Board drops retest results

Adequate progress not made at four schools.Sanctions could be imposed; 3 elementaries may lose award money.

The decision to discount the results of last summer's fifth-grade math retest from all three schools —- and a fourth in coastal Glynn County —- means the schools did not make AYP, or adequately yearly progress, for the 2007-2008 school year.

It puts all four in jeopardy, if they also don't make AYP this year, of being tagged "needs improvement" and facing a variety of sanctions under the federal No Child Left Behind Act. At least three also may have to return some money.

The state Board of Education tentatively agreed late Wednesday afternoon to throw out the retest results from the four schools —- Atlanta's Deerwood Academy, DeKalb County's Atherton Elementary School, Fulton County's Parklane Elementary School and Glynn County's Burroughs-Molette Elementary School. An official vote today is considered a formality.

In June, the Governor's Office of Student Achievement brought the board the results of an investigation that showed someone at each of the four schools had erased wrong answers on some of the tests and replaced them, largely with correct responses.

The state investigation followed an analysis by The Atlanta Journal-Constitution in December about improbably steep gains at some schools on tests taken first in spring and then in summer.

The AJC's analysis showed a handful of schools' increases on the retest were statistically unusual to a high degree. One researcher who reviewed the newspaper's findings said Atherton's gains were so statistically unlikely that they were akin to a snowstorm in July —- in Atlanta.

Passing rates on the state's Criterion-Referenced Competency Test spiked after the retests, enough that each school was able to make AYP, a requirement of No Child Left Behind.

Administrators at all the schools initially denied any wrongdoing. But after the findings were made public, the principal at Atherton resigned, and the assistant principal was reassigned. Both are now facing felony charges of tampering with a state document.

The State Professional Standards Commission, the licensing agency for educators in Georgia, also said that it will look into whether any administrators should be disciplined in the cheating scandal.

Kathleen Mathers, executive director of the Governor's Office of Student Achievement, told the board Wednesday that administrators in Atlanta Public Schools are the only ones still refusing to acknowledge that the test results are tainted.

"They're taking the position that there's no confession, so it didn't happen," she said.

In the meantime, Mathers said, there's mounting evidence that employees at all four schools did change answers on the tests.

Among the strongest evidence, she said, was the results of this year's sixth-grade math tests.

Students who had failed the 2008 spring CRCT and then appeared to make stunning turnarounds on last summer's retest didn't do well on the 2009 Spring CRCTs, Mathers said.

For example, 100 percent of Atherton's fifth-graders passed last summer's now-suspect retest. When they took the sixth-grade math test this year, only 41.4 percent of them passed, Mathers said.

It was the same at the other three schools —- high passing rates on the retest and dramatic failure rates on this year's tests, she said.

"Of course we would expect if they did very well in fifth grade, they would be able to stay in that same level of achievement," Mathers said.

"The evidence is extremely compelling," said Larry Winter, a member of the state school board.

The board plans to require that each system send a letter to the parents of the students who took the retest, advising them what happened and developing a plan, if needed, to provide extra math help to those students.

DeKalb school system spokesman Dale Davis said parents at Atherton were contacted "immediately" and monitoring is in progress. Davis said a support plan for the students is being provided at no charge.

The money that's at issue is offered to schools with significant populations of poor students as a financial reward if they meet federal standards —- which include improving test scores —- at least four years in a row. The amount each "Distinguished School" receives is determined by its level of poverty, the number of schools that qualified statewide and the total money the federal government provided for the program.

State records show Atherton received $10,130 after meeting federal standards last year. Parklane received $8,104, and Deerwood won $842.

Parklane spent the money on supplies such as paper, pens and books, and on replacement bulbs for high-tech educational tools, state records show.

Deerwood used its grant for a field trip to Atlanta college campuses, the Martin Luther King Jr. Historic District and a museum.

Atherton, however, elected to spend the money on cash awards to school staff, giving about $142 to each of 65 staff members, records show.

The decision to pay cash to staff was made by the Atherton Elementary School Leadership Team, according to records.

Newest revelation

The Governor's Office of Student Achievement said Wednesday that it has more evidence that retests of the CRCT were altered. The office compared the passing rates for students at the four schools on the fifth-grade math retest and the spring CRCT for sixth grade. Officials said they would have expected results to be consistent, but they varied widely.

Deerwood Academy in Atlanta

Fifth grade: 87.5%

Sixth grade: 41.7%

Atherton Elementary in DeKalb

Fifth grade: 100%

Sixth grade: 41.4%

Parklane Elementary in Fulton

Fifth grade: 83.3%

Sixth grade: 29.2%

Burroughs-Molette Elementary in Glynn County

Fifth grade: 85.7%

Sixth grade: 33.3%

Staff writer Heather Vogell contributed to this article

Support real journalism. Support local journalism. Subscribe to The Atlanta Journal-Constitution today. See offers.

Your subscription to the Atlanta Journal-Constitution funds in-depth reporting and investigations that keep you informed. Thank you for supporting real journalism.