For those of you who haven’t seen the ubiquitous headlines this summer, our country’s largest teacher cheating scandal came to light in Atlanta this July. According to the Georgia Bureau of Investigation, at least 178 teachers and principals in Atlanta Public Schools cheated to raise student scores on high-stakes standardized tests. This was in no way passive cheating; teachers changed their student’s answers before handing over their answer keys and one principal even went so far as to host a cheating pool party with her teachers! According to Atlanta Mayor Kasim Reed, “There is no doubt that systematic cheating occurred on a widespread basis in the school system” and now Atlanta’s award winning gains on standardized tests are all being called into question.
Yet Atlanta is in no way the first major metropolis to be implicated in scandal. Baltimore is now under the gun for similar reasons after numerous erasure analysis tests of student answer sheets have indicated that an extremely high proportion of answers were changed. Of course, all of this comes on the heals of the Chicago cheating scandal, which became famous in Steven Levitt and Stephen Dubner’s bestseller Freakonomics. All in all, our executive and legislative branches have plenty of good reasons to ensure that scarce funds and grant monies are going to deserving schools (and not cheaters!) and with the prominence of Obama’s Race to the Top Program and the re-authorization of the Elementary and Secondary Education Act on the horizon, this phenomenon needs to be better understood so quality control measures can be put into place. There is no faster way to lose public support for the new education reform and accountability movement then to make voters think that recent improvements in scores are all illusory.
In my research paper, I want to use a dataset of Atlanta schools to statistically analyze the characteristics of the schools implicated in the cheating scandal and come to some understanding about which factors likely influenced the cheating. Are schools that received money from private foundations more likely to cheat? What about schools facing sanctions under No Child Left Behind regulations? How do policies such as teacher promotion based on student scores, value added models of teacher evaluation, and merit pay factor in to this cheating?
I then want to pivot to the question of how to best design these teacher incentives to maximize student performance while minimizing the possibility of cheating and discuss how these newly heralded reforms square with psychologists understanding of motivation. After all, in Daniel H. Pink’s new bestseller Drive, Pink writes “external rewards and punishments-both carrots and sticks-can work nicely for algorithmic tasks. But they can be devastating for heuristic ones. Those sorts of challenges-solving novel problems or creating something the world didn’t know it was missing-depend heavily on Harlow’s third drive which holds in part that ‘intrinsic motivation is conducive to creativity; controlling extrinsic motivation is detrimental to creativity.’” So in the end, are these cheating scandals cause to reevaluate policies like promotion based on scores? I certainly don’t know the answer, but hope to learn a lot more by the end of my paper!