The Problem With Demanding Proof on Teacher Evaluation
Looks like it's gonna be a full week without school for Chicago kids.
I've steered clear of saying much on the CTU strike because with so many people commenting and writing about this it gets hard to have anything original to say, but with so many people weighing in it's inevitable some of them will be saying ridiculous things that deserve calling out. One of the weirder memes I've seen going around the last few days is the notion that "the real problem here is that there's no evidence the teacher evaluations Rahm Emanuel wants to put in place work."
Let's think about this for a second. It's by and large true that there's no evidence that the evaluation system proposed in Chicago will improve student achievement. But that's because such evaluation systems have been implemented to date in only a few places, most of them for too little time to see any evidence of their effects. Until very recently, virtually all school districts in the country had very similar teacher evaluation systems that didn't take student learning (by any measure you like) into account or give teachers much real feedback about performance--and virtually all teachers were rated satisfactory or excellent. Those systems are still in place in most districts and, just for the record, there's not really any evidence that they improve student achievement either. To say that the Chicago Public Schools shouldn't try a new approach to teacher evaluation because there's no evidence it works is basically to say that we should never try anything new in public education--because things that have yet not yet been done by definition can't prove that they're effective.
Having spent much of the past year looking at state teacher evaluation legislation, I'll be the first to admit that there's a great deal we don't know about what works and what doesn't here, and that we really don't know how a new teacher evaluation system is going to work. Anyone who claims to know for a fact that the new teacher evaluation proposed in Chicago will or won't improve student learning has a better Magic 8 ball than I do. But we do know that the current system is broken: From 2003-2008 only 1% of Chicago teachers were rated "unsatisfactory" a fact that jives with neither research finding wide variations in teachers' impact on student learning, nor test data showing that less than 1/3 of Chicago high schoolers passed reading and math assessments in 2008. And the only way that the only way to learn how to do better is by trying new things, carefully monitoring what does and doesn't work about them, and making improvements and adjustments over time. It's worth noting that the system on the table in Chicago is designed in part to do exactly that: The percentage of a teacher's evaluation that is based on student learning indicators would increase gradually over time and after 5 years a committee will make recommendations for the longer-term design of the system, based on lessons learned.
The question on the table here isn't "do we know the new system will work," but "is there any reason to believe that some changes might improve upon the system that we currently have?" And the answer is yes.