Guest post by John Thompson.
What would have happened if the Gates Foundation's Measures of Effective Teaching (MET) research had been conducted before the "Billionaires Boys Club's" preferences were codified into law in Race to the Top and the Department of Education's NCLB waiver requirements? How would their final report, "Ensuring Fair and Reliable Measures of Effective Teaching," read if its findings were reported before value-added evaluations were imposed on the nation's schools? Even if the same researchers had used the same methodology and made identical findings, how would that evidence have been presented?
Had the Gates Foundation looked at the evidence before foisting test-driven accountability on states, their prime conclusion might have been rephrased as, "our analysis should give
heart pause to those who have invested considerable effort to develop practices and policies to measure and support effective teaching."
It would be too much to ask for scholars to analyze the policy evidence and then echo the old Gilda Radner skits on Saturday Night Live, "Never mind!"
But, surely the Gates Foundation knows that it is improperly spinning the evidence when it asks the concluding question, "When high-stakes decisions must be made, can these measures support them?" (emphasis mine)
Of course, those measures can be valid and reliable. However, the real world question is how often would those metrics prove to be reliable? How often would those measures drive the unfair firing of effective educator? Systemically, how many unfair terminations can districts make before educators stampede out of schools where it is harder to meet value-added targets? And, how often will value-added lead to more primitive bubble-in instruction and curriculum narrowing?
Similarly, had Gates done the research and identified the relatively better and worse value-added models, maybe they could have avoided the hypocrisy that Bruce Baker describes. Baker explains how districts have already adopted models that are not reliable. But, after pressuring systems to rush value-added into high-stakes decisions, Gates now remains silent about the improper use of its evaluations. So, Baker concludes, "the new Gates findings - while not explicitly endorsing use of "bad" models - arguably embolden this arrogant, wrongheaded behavior!"
As their previous research shows, the MET found little reason to believe that forcing teachers to concentrate on basic skills, to impose rote instruction, to mandate nonstop test prep, and to narrow the curriculum is a viable way of increasing student performance, even on primitive state tests. But, in the three years since the foundation started to help coerce states into high-stakes value-added that is precisely what has happened. And, now, those retrograde policies are codified into law.
The MET got one thing right when it finally observes that "heavily weighting a single measure may incentivize teachers to focus too narrowly on a single aspect of effective teaching and neglect its other important aspects." Had they entered into the study without preconceived notions, however, might the Gates scholars have noticed that teachers aren't the only stakeholders who were incentivized to engage in educational malpractice and wonder whether we were voluntarily driving the joy of teaching and learning from our classrooms? Had they conducted research before engineering a change to state laws, researchers might have also stumbled across principals, districts, and state systems predisposed to mandating the teaching-to-the-bubble-in-test. In the course of conducting research, they might have questioned their assumption that teachers are like Lemmings jumping off the cliff and seen that it is our bosses that are pushing us off by requiring paced instruction in order to cover their own rear ends.
If "reformers" had researched first and made the policy decisions after evidence was gathered, would they have mandated value-added on the eve of implementing Common Core? We can debate the merits of Common Core, but it is hard to see how value-added is not helping to doom that experiment. The MET now finds lower correlations between value-added ratings for teachers on state tests and ratings for the same teachers with the same kids on higher order tests - tests that resemble Common Core assessments. In other words, educators, who are now coerced into teaching basic skills to protect their jobs and their schools' survival, will soon be expected to turn on a dime and teach a college-prep curriculum.
If districts want Common Core to succeed, they now face a number of frightening scenarios. The least-destructive approach would be for districts, having already spent big bucks on statistical models, computer systems, and the rest of the Gates-endorsed rubrics, to put the value-added on the shelve for several years. After there are enough years of Common Core results to put into their algorithms, and if they still wanted to continue the teacher-bashing, they could still fire away. But, that would require a level of patience that contrasts with the corporate reformers' ethos of "fire," "ready," and then "aim." Or, more likely, districts will terminate teachers using the results of very different tests in a statistical model that was not designed for Common Core-type scores.
I would hope that the courts would make sure that districts who take that route pay dearly for that impatience. In fact, I would hope that Gates scholars would seek to testify in court regarding the impropriety of misusing value-added in such a manner.
On the other hand, the foundation provides a valuable service in documenting the common sense conclusion that, "Additional shorter observations can increase reliability. Our analysis suggests that having additional observers watch just part of a lesson may be a cost-effective way to boost reliability by including additional perspectives." Whether those solid conclusions were worth $50 million is their call.
And, were it not for the accountability hawks' Sword of Damocles we could also gain a great deal of insight from the MET's video library of classroom instruction. Above all, we could use student surveys as essential diagnostic tools for setting policy and creating respectful learning environments.
I'm afraid, however, that such findings are too mundane and incremental for billionaires. Apparently, their hubris can only be satisfied by "transformational" theories that produce "disruptive innovation." So, it is no surprise that the Gates scholars now conclude the preferences of Bill Gates et. al were right all along. It takes a lot of convoluted wordplay to make the MET policy findings seem consistent with the evidence, but that is what happens when the big boys act on their whims.
What do you think? Why would we have gambled so recklessly on value-added evaluations before thinking through their policy implications? Wouldn't it have been nice to read an analysis of the MET research that did not have to be shoe-horned into predetermined conclusions?
John Thompson was an award winning historian, with a doctorate from Rutgers, and a legislative lobbyist when crack and gangs hit his neighborhood, and he became an inner city teacher. He blogs for This Week in Education, the Huffington Post and other sites. After 18 years in the classroom, he is writing his book, Getting Schooled: Battles Inside and Outside the Urban Classroom.