When I first read over the summer about the unbelievable news that 176 teachers and principals in Atlanta, Georgia were being accused of cheating on state standardized exams, I immediately knew that I wanted to write my paper on this relatively new phenomenon of large-scale teacher cheating scandals. To me this was a fascinating topic: the motivation for standardized testing is the collection of data that policymakers hope can be used to improve student outcomes, identify best practices, and reward the best of the best. These teacher cheating scandals were therefore the absolute worst thing that could happen in this new era of education reform: they were giving policymakers factually incorrect data. While originally I planned to perform an econometric analysis of which teachers were cheating by creating a mathematical model using student characteristics and teacher characteristics (age, experience, previous test results) as predictors of cheating in a regression framework, I found this had already been done by Steven Levitt and Brian Jacobs and published by the NBER.
I then turned my focus to more qualitatively exploring why teachers feel the need to cheat (by reviewing state investigator’s reports and first hand teacher confessions) , reviewed the literature on how common experts believe the phenomenon to be, described the methods teachers use to cheat (by reviewing investigators’ reports), discussed cheating in the context of the psychology of motivation and rewards (by reviewing the findings of self-determination theory, which moves away from the totally rational model of human behavior), and proposed policy changes to reduce the incidence of cheating and its deleterious effects on education reform.
Essentially, I propose two overarching solutions designed to incentivize people to care about the phenomenon. First, I propose taking the investigations of possible cheaters out of the hands of schools who have an incentive to cover up cheating and putting the state in charge of conducting thorough investigations. Secondly, I propose that the federal Department of Education include testing integrity policies as part of its holistic rubric that it uses to grade the grant applications that states submit in an effort to win Race to the Top grant money (meaning that a state gets points for actually taking the preventive and investigative steps to ensure that scores reflect true student ability). The hope is that federal involvement will help offset the lack of incentives states and localities have to address the problem on their own.
It is really interesting that this topic has gotten more and more attention as the year has gone on. In September, I thought I was one of the only people looking at this phenomenon but now Secretary of Education Arne Duncan has written to all state superintendents regarding the issue, the Dept of Education has issued a “Request for Information to Gather Technical Expertise Pertaining to Testing Integrity,” (Federal Register), the Wall Street Journal just ran a cover story on this phenomenon in March, and even more exciting to me is the fact that New York State has adapted a bill that redesigns the investigative procedures of teacher cheating scandals towards a state-centric process, just as I advocated in my paper. It now seems that despite picking an “out-of-the-box” topic, people are finally beginning to talk about this phenomenon and prevention techniques. This is really exciting for me because it is really cool to be writing about an issue that has not received much press or scholarship and is finally catching on in the popular press and in academia too.