Yesterday, we met three teachers – Ms. Holmes, Mr. Chandler, and Ms. Spade – along with their principal, Ms. Doyle. Ms. Doyle observed each teacher deliver a tenth grade English lesson, and was now in the process of evaluating their work.
When evaluating each teacher, Ms. Doyle examined the results of the assessments each teacher had given his or her students to measure mastery of the lesson’s objective. She also asked each teacher to examine the results independently and offer their own conclusions during the observation debrief.
For the methodical Ms. Holmes’s lesson on identifying elements of similarity between poems, her mini-essay assessment found most students able to make connections at the level of plot, setting, and character and roughly half of students able to make connections between the themes of poems. It was clear from reviewing her students’ assessments that the difficulties with thematic connections came from an incomplete understanding of theme as it relates to poetry.
The discussion-focused Mr. Chandler found that half of his students were consistently able to write research questions that met his expectations for clarity, focus, and depth, a quarter wrote good questions about half the time, and a quarter wrote questions that all needed improvement. Since his assessment only asked for students to submit a list of questions, he and Ms. Doyle couldn’t be sure based on the data about where learning broke down for the students still struggling with the skill.
As for Ms. Spade, two-thirds of her students ranked themselves as comfortable connecting an author’s text to its historical context, while a third expressed less comfort. Ms. Doyle, however, noted that assessing students’ confidence doesn’t say anything about how well they can actually perform the skill in question.
These three results demonstrate the importance of having good assessments of student learning. Creating assessments that (a) measure mastery and (b) are sufficiently detailed to explain which particular sub-skills were mastered and which require more work is an important skill for teachers that strengthens over time. Good assessments are critical to understanding the real effects of actual teacher actions, and assessments – like the MCAs – that don’t provide that kind of information are useless for meaningfully evaluating teachers.
Good teacher evaluation requires actual knowledge of a teacher’s classroom and practices. It also requires good evaluators. The three teachers we’ve looked at are fortunate to have a principal who understands their content area and knows how to conduct an evaluation with an eye towards encouraging growth rather than rooting out teachers to judge or punish. Too many teachers don’t have that advantage, which means that too many students have teachers that aren’t getting the kind of support they deserve.
Even given a universally excellent corps of knowledgeable, fair-minded principals, teachers should be evaluated by more than just their principal. Peer evaluations, especially by master teachers in their school or district, can be very helpful. Ms. Doyle may be a great evaluator for her English teachers, having previously taught that subject, but the social studies and music teachers at her school deserve evaluators who know their subjects, too.
Evaluators also need to have a fair system for assessing teachers, and they need to understand how to use that system. Giving an evaluator a rubric and a one-hour workshop is not enough, and poor use of an evaluation system is a great way to discredit its worth in the eyes of those being evaluated.
Beyond who is doing the evaluating, we also need to consider what the response to those evaluations will be. In general, the goal of teacher evaluation should be about encouraging growth and development. Teachers who receive negative evaluations from multiple evaluators deserve support (in the form of mentoring from master teachers and/or the school’s instructional coaches) to help them improve. It is only when a teacher shows a persistent unwillingness or inability to improve that evaluations should be used to remove them from the classroom.
As for determining which teachers to lay off when the budget requires it, we need a clear, fair system with an appeals process to be in place and well-understood before we replace seniority as the default mechanism for determining layoff order. The quality of teaching demonstrated by Ms. Holmes, Mr. Chandler, and Ms. Spade has little or nothing to do with their students’ test scores, and those scores do not have a place in determining which of the three to let go (assuming layoffs came down to one of the three).
Quality teaching is demonstrated by so many things other than test scores. “Student data” or “student outcomes” deserve a central role in evaluating teacher effectiveness, but policymakers should take care not to conflate those concepts with standardized test scores. We need many strong evaluators participating in a growth-oriented, fair process that encourages a professional mindset of continuously increasing effectiveness instead of constant fear about losing one’s job based on test scores or a single dog-and-pony-show observation. Our students deserve teachers who are supported in getting better, not teachers intimidated by an unfair system.
This story originally appeared at mn2020.org