On Cheating, Blame, and Accountability
I must confess that I do check my blog stats once or twice a day. I’m not driven by pursuit of big numbers, though of course I’m pleased when a post is widely shared and read. But I’m drawn to the stats page because of curiosity about search terms that lead to this blog, about old posts that suddenly find new life for unknown reasons. The stats also show that I’m often wrong about which posts I think will be more popular. The ones I like the most often fade quickly, while posts I’m less invested in sometimes take off.
Case in point: a Facebook friend from New Zealand shared the story about a school cheating case where teachers tampered with test results reported to the national authorities. I noticed similarities to incidents in the U.S., and made the connection to Pasi Sahlberg’s talks and writing about the Global Education Reform Movement – or GERM. I cranked out that blog post in a few minutes and figured it would be a blip on the stat sheet. Surprisingly, it has been the most read and shared post in the past month.
The post also drew a response from Benjamin Riley of the New Schools Venture Fund, who may have noticed this particular post because he’s currently on leave from NSVF and spending a year working in New Zealand. (Nice work if you can get it! I loved my visit there a few years ago).
Here is the distillation of Riley’s response and suggestions (though if you go read the full version at his blog, you get the benefit of his use of White Stripes lyrics). He argues that the inevitability of cheating on tests shouldn’t be used to argue against testing any more than cheating by golfers leads to the end of golf; it simply means we must guard against cheating. Riley also quotes Kevin Carey suggesting that such cheating is short-sighted if inflated scores will end up inflating expectations for subsequent years. Carey adds that “cheating also means that public schools finally care enough about student performance that some ethically challenged educators have chosen to cheat. This is far better than the alternative, where learning is so incidental and non-transparent that people of low character can’t be bothered to lie about it.”
Overall, I can agree with Riley that individuals are responsible, and that cheating by itself is not an argument to eliminate testing. There are some appropriate uses of large-scale standardized assessment. I don’t agree with Carey that an uptick in cheating indicates people “care enough about student performance.” I think it means those people are mad or fearful about the public uses of what passes for “student performance” but really isn’t.
Riley and Carey seem to assume that standardized testing generally produces useful information about students, teachers, schools, and systems – from the individual level on up – and so they tackle this issue with a focus on what educators should do without engaging around one of the key underlying problems: weak tests, or good tests used for weak policies, are central to this story. So I’m asking what education leaders should do to address the problem. I wouldn’t be satisfied with an answer that puts the problem entirely on the teacher, any more than I would accept a teacher who says all the problems in the class are the students’ fault.
Their perspective also seems a bit removed from an understanding of classrooms and schools. I’d suggest that Carey’s view suggests people think more about cheating rationally rather than emotionally. I don’t think it works that way. People who cheat are likely angry or insecure. It’s also important to acknowledge that in some of the high-profile cases in the U.S., there’s evidence of cheating at the school and administrative levels, which calls for a different set of models in trying to understand the psychology of the act, throwing in group dynamics and the possible role of intimidation.
Riley’s main point, the one linked to the lyrics, is that you can’t blame the test for the cheating, any more than you can blame the bank for the robber. And if you’re addressing the cheater, or the robber, I agree: pushing off one’s own misdeeds on others doesn’t negate or excuse the misdeed.
But as someone who has administered thousands of tests and been responsible for learning outcomes, I accept an accountability that Riley and Carey seem less interested in ascribing to “the system” that gives the larger tests, and should be responsible for broader outcomes. This is my fundamental disagreement with much of the education reform notion of accountability. The people with the most power are supposed to bear the most responsibility. When I give a test, I try to design an assessment that is fair, useful, valid, productive, and worth the effort. To attach high-stakes to an exercise that doesn’t meet those criteria is to invite cheating. I’m not excusing the cheaters, but it would be sloppy, unprofessional work on my part to create conditions that I should have known would ultimately undermine my work. If a bank has inadequate safeguards against robbery, it’s not exactly their fault if they’re robbed. But isn’t someone supposed to be accountable for having the foresight to reduce the chances? If a CEO creates the conditions that push more of his managers and accountants to cook the books, and at the same time says be honest, then does the CEO bear any responsibility for corrupt practices that should have been anticipated?
Where’s that accountability in education policy? Okay, punish the cheaters. You can even try to tighten test security, but ultimately, that system must rely on local practitioners without creating an undue administrative burden. So, policy makers, we can easily predict that the more you rely on standardized tests for purposes they can’t adequately measure, and the more you raise the stakes, you are creating conditions that lead to more cheating. It will happen. It shouldn’t. People should resist. They should do the right thing. They should raise objections in appropriate venues. They should be honest and transparent. Those are wonderful sentiments that serve to distance you from knowingly following a series of steps that will have a negative effect on schools and on your own accountability measures.
Our Secretary of Education, and many state superintendents and legislatures have pushed more and more use of mediocre tests for inappropriate, high-stakes uses. People shouldn’t cheat, but neither should our leaders avoid criticism for their failure to produce an accountability system that works. They’ve ignored too much of what we know about students, teachers, schools, learning, and human nature. They’ve failed as policy architects and leaders who should have foreseen the mess they’ve helped create.