Join the conversation

We all know that good teachers are important. But I think it is commonly acknowledged that we have trouble assessing whether a teacher is good.

Some claim that teacher quality can be determined based on whether a teacher raises the standardized test scores of her or his students. Others disagree, believing that good teaching is about more than test scores and/or that even if we agreed test scores were the basis on which we wanted to judge our teachers, current systems for doing so are unreliable. And still others argue that yes, good teaching is about more than test scores and that teacher assessment systems based on those scores can be unreliable, but they are the best way we currently have to assess teacher quality...so we need to use them.

A recent report from the National Education Policy Center (NEPC) may help us work through some of the very complicated issues surrounding teacher assessment. It reviews the evidence to date on a number of concerns on the topic.

I will spare you the suspense: The author concludes that the best teacher assessment systems use multiple sources to determine whether a teacher is performing. In reviewing evidence on assessment tools - including classroom observation, instructional artifacts (lesson plans, assignments, handouts, etc.), portfolios, teacher self-reports, student surveys and value-added assessments - she ultimately finds that all have strengths and all have weaknesses, and we need to balance them.

I completely agree with her - and I think the report is well worth a read just to familiarize yourself with some of the issues around not only the tools of assessment, but also who conducts those assessments and what the purpose of them are.

In fact, what struck me almost as soon as I started in with this report is how shockingly superficial most (of the loudest) debates on teacher quality are. For example, the report points out that prior to designing a teacher assessment system, one should ask what it is the system should 1) assess and 2) achieve. And I cannot remember the last time that I heard that conversation on a national scale…or at least, not an honest conversation.

It seems that the politicians and journalists who cover this issue take as a given that we are only interested in assessing outcomes. If that were true, then the discussion that we are having on a national scale - about whether standardized test scores are the outcome of interest - could be the next step.

But as this report points out, imagine that we design a system that allows us to capture whether a teacher has achieved what we believe a good teacher should. If all we measured were the outcomes, how would we know what teacher characteristics and instructional practices were associated with them? How will we help other teachers achieve them?

The report goes into detail about the difference between summative assessments (those used to make judgments) and formative assessments (those used to guide improvements), the importance of each and the challenges of designing a system that can do both simultaneously (it can be done – it just has to be done carefully).

But the report makes clear that if we want a good teacher in every classroom, we need to help educators improve. We need to assess their practice so they can strengthen it. And that means that discussions of teacher assessment systems must go beyond student outcomes.

UPDATE: Another study on this issue was released today. Check out what EdWeek has to say about the Bill & Melinda Gates Foundation's Measure of Effective Teaching study.