From scandal at APS to suspicious scores nationwide

Beverly Hall always boasted that her transformation of Atlanta Public Schools was driven by data. The Atlanta Journal-Constitution’s reporting on the district was also driven by data, but the AJC’s numbers increasingly suggested that Hall’s numbers were an illusion.

The newspaper uncovered the first piece of solid evidence of a widespread cheating conspiracy at APS more than four years ago – a limited analysis showed statistically unlikely test score gains at a single Atlanta school.

As Hall and her lieutenants grew ever more emphatic in their denials, the Journal-Constitution’s data analysis grew ever more sophisticated and wide-ranging. And the newspaper’s reporters kept finding teachers and administrators who whispered into the phone that something was wrong at APS.

In 2009 the Journal-Constitution developed a computer analysis that flagged schools with unusual changes in test scores. This analysis calculated the probability that a class’s improvement from one year to the next could have occurred by chance.

The data showed anomalous gains at a dozen Atlanta schools. At some schools, the AJC found, the odds were greater than a billion to one that scores improved without some intervention, such as cheating. The state soon began a lengthy investigation that culminated in Friday’s indictments of Hall and 34 others at APS.

As the criminal investigation plodded on, the AJC decided to look beyond Atlanta: Was it possible that this was the only big-city school district with artificial gains on standardized achievement tests?

In the fall of 2010, a team of four reporters and data specialists began to assemble several years of test scores from every public school in America. Eventually they analyzed scores at 69,000 public schools; they also vetted their statistical methods with national experts.

In the spring of 2012, The Atlanta Journal-Constitution reported that 196 school districts throughout the U.S. exhibited suspicious patterns of test results that, in Atlanta, indicated cheating. A month later, it reported that dozens of schools awarded the nation’s highest educational honor – designation as a National Blue Ribbon Schools – had highly unlikely test scores.

The response from many of those districts and schools sounded eerily familiar: We don’t have systemic cheating here. We just have high standards.