Only a few months ago Atlanta was held up as the city that improved schools for poor kids without relying on chartering, school closings, or new sources of teachers. Atlanta stayed inside the box, relying on professional development, coaching, and other traditional methods. Now it turns out that positive results from Atlanta could have come from widespread falsification of test score sheets by school staff.
Clearly, it doesn’t take the threat of widespread teacher firings or school closings to induce test cheating. What does this mean? Should districts abandon testing, thus giving up the only means they have to identify situations where kids aren’t learning in time to do something about it? No. But it does mean that educators are not morally different from other people: many wouldn’t cheat under any circumstances, but some will cheat if they can benefit and expect to get away with it. Districts need to anticipate this and take sensible precautions, like delivering tests to schools minutes before testing periods begin, and picking up shore sheets immediately after test periods end. Longer-run and better solutions would involve online adaptive tests for which no one knows what the questions will be before they are asked, and results are available immediately.