In one of his first moves as interim Atlanta school superintendent, Erroll Davis announced that any unusual gain in test scores would trigger an automatic investigation into how those gains were achieved.
It’s such a simple step, and so obvious. If you have any real interest in protecting the integrity of a high-stakes testing regimen, instituting that kind of policy would seem essential.
In fact, if Atlanta Public Schools officials had implemented such common-sense safeguards back when credible allegations of cheating on test scores first began to draw attention, a lot of this controversy could have been avoided. There might have been no state investigation, and no international scandal.
But I guess that’s the point. Character is fate. The previous APS leadership, including the school board, was incapable of taking such a simple, proactive step. Maybe they feared that by implementing safeguards against cheating, they would have signaled that cheating was a problem. Maybe, at some level, they understood that tighter oversight would have made it difficult to maintain the system’s Ponzi scheme of ever-rising test scores. Whatever the reason, it did not happen, and an opportunity was lost.
As a silver lining to a very dark cloud, however, the Atlanta scandal has at least forced a national conversation about high-stakes testing and related issues. As that debate continues, it’s important to note that changing wrong answers to right is only one way — perhaps the least subtle way — of manipulating educational data and offering a falsely rosy image of achievement.
For example, then-Gov. Sonny Perdue and his Governor’s Office of Student Achievement deserve a lot of credit for refusing to ignore evidence of widespread cheating in the Atlanta public schools and pushing for a full investigation.
However, as Perdue was helping to drive that probe, he was also citing GOSA statistics claiming that during his two terms as governor, Georgia’s high-school graduation rate had risen from 63 percent to more than 80 percent, just as he had promised.
To be blunt, nobody believes those numbers, which are barely more plausible than the data that Beverly Hall was touting. The equation used to generate them is widely acknowledged as flawed. While Georgia’s high-school graduation rate has probably increased over the past decade, the real number is probably closer to 60 percent than 80 percent.
In another example, state officials announced this spring that 96 percent of Georgia eighth-graders had met or exceeded state standards for reading. According to the same state-mandated standardized tests involved in the Atlanta scandal, more than a third of Georgia eighth graders actually exceeded state reading standards.
That kind of number — only a 4 percent failure rate — tells parents, students, voters and politicians that Georgia schools are doing well. But do you really believe that 96 percent of Georgia students can read as they leave the eighth grade? Or should that number set off the same internal warning system that Hall and others at APS chose to ignore?
The National Assessment of Educational Progress, given to a cross-section of students from around the country, offers a contradictory assessment. While just 4 percent fail Georgia’s test, NAEP reports that 28 percent of Georgia eighth-graders lacked basic reading skills. Among students eligible for the national school lunch program, 40 percent failed to reach the basic level of reading skills.
Nationwide, eighth-graders in 32 states outperformed Georgia students on the NAEP reading test; only nine states did worse.
“We have a duty to require that student test results reflect real learning,” Gov. Nathan Deal said earlier this month in announcing the results of the Atlanta cheating investigation. Unfortunately, Atlanta isn’t alone in falling short of that goal.