Former Atlanta schools chief Beverly Hall is defending her reputation in a national education publication.
Investigators released a report in July that found soaring scores on the Criterion-Referenced Competency Tests that made Hall an education star were the result of widespread cheating. Critics have called her a crook and a liar. Some want her to return more than a half-million dollars in bonuses. Others want her behind bars.
So when the trade publication Education Week ran a commentary by Hall on Aug. 10, PolitiFact Georgia took notice. What could she possibly say to clear her name?
Hall’s piece, "The Scandal Is Not the Whole Story," argued that because the district improved test security in 2010 and 2011, cheating did not taint all of APS’ gains.
"The results of the standardized tests administered in 2010 and 2011 under this enhanced security have not been questioned -- and most important of all -- the dramatic improvement in test scores has remained," Hall wrote.
The 2010 and 2011 results "have not been questioned"? We reached for our Truth-O-Meter.
Before we launch into our analysis, we need to make two points. The first is that this story will not address whether APS tests scores registered "dramatic improvement." We’ll leave that question for another time.
More importantly, when Hall got wind of our story, she decided to revise the sentence we are checking. Now it reads "the results of the standardized tests administered in 2010 and 2011 under this enhanced security indicate that dramatic improvement in test scores has remained." Since her unrevised statement was online for more than a week, we think it’s still fair game to check it.
Now on to Hall’s statement. We asked Hall’s lawyer for proof, but he did not respond. Fortunately, we had plenty of other sources.
One was the Governor’s Office of Student Achievement, which launched a statewide analysis of 2009 CRCT tests the same year a probe by The Atlanta Journal-Constitution found improbable gains in test scores. At two schools, the odds of making such leaps were less than 1 in a billion.
GOSA Executive Director Kathleen Mathers said Hall is wrong about 2010’s results.
GOSA flagged 16 Atlanta elementary and middle schools for possible cheating on that year’s test. This means that 11 percent or more of their classrooms showed high numbers of wrong-to-right erasures, which is a sign of possible cheating. Thirteen of these schools had already been flagged for their 2009 scores.
Mathers also referred us to the state investigation that found widespread APS corruption. It uncovered even more evidence of testing impropriety or outright cheating on the 2010 CRCT. Here are some examples:
*Parks Middle School: A teacher saw evidence that a paraprofessional gave the school’s principal a testing booklet that he was not supposed to have. When she reported the incident, the principal pressured her to change her story.
* Dunbar Elementary: One teacher found what he thought to be math formula cheat sheets on every student’s desk during math CRCT testing. He also discovered that students had their social studies books out before the social studies CRCT tests.
* D.H. Stanton Elementary: Teachers said they think cheating occurred because the school’s testing coordinator allowed teachers who were friends to administer the test to each other’s students. Some had a reputation for cheating.
* Dobbs Elementary: A student told a teacher that the person who administered their CRCT gave them test answers.
* Bethune Elementary: A teacher said that the principal allowed students who live out of zone to stay at the school if they had good scores.
So much for Hall’s claim about the 2010 CRCTs. What about the 2011 tests?
GOSA has not yet performed its statewide erasure analysis on these newest APS scores, and the special investigation did not address 2011.
The AJC did publish its own analysis July 28.
Five schools registered unlikely increases in their CRCT test scores this year. Three of them -- Toomer, White and West Manor elementary schools -- had also been named in the special investigator’s report for 2009 cheating. West Manor and White were also flagged in 2010.
The odds of such increases range from about one in 700 to one in 21,000, according to the AJC analysis -- not as unlikely as gains in prior years, but still statistically unusual.
School supporters say teaching methods or changing enrollment were behind the increases, but APS’ interim Superintendent Errol Davis and Mathers said they may deserve further scrutiny.
Hall’s claim is clearly incorrect. So how do we rule?
Hall submitted her commentary to Education Week on Aug. 9, according to its opinion editor. This is days after the AJC released its findings on the 2011 scores and more than a month after special investigators released their report.
She had more than enough time to figure out that GOSA, special investigators and the AJC have questioned her results.
Hall’s pants are more than singed. They’re blazing.
She earns our lowest rating on this one. Pants on Fire.