by Emile Session
Teachers, like students, respond how they’re assessed. I used to work as a classroom observer for an opt-in teacher evaluation program, grading teachers on a rubric that was used for coaching and feedback. The reactions I got when I showed up varied: great veteran teachers had specific questions about the rubric and how to improve; new teachers were curious but the rubric was just one more piece of an overwhelming amount of things they were learning; and ineffective, checked-out teachers just wanted to know if their union rep had been informed. Now, New York City is implementing a new evaluation system based on a mix of observations and student test scores. On paper, the goal is better teaching. Based on my experience, if the only teachers who productively engage with the feedback are the ones who are already teaching well, then no amount of statistical sophistication and labor negotiations is going to make the new system any more effective than the old one.
Observations are better than tests for evaluating teachers. When I was a teacher I got an effectiveness score that stated that New York City was 95% certain that I was between the 10th and 54th percentile ranks of NYC teachers. In plain English, they couldn’t be sure if I was in the bottom tenth of teachers, or slightly better than average. It is that hard to give a precise estimate of how good a teacher is based on test scores, because of how many other things impact those scores. However, politicians and policymakers are under pressure to publish hard numbers even when they haven’t finished assessing the quality of the test they’re using, so New York City published that within that 44-point wide margin of error, its best guess was that I was in the 31st percentile rank. Meanwhile, the fact that after 3 years I was at best a middle quality teacher was obvious to anyone who spent an hour in my classroom.
This is why the Gates Foundation, generally a supporter of using test scores in evaluations, says that observations are the most accurate assessments of teacher skill. The hard part is the cultural change: how do you implement an observation system so that teachers and their supervisors co-create feedback that is specific, timely, and achievable for the teacher, and ultimately positive for students? You have to balance principals’ already-packed schedules, coordinate meaningful feedback and follow up conversations with teachers and their coaches, and build the whole system into the rhythm of the school year. Taken together this means observations either need to be prioritized as a core part of a principal’s job, or spun off into a dedicated position.
An older researcher once told me that “Accountability is what’s left when responsibility has left the building.” To grow as a profession, teachers need greater say the selection, evaluation, and in some cases expulsion of people from the profession. They need a career path for becoming instructional leaders, instead of administrative work being the only way up. A functioning observation system can work better than a value-added measure (or a union-negotiated pension) for attracting and retaining the right people in the system.
About the Author, Emile Session
Emile Session lives in Brooklyn, NY and works for Achievement First, a network of high performing public charter schools. He has worked as a math teacher, education researcher, and classroom observer, and is a graduate of Cornell and Columbia University.