Something curious happened in “The Mirage,” the recent TNTP report on teacher development: the authors appeared to ignore one of their most compelling findings. It’s too bad, as it was a finding that seemed to corroborate what I have been hearing in my own research.
I am currently working on my dissertation, an interview-based study of teachers describing their most powerful professional learning experiences, and in considering the full complement of stories I have been struck by just how varied they are. I have heard about self-designed courses of study, weeklong university-based institutes, and “expert”-run afterschool PD; experiences that were content-focused, climate-focused, and pedagogy-focused; experiences that drew on district priorities and that followed from participants’ interests; experiences that were required and elective; experiences that involved active inquiry and that involved relatively passive or providential learning. And notably, teachers’ reports of “typical” or memorably negative PD experiences were just as varied. This has led me to see professional learning as a highly individualized experience and to wonder whether the most influential PD experiences are perhaps those that best match each individual teacher’s most pressing dilemmas or keenest interests. (This is hardly a revelation: situated learning theorists and education researchers have been suggesting as much for years.)
Given the questions raised by my own research, then, I was glad to read the TNTP authors write that professional development is a deeply individual enterprise. Across three large public school districts and one charter management organization, they found no single approach to professional development seemed consistently to result in teacher improvement. “What works for one teacher may not work for another,” they wrote (p. 18). And then summing up their findings, they concluded that, “we cannot try to force one solution on some 3.5 million individualized challenges” (p. 34). Such statements seemed refreshingly aligned with the notion of differentiated instruction, namely that students have diverse learning needs and strengths and ought to be taught accordingly. In other words, it is good to have clear learning objectives, but it is maybe less reasonable to have too narrow a path for achieving them.
And so it seemed odd that the TNTP recommendations involved funding a series of experimental impact evaluations of PD programs designed to weed out those least likely to lead to “observable, measurable progress toward an ambitious standard for teaching and student learning” (p. 35). This vision of instructional excellence and the path to achieving it – seemingly intended for the full, varied education sector as a whole – was based on surveying and observing a single charter organization. Generalizability is an elusive concept in even the best designed and most rigorously executed experimental research, so I was dubious that it could be confidently applied here. But more curiously, such a recommendation seemed squarely at odds with the report’s own finding that teachers develop along individualized and perhaps unpredictable paths.
I admit that it is alluring to think that the puzzle of professional development – with its vast potential and decidedly uneven results – can be solved through such objective and scientific means: set a goal, measure progress toward that goal, accept no excuses. Indeed, this has been the operating assumption of PD policy for decades, but I think this very idea that effective PD can be isolated and scaled is the real mirage. It seems possible, but in reality it’s not.
Despite this mirage, the findings in the TNTP report offer a rare and vivid opportunity for us to re-examine this fundamental assumption. If teacher development is indeed so individualized, then maybe we need to consider the possibility that scaling up unitary best practices is likely to leave more teachers behind than to lift them up. Maybe teacher development needs to be predicated on a clear vision coupled with broad and deep flexibility regarding the process for achieving that vision. Tight on ends, loose on means, as Karl Weick might say. We give teachers more time for development and let them choose how to fill their time. As long as students learn what we want them to learn, we let their teachers learn however they best see fit. Prolific teacher-blogger Peter Greene offered a similar vision in his review of the TNTP report. (Of course, there is precious little agreement about how to measure these things, as many people have pointed out, but that's for another blog post.)
Admittedly, a more varied universe of PD options is less conducive to randomized control trials and impact evaluations, but I would argue that it may better model the way teachers learn and improve. And as a result, we might better be able to expect to see what we all agree we want: teacher improvement that leads to student learning.