Gov. Sonny Perdue took the extraordinary step Wednesday of announcing that a special investigator will look into the cheating scandal at Atlanta Public Schools, calling a local review “woefully inadequate both in scope and depth.”

“This is about individual students being robbed and cheated of their one fair shot at a good education,” said Perdue, who made the announcement Wednesday during a rare appearance before the state Board of Education.

Perdue also charged that APS’s botched investigation delays cases against 100 of the 108 Atlanta educators referred to the state for possible disciplinary action because the system did not supply the proper documentation to the state.

Atlanta school officials, in a short statement issued by their spokesman, said they will cooperate fully. “APS welcomes the governor’s call for a special investigator to look into this matter,” they said.

Perdue said an investigator will be named as soon as today.

Perdue’s special investigator will have subpoena power, meaning he or she will be able to compel witnesses to testify. Under state law, an investigator can refer cases to local district attorneys or to the state Attorney General’s Office for prosecution. State law prohibits falsifying state documents but there is no law specifically against cheating on tests.

Atlanta employees who refused to talk to the district’s investigators can no longer hide, Perdue said.

In addition, the state will continue withholding federally-mandated Adequate Yearly Progress status for the 58 public schools, pending the results of the investigation.

‘Different tools’

The inquiry started by Perdue, who leaves office at the end of the year, will start immediately and will focus on Atlanta and Dougherty County schools, which state officials said also did an “incomplete” review of schools that may have cheated on state tests.

Perdue’s action is a response to Atlanta’s and Dougherty County’s investigations, ordered by the state, of schools found early this year to have a high number of suspicious erasures on the 2009 Criterion-Referenced Competency Tests.

Gary Price, a partner with Pricewaterhouse-Coopers and chairman of the commission the Atlanta school board named to conduct the local investigation, said the panel has “no problem with it whatsoever.”

The governor, he said, can “extend the investigation and use different tools.”

The Atlanta investigators reported that widespread cheating appeared to be limited to 12 schools among the 58 flagged by the state as requiring further scrutiny. Some problems were identified at 13 schools. The commission noted fewer concerns among the other 33 schools. It recommended 109 employees for further investigation.

Atlanta Superintendent Beverly Hall last week began referring 108 of those employees to the Georgia Professional Standards Commission, which polices state teaching credentials.

But on Wednesday Perdue said the city’s incomplete report means only eight of those employees face “legitimate complaints ... with allegations of wrongdoing that can be investigated.”

It is up to Atlanta to take action before referring educators to the agency, he said. “The PSC is not an investigative agency to be used by local school systems to ferret out wrongdoing,” Perdue said.

APS probe criticized

Atlanta submitted its report to the state early this month and it immediately drew criticism because the investigative panel’s approach, including its reliance on the work of its consultant, Caveon Test Security, seems to have conflicted with instructions from the state. Caveon used a different standard to assess test data.

Instead of rigorously investigating all 58 schools, the commission gave lower priority to 31 schools after Caveon reported that its own data analysis turned up fewer concerns than the state had reported.

On Wednesday, Kathleen Mathers, executive director of the Governor’s Office of Student Achievement, the state agency that conducted the original testing audit, blasted that approach.

“Children in one school don’t deserve less of an investigation than in another,” she said in a presentation to state board members that preceded Perdue’s appearance.

Caveon’s analysis, Mathers said, did not compare data on suspicious erasures within Atlanta with data found in other districts statewide. As a result, the analysis compared outliers “against outliers,” potentially minimizing unusual findings.

As such, the Caveon analysis appeared to have been used to clear schools that the state had identified as potentially suspicious – not to flag all potential problems, she said.

Didn’t have all data

She cited two elementary schools, D.H. Stanton and Boyd, that Caveon’s analysis helped place in a lowest-concern category – essentially considered clear of wrongdoing – despite indications scores might have been manipulated.

At D.H. Stanton, more than 58 percent of classrooms produced suspicious erasures, two teachers had highly unusual such marks across all three subject tests and the school’s CRCT scores dropped in 13 of 15 areas during the closely monitored 2010 test, she said.

At Boyd, a similar percentage of classrooms were flagged, one teacher had 100 percent pass rates on more than one subject and CRCT scores dropped in 11 of 15 areas.

Told of the state’s comments Wednesday, John Fremer, president of Caveon, acknowledged the analysis was limited because his company did not have all the data it would have liked, such as student-level responses to questions.

“The bigger the body of evidence you have to do comparisons, the better off you’re going to be,” he said. “We would have been better off to have all the schools in Atlanta Public Schools than to have just a subset.”

Still, he said, the company’s work helped investigators focus their work on the most questionable schools. He said his firm, which works with districts nationwide, found “virtually no evidence” of suspicious erasures at the 33 schools in the lowest-concern category.

If the state wants a more thorough investigation, it “ought to make more information available to whoever is doing that investigation,” he said.

Mathers also took issue with the district’s investigation, saying the state had received evidence of interviews for only 32 of the 58 schools identified. Some employees gave non-responsive answers to questions at half the dozen schools the district found most suspicious. The state does not believe the interviews that were conducted followed state guidelines, she added.

Mathers said the investigative report lacked summaries of interviews at some schools, leaving state officials with no information about the sessions. The state expected details, school by school, describing what questions were asked, who was interviewed and what investigators learned, she said.

“We don’t have a clear picture of that from the information contained in the report,” she said in an e-mail.

In the meantime, parents wait and wonder.

Rachel Mensah, a new parent at Dunbar Elementary, said she has been following the allegations. She supports a state investigation into the matter so she can be certain her child is in good hands at school.

“We teach our kids to be honest,” Mensah said. “If there is any dishonesty, I think we need to get to the bottom of it so we can know for sure.”

Contributing to this article: Staff writers Nancy Badertscher, Katie Leslie, Megan Matteucci, Bill Rankin, Craig Schneider, Ernie Suggs, Steve Visser and D. Aileen Dodd