A Teacher's Advice on Getting Teacher Evaluation Right
In the last three years, I've been evaluated three different ways. I'm not sure how I'll be evaluated this year. No one has told me, and the information on the district website is a year out of date. That's problematic, but probably not unique to Baltimore. A wave of education policy changes have combined to create an increasingly confusing situation.
Here's the rundown of what's happened recently in Baltimore. Our old evaluation system, PBES, was based on two observationsalmost no one got the lowest ranking. This was thrown out, and no one, myself included, is advocating bringing it back. The new multiple measure evaluation is supposed to include
- value-added measures for teachers who teach in the "tested subjects";
- student-learning objectives for teachers that don't teach tested subjects;
- two classroom evaluations where teachers are evaluated by a nine-indicator teaching framework and rubric;
- student perception surveys;
- a professional expectations checklist; and
- a school index, which is a number determined by plugging school data into a formula (graduation rates, attendance, test scores, etc.).
However, these components are at various stages of being tested and normed. Not all of them are ready to be rolled out formally, especially when PARCC testing will most likely show lower student performance than the previous measures. It's clear that my district hasn't fully decided what parts of the evaluation they'll use this year.
I'd like to offer the district, and others in our situation, some suggestions. I know a little something about evaluation. After all, as a teacher, a big part of my job is evaluating students. I use formative assessments to evaluate where students are struggling, so that I can create supports to help them reach higher outcomes. I also use summative assessments to evaluate my student's learning and give them final grades. It seems like since we need an evaluation system that can both support and "grade" teachers, a teacher would be the best person to ask about how to do both fairly and accurately. I've got three suggestions.
First things first, don't make an evaluation high stakes until you're sure that it's well designed. If I design what I think to be an awesome project, then I give it in class and the kids absolutely bomb it, chances are that the project wasn't as great as I thought. It wouldn't be fair to those kids if that project was a huge part of their grade, because unbeknownst to all of us, the project was flawed. Likewise, with teachers, don't rush to make evaluation high stakes until all the bugs are ironed out. We don't need to trade one bad evaluation system for another.
Next, support early and support often. A final evaluation of a teacher should only come after there have been meaningful opportunities for the teacher to improve. Any major project I give contains multiple benchmarks, so that I can review student progress toward the goal, and provide additional supports where necessary. Similarly, a teacher-evaluation system needs to let teachers know early that they're not on track to be rated effective. (That's a significant weakness for VAM, which at this point would be released to teachers towards the end of the school year. Data is best when we can use it to inform instruction. Surely with all the progress we have made, we should be able to design measures that provide data quickly to teachers, and then support us to tailor our instruction to what the data does and doesn't show.)
Lastly, when you're designing evaluation, don't be afraid to innovate and radically rethink the way things are done. I've evolved from a teacher focused on multiple-choice tests and objective data to a teacher whose class revolves around historical investigations, Socratic seminars, and Oral Histories. I've stepped way out of my comfort zone time and time again because it's worth it to experience authentic learning with my students. When you're designing a new teacher-evaluation system, don't do something because it's common or trendy, do something because it delivers authentic results. It's worth it.