Add Detroit to the growing list of states and districts looking to revamp their teacher evaluation systems. Test-based accountability reached mania-like proportions during the Arne Duncan years but has slowly abated, with 34 states now requiring the use of test scores to evaluate teachers, down from a high of 43 in 2015. Maine made news earlier this year when its legislature voted to drop a requirement that test scores be used to rate teachers. Thanks to a new law in New York, there’s no longer a requirement that evaluations be tied to the state’s standardized grade 3-8 math and English assessments. Pennsylvania is considering similar reforms. And many states have made other changes to their evaluation systems, with some reducing their frequency and others washing their hands of the whole sordid thing by allowing districts and local bargaining units to work out the details.
What they all have in common is a tacit admission that evaluating teachers on the basis of student test scores is a stupid idea. Of course, teachers could have told them that all along (and we did), but even without test-based accountability, designing a system that fairly evaluates teachers has proven to be a Herculean task. There are a lot of reasons for that, and Detroit’s proposed new teacher evaluation system illustrates some of the challenges.
According to this article from Chalkbeat, Detroit’s new system, which they’re calling Thrive For Teachers, will rate teachers on a 100-point scale. Here’s how that breaks down:
- 40 points from classroom observations
- 40 points from student growth on standardized exams
- 10 points from student surveys in grades 3-12 (not sure what replaces these points for K-2 teachers)
- 10 points based on teachers’ commitment to the school community
Additionally, the district will “provide stronger guidance on how administrators should observe teachers” and teachers “can expect to engage in regular conversations with leaders about their growth and receive individualized professional development.”
I appreciate that Detroit’s leaders recognize what teachers have been telling them for years: that the old system didn’t work and there needed to be a new one. And I recognize that designing a teacher evaluation system is a difficult thing to do. But this new system stinks, and it’s worth looking at why it stinks because Detroit’s proposed system stinks in many of the ways all teacher evaluation systems stink.
Regrettable Truth #1 – Classroom Observations Only Work in Theory
It makes sense on paper. A principal, a highly-trained educator with years of effective teaching experience on her resume, stops in frequently to watch her teachers. Using her vast knowledge of education research and the hard lessons she won through years of classroom teaching, she acts as a mentor to young or struggling teachers. She meets with them regularly. She sends them to conferences. She pairs them up with skilled veterans. She directs them to research that will help improve their instruction or classroom management. Her teachers see her as an ally in their quest to raise student achievement.
Unfortunately, this ideal rarely happens in real schools. Many principals weren’t particularly strong teachers; it’s one reason they got out of the classroom in the first place. Even those who were good simply don’t have the capacity to do what would need to be done to truly improve the teaching in their building. Principals, like everyone in education, don’t have the time to do everything they need to do at the level they need to do it. They make do. And they often make do by checking boxes, by minimally complying. They perform the required number of observations for the required amount of time and no more. They have the follow-up meeting with the teacher and never meet again. They often don’t have much to offer even struggling teachers because they either don’t know, are humble enough to realize that what works for one educator won’t work for a different one, or they just don’t have the time to devote to mentoring.
And that’s if the principal is acting in good faith. Putting 40% of a teacher’s evaluation in the hands of one individual is problematic for a host of reasons, the most obvious being that principals are humans and, like all humans, some of them are petty, vindictive assholes. As I’ve written before, there’s a reason Olympic gymnasts are scored by multiple judges with the high and low scores thrown out. People have biases. One person watching a performance does not see or appreciate all of the things another person watching the same exact performance sees or appreciates.
Regrettable Truth #2 – The Numbers Don’t Mean What You Think They Mean
In a nice, clean world we would be able to assess the effectiveness of a teacher by measuring what her students know and can do at the start of the year and then measuring those same things again at the end of the year. We would then be able to say how much of an impact the teacher had on each of her students and we could determine levels of poor, average, and exceptional growth and rate the teacher accordingly.
But learning isn’t a one-way street. It involves both the teacher and the learner, and the student’s learning is dependent upon more than just his teacher’s talents and efforts. What if a student missed 40 days of school? What if she refuses to wear her glasses or her parents neglect to have her hearing tested? What if she doesn’t care about the tests, since she has no skin in the game? What if the computer glitches on the big day and she balks at having to do the whole thing over again and then gives less than optimal effort?
When it comes to student performance on a test, there are too many inputs, but teacher evaluation systems pretend there’s only one.
And those are for the teachers who teach tested subjects. Detroit’s new system does nothing to remedy the problem with test-based accountability systems: that as many as half of the employees in the system are required to be rated on tests they don’t directly affect.
Finally, systems like Detroit’s play fast and loose with the language. When Detroit says they will use “standardized exams” for the growth portion of the evaluation, they don’t mean the state standardized test. Districts don’t get that data back quickly enough to use it for teacher evaluations for the current year. Instead, teachers are usually tasked with designing common assessments that all students at their grade level or in their subject area will take or districts misuse benchmark screeners for the purpose. In Michigan, the state test is only given at the end of the year, so there is no standardized test that can provide a beginning and end of the year measure that could be used to evaluate a teacher’s impact. Many of the tests districts use to evaluate teachers were never designed for that purpose, and some were created by people with no education in test design.
Regrettable Truth #3 – Students Are Easily Manipulated
The most novel change to Detroit’s system – and the one that’s generated the most headlines – is the decision to include student survey results in teacher evaluations. 10% of every grade 3-12 teacher’s evaluation would be based on what their students think about them. I’m a third-grade teacher and I can tell you two things. First, I survey my students at the end of each year, and without fail, the things my students like best about my class are not things that prove I’m an effective teacher. Last year, my students liked that I gave them breaks, that I played music during those breaks, and that they were allowed to read on their Chromebooks on Friday. Second, if I thought for one second that I was in any danger of being laid off due to a low evaluation, I would bribe my students on the day they were to fill out their survey. And let’s not pretend lots and lots of other teachers wouldn’t do the same. It wouldn’t even cost me much; you should see what students will do for a sticker.
Regrettable Truth #4 – Teachers Get Exploited. Teachers in Poor Districts Get Exploited More
Classroom observations are defensible because if they were performed as designed by people who should be leading our schools then they could conceivably lead to rich conversations that might improve teaching. You can also forgive Detroit for including student growth data in their teachers’ evaluations because, for the time being, they don’t have a choice; Michigan law will require districts to do exactly as Detroit will do in 2020-2021. You can even understand a district wanting to give students a say in their teacher’s evaluations (although I do wonder if Detroit uses teacher surveys in their evaluations of their administrators). After all, it’s the students who are the “clients.” We’re providing them with a service, so we should care about what they think of that service.
All of that is understandable, even if, in the real world, none of it will work the way it’s intended. What’s totally unacceptable is the final 10% of every Detroit teacher’s evaluation. That 10% is based on a “teacher’s commitment to the school community” and the Detroit Federation of Teachers should fight it with everything they’ve got.
Because all that part of the evaluation does is provide administrations with a cudgel to use whenever they need free labor or when they want teachers to keep their mouths shut.
Some, maybe even most, principals will interpret “commitment to the school community” as “going above and beyond” and “doing whatever is necessary.” They will tell teachers to “do what’s best for kids.” That’s the language of manipulation and exploitation.
Don’t feel like donating two hours after school to attend literacy night? Then you’re not “committed to the school community,” are you?
You aren’t willing to meet with parents whenever it is convenient for them? That doesn’t sound very committed to the school community.
You’re the first one to leave at the end of the day? Not enough commitment.
You expect to be paid for committee work? Well, the district doesn’t have the funds, and besides, you want those 10 points on your evaluation, don’t you?
But the most egregious use of the commitment standard will surely be when something is wrong and a teacher considers speaking up about it. Like, I don’t know, when teachers in Detroit blew the whistle on the deplorable conditions of some of their buildings, including leaking gym ceilings and black mold.
Any teacher who dares to publicly embarrass the district in such a way is surely not “committed to the school community.” You can bet they’ll receive a zero for that part of their evaluation, and with 40% of the rest of the evaluation in the hands of a principal who will observe their teaching, it wouldn’t surprise anyone if such a teacher were suddenly at the bottom of the food chain when layoff time comes around.
Detroit’s new evaluation system is more of the same, with the added insult of language that will make it easier to take advantage of teachers. The Detroit Federation of Teachers should oppose it, and if they can’t defeat it, then they should at least advocate for a name change.
Thrive for Teachers?
That’s some Orwellian bullshit right there.