Building a Better Teacher: Still More Questions Than Answers

Thursday, Apr. 21, 2011

By Andrew J. Rotherham

Teacher effectiveness matters more to student learning than anything else schools do, and there are substantial differences between teachers. Those two points often get lost in the din about teachers unions or tenure. Underneath all that noise, however, researchers are quietly looking at teacher quality. Two new studies that didn't get a lot of attention challenge beliefs of reformers, teachers unions, and reform critics. (Beyond Unions: See Five New Rules for Teachers.)
In a paper published by the National Bureau of Economic Research, economists Eric Taylor and John Tyler used data from Cincinnati, Ohio to look at what happens when teachers are actually evaluated. Taylor and Tyler tracked teachers during the year they were evaluated and the following years. They found that not only did performance (as measured by math achievement of students) increase during the evaluation year, but the gains were sustained in subsequent years. That's a big deal—it means teachers were not just responding to being evaluated but using the feedback to improve their work.
Two aspects of the Taylor and Tyler study are especially relevant to when and why evaluation works. First, the teachers in the study were all veterans with at least five years under their belts. That means that they were experienced teachers and still became better through evaluation. In addition, the teachers were not evaluated using student test scores, but rather through observation based on a set of metrics developed by the Cincinnati schools. That matters because most teachers around the country do not teach in subjects — for instance math and reading — that are regularly assessed with standardized tests. (See what the NFL can teach teachers.)
The second study points to shortcomings of using National Board certification and pay incentives as a way to get teachers to work in more challenging schools. National Board certification is credentialing program for teachers that has been around since 1987. The process of earning certification is an arduous one involving tests, video, and a portfolio of work, but research does show that teachers who complete it are modestly more effective than those who don't. The credential is used around the country and most states recognize and reward it in some way. Jim Simpkins of the Center for Reinventing Public Education at the University of Washington, looked at the results of Washington state's efforts to pay teachers $5,000 more annually for earning the National Board credential and an additional $5,000 for teaching in a high-poverty school.
Simpkins found that the number of National Board teachers in challenging schools is increasing across Washington, but mostly because teachers in those schools are earning the credential — not because teachers with the credential are changing schools. He also found that National Board teachers are no more likely to stay in those schools than other teachers. These results are disappointing for proponents (and I'm one) of tying National Board bonuses to service in high-poverty schools as a way to improve equity for low-income students. (Rating Teachers: The Trouble With Value-Added Data.)
So what's the takeaway? Most immediately that there are probably better ways to spend $10,000 to attract and keep one teacher in a high-poverty school (and Washington's governor has proposed ending the bonuses). More generally, both these studies illustrate how we had better get used to a lot of trial and error as the education field at last tries to grapple with this fundamental issue of human capital.
Last week, teachers unions and school reform groups in Illinois agreed on some policy changes there — including common sense reforms to teacher seniority rules. And the current emphasis on teacher evaluation because of Race to the Top will produce some new ideas and approaches, too. These are obvious and foundational steps that policymakers should take, but the reality is that because of years of inattention to teacher effectiveness, we still know relatively little about what makes a teacher great and how to build systems full of great teachers and high quality instruction. That frustrates policymakers — and it should terrify parents. But it's also an enormous opportunity for a field that is ostensibly about learning to perhaps learn something itself