentation.
That included Toledo City Schools, which automatically withdraws students who have had a total of 20 unexcused absences for the year after five consecutive days of them. Auditors found the district dropped this "5/20 rule," then reinstituted it after seeing a decline in local report card rankings.
Cleveland failed to document student withdrawals under a similar policy as a matter of routine, Yost said. "So we have to report that Cleveland is unauditable," he said.
In Columbus, 81 of 82 files for questionably withdrawn students lacked proper documentation. Their cases were selected from among about 10,000 students who withdrew during the 120-day window when student performance counts toward their school's overall ranking yet showed up as withdrawing after the window had closed, the report said.
State auditors found that 32 of 40 student withdrawals flagged for examination in Campbell were similarly unsupported.
"This is more than nit-picking about pieces of paper," Yost said. "Without the records, there is no evidence that the actions were properly taken by the government."
He said Marion's scenario was slightly different. Chronically truant students were automatically transferred to a new "digital academy" that administrators believed might serve them better academically. The policy was in place for only one year.
He said administrators or administrative staff, not teachers, generally perform the withdrawal function.
"It may occur in the superintendent's office, it may be part of the (role of the) principals, assistant principal. It may be done by clerical staff under the direction of someone else, or perhaps (by) the attendance officer at the individual schools," he said. "The practice varies by district, even by building."
His report recommends that the Ohio Department of Education improve the independence of its accountability measures and that report card performance ratings be removed from a location where many people can manipulate the outcome.