In August of 2010, a briefing published by a formidable team of researchers at the Economic Policy Institute warned policy makers against emphasizing student assessment outcomes in teacher evaluation formulas. At the time, many states were proposing new data models for measuring individual teacher contribution to student growth (e.g., value-added model, or VAM) to help systematically weed out “ineffective” teachers and reward those deemed “highly effective.”
The briefing stated concerns with the growth models themselves, calling them “unstable” and incomplete in accounting for influences on student test scores that could wrongly be attributed to individual teachers. And then they predicted what would happen as a result of associating these metrics with teacher evaluations. The list is bleak—
- Teachers would be disincentivized to work with students perceived to be most challenging to move (e.g., English language learners, those with disabilities).
- Teachers would “narrow the curriculum” even further than what school accountability formulas encourage, prioritizing content and skills likely to improve teacher effectiveness ratings rather than that which will be required for all students to become “effective participants in a democratic society and contributors to a productive economy.”
- Schools would become less collaborative as teachers compete for individual bonuses.
- Even effective educators would become demoralized with increased pressure and decreased autonomy, possibly leaving the field altogether.
Any of this sound familiar to you now? Reading this brief a decade down the road is chilling, like the authors had already seen the horror film and were yelling at the screen, “Do not go in there!”
But we did go in there, many of us with pure, but naïve, intent. And I would offer the list of unintended consequences was even longer and more damaging in the end. [What would you add to the list above?]
Rubbing salt in the wound is this EdWeek headline from November 2021: “Efforts to Toughen Teacher Evaluations Show No Positive Impact on Students.” The article refers to the findings of a national study of teacher evaluation systems that stemmed from Race to the Top. After all the money spent and the emotional toll extracted of teachers, there was no real improvement?? [shakes fist at the article]
Given the “tough on teachers” mentality propagated legislatively over the last ten years in many states, we should not be surprised that VAM is now a dirty word in most places. Rather than embrace academic growth measures as useful and appropriate tools for reflection, teachers often reject them for fear they will be unfairly judged.
The good news is that so many people are so dissatisfied with the status quo that legislators now appear open to reconsidering some basic assumptions, like whether states should have end-of-year assessments at all, and whether data from those assessments should be included in teacher evaluations.
It’s helpful to remember we got into this mess by attempting to answer the question, “How can we use state testing data to hold individual teachers accountable?” In hindsight, it appears this was the wrong question entirely. Rather, we should have been asking ourselves, “How can we use academic outcomes data to improve teaching?” Not because teachers are currently doing it wrong, but because our current outcomes (e.g., NAEP results captured just before the pandemic) tell us it could be better.
At K12 Lift, we help principals, teachers, and coaches routinely use their assessment results to figure out which students are benefiting most and least from current instructional practice, reflect with their peers on why that is, and make well-considered adjustments in response. And we have seen first-hand how these healthy habits around the use of data build collective efficacy across teams.
Our plea is that policy makers use this seeming inflection point to cleave student outcomes data from teacher evaluations altogether, thereby leaving the data unadulterated for their noble purpose—answering student-centered questions motivated by a desire to incrementally improve teaching practice.
What would change for teachers and students in your district if academic outcomes data were used solely for purposes of reflection rather than judgment?
