Teacher Evaluations: We Need Trust, Not Just Tools
Note: Robin Lake, director of the Center on Reinventing Public Education (CRPE) at the University of Washington, is guest-posting this week.
As almost everyone admits, the traditional teacher evaluation system is about as useful as the old Mac Plus buried in my basement. As we work to improve it, we find ourselves immersed in meetings, studies, and nasty political fights. The friction is fine; we need change, and change begets conflict.
But all of this talk is missing a critical question. We're so focused on the evaluation tools themselves--mechanical issues, like how to factor in student performance indicators and what form observations should take. But no matter how well we design the tools, they won't be effective as long as schools are so deeply steeped in traditions of isolation and mistrust.
I spend a lot of time studying charter management organizations (CMOs), which are often obsessive about ongoing evaluation and improvement. But as an upcoming report from my CRPE colleague Michael DeArmond shows, the evaluation tools that CMOs use are not especially sophisticated. In fact, in effective charter schools, annual or twice-annual evaluations play a far less important role than ongoing feedback and coaching. Test scores often have a place in CMO evaluations, but so do many other factors. CMOs tend to base bonuses and career advancement on a definition of high performance that includes principal judgment but is not necessarily hard-wired to test-score gains.
What effective CMOs do rely on heavily is trust, relationships, and clear communication. In well-run charters, there is a common belief about teaching and learning, and teachers are hired and retained based on whether they share that belief. Teachers know they are getting ongoing feedback and no surprises. They know that the principal doing their observations and evaluations is a master teacher operating on the same definition of good instruction as they are. They know that every other teacher in the building is a potential collaborator. In other words, they trust their coworkers and operate in a culture of common understanding and mutual respect. Evaluation is understood to be more about organizational improvement than about passing judgment on an individual. In fact, some CMOs have tried and dumped merit pay because they felt it disrupted this collaborative culture.
It would be a major omission not to address the need for organizational trust and coherence as we think about how teacher evaluations might impact teaching and learning. Tony Bryk and others have written about the importance of trust, which is a foundational aspect of effective schooling. Without trust, educators come to rely on rigid practices and rules, grievance procedures, formulaic annual evaluations, and highly structured pay ladders. But they also come to resent rigidity and the people who enforce it.
So when teachers unions complain that they don't trust that evaluations will be used well, they have a point. According to the Center for Educational Leadership, most principals are not sufficiently trained to recognize good instruction. Principals have traditionally been selected for their ability to manage programs and crises, rather than how well they provide feedback to teachers. Districts rarely hold principals accountable for results or conduct meaningful evaluations of their management abilities. And when teachers get feedback that they need to improve, they rarely get the training they need to do so, because professional development is so rarely linked to what is happening in schools and classrooms.
Still, teachers unions are often obstructionist and disingenuous when they say they won't support new evaluation systems because they don't trust principals to administer them fairly. They also assert that the problem is not poor evaluation systems, but instead the unwillingness of principals to address poor performance. Yes, principals and central office staff are reluctant to act on incompetent teachers, but that is largely because of aggressive, antagonistic union tactics and a deep-seated mistrust that certainly flows both ways. Principals often balk at firing bad teachers because of the massive time commitment involved, fear of being grieved (or worse, sued) if they don't adhere strictly to contract procedures, and the great likelihood that for all their effort, the teacher in question will find a loophole that extends the process another year or more.
Even if evaluations improve, it's hard to imagine teachers will adjust their instruction based on evaluation feedback if they don't respect their principal. It's also hard to imagine principals will do meaningful evaluations if they think they'll be grieved at every turn. And even good-quality evaluations are unlikely to be powerful if they happen at most three times a year, as opposed to the regular coaching, feedback, and support we see in high-performing CMOs.
As we move forward with new evaluations, we need to step back from the mechanics and pay attention to organizational capacity and relationships. How pervasive and legitimate is mistrust between teachers and principals? How can principals create management practices, coaching supports, and organizational cultures that will allow them to use evaluation tools to their greatest potential? What must unions do to build trust and ensure that principals get rid of bad teachers? And what kinds of school accountability programs put principals and teachers in the same boat so they all appreciate the importance of making their schools as effective as possible?
We should strive to answer these questions just as fervently as we work to perfect inter-rater reliability and our value-added formulas. Achieving any of this is elusive, but it's all imperative.