The Madness of Teacher Evaluation Frameworks
Schmoker, Mike, Phi Delta Kappan
The clouded language of educational theorists hinders thought and under-standing.
Once again, we're rushing headlong to embrace yet another unproven, hastily conceived innovation in the hope that it will improve school quality. We've seen this hap-pen before with "strategic planning," the development of state standards, the worst aspects of No Child Left Behind (some were good), and with so-called school turnarounds.
All of these are good ideas gone bad, and they have the same essential features: They were overly complex, unproven, and premature. They were implemented on a national scale before they were sensibly piloted, then refined on the basis of early, small-scale implementation. They suffer from what Jay Matthews calls "all at once it is"--our pathological insistence on launching all aspects of an innovation simultaneously, everywhere, in the absence of evidence that it works. No one asked the obvious questions: Does this innovation have a track record? Could it have unintended consequences or could it displace much higher priorities that would guarantee a better education for all, e.g., ensuring that every teacher is furnished with a decent, coherent curriculum, without which effective teaching is difficult or impossible?
Our silver bullet du jour is teacher evaluation (a good thing)--on steroids (a bad thing). It is being driven by popular, time-gobbling, anxiety-inducing evaluation "frameworks." Don't misunderstand me: I have always been a fan of simple, effective teacher observation and evaluation, which I'll describe in a moment. Good teacher evaluation is a critical force for improvement. I'd even like to see carefully piloted inclusion of assessment scores in evaluations, but only if the assessments truly represent legitimate, curriculum-based knowledge and skills for each respective course. (We've never had this; we don't have it now.)
My complaint is with the frameworks themselves--their sheer bulk and their sloppy, agenda-driven language. They're absurdly long; teachers are desperately trying to design lessons to meet criteria described in as many as 116 categories (Anderson, 2012). Administrators are expected to use these unwieldy instruments to conduct up to six full-period observations per teacher per year and to conduct both preobservation and postobservation conferences for each observation with every teacher.
Much of the criteria itself is both misguided and ambiguous - written in that thoughtless, tortured prose that continues to mar the education profession. For example, one framework calls for lessons to include "simultaneous multisensory representations" and "facilitation... that results in students' application of interdisciplinary knowledge through the lens of local and global issues." Teachers must "facilitate content accessibility" by assembling or modifying curricular materials at the "individual and subgroup level"--even though the best teachers do no such thing (Poplin et al., 2011). The effective educator is supposed to "solidify learning after constructed experience with clear labels" and with "articulation of metacognition" (among the murkiest words in the education lexicon).
The designers of another popular framework defend their similarly elaborate instrument with talk about "proximal processes" based on "multi-level, latent structure" and "varying degrees of molarity/discreteness." The instrument purportedly "reflects the developmentally relevant construct of heterotypic continuity" in the pursuit of (the ever-present) "metacognitive skills."
I don't know about you, but I'm very nervous entrusting our children's futures to people who write--who think--in this fashion.
Would it work?
Another popular framework puts teachers on notice that lessons must "accommodate prerequisite relationships among concepts and skills," as well as "reflect understanding of prerequisite relationships among topics and concepts and a link to necessary cognitive structures. …