A simmering scandal in Atlanta over cheating on standardized tests came to a head this week as state investigators released a report that found in the city’s schools “an enterprise where unethical—and potentially illegal—behavior pierced every level of the bureaucracy,” according to the Atlanta Journal Constitution. The scandal follows closely on the heels of a USA Today investigation into possible cheating in the Washington, D.C. schools. The Hechinger Report talked with Robert Tobias, director of the Center for Research on Teaching and Learning at NYU’s Steinhardt School of Culture, Education and Human Development, and former head of assessment and accountability for the New York City schools, about whether high-stakes testing inevitably leads to cheating, and how it might be avoided.
Hechinger Report: There have been two major stories this year about cheating, there was the USA Today series including the reports on cheating in D.C., and now there’s the Atlanta scandal. What do you think went wrong?
Tobias: First of all, this is not the first time we’ve had cheating scandals. You can go back to New York City, back in the late 90s, where we had a pretty major cheating scandal. In many ways it’s not surprising, because of the current emphasis on high-stakes accountability using standardized tests, and standardized tests almost exclusively. It almost encourages some people to do the wrong thing. So as you kick the stakes up, people are going to focus on the metrics that will be used to determine their fate. They’ll be looking for ways to elevate those metrics, and some people will try to take a short route.
HR: Given that, is cheating inevitable now?
Tobias: I’m not going to say it’s inevitable, but you’re increasing the motivation and the probability for this kind of thing to occur.
HR: Is there a way to use testing to judge schools and teachers, without having cheating as an outcome?
Tobias: It would be idealistic to believe that we would have a system that focuses on only on improving, using data for teacher improvement and student improvement. As long as you require accountability – and I don’t think accountability is a bad thing, I think you have to have accountability – but as long as the accountability is going to be so heavily dominated by testing, and that coupled with targets and goals that in many cases are unrealistic, that encourages cheating. If teachers and administrators believe that the system has established standards that are just unattainable, and their futures are determined by those standards, it leads them to seek ways to beat the system. Cheating is one of those ways.
HR: What would your ideal system look like for accountability?
Tobias: Certainly it has to have multiple components. At the level of the teacher, and teacher effectiveness, you have a movement now to base the determination of teacher effectiveness on value-added test performance, and that’s all based on standardized tests. To me, that is not only narrowing the curriculum, but that’s the kind of system that creates an atmosphere where more cheating will occur.
If instead you have an accountability system that has multiple components, standardized tests as part of that system, but in addition to that, teacher observations, analysis of student work, teacher self-portraits and reflections, so multiple components to evaluate a teacher, and those multiple components [are] used to identify ways to provide supports for the teacher, professional development, to assign coaches, ways to focus on improving teacher performance and effectiveness – that’s the kind of system that would, I think, reduce the incentive to cheat. It creates a whole different culture and atmosphere in a school.
This transcript has been edited for length and clarity.