What We Talk About When We Talk About "Good Teachers"

It's always interesting to watch the Bill & Melinda Gates Foundation tackle a big problem. With so much more resources than other foundations, Gates can shell out millions just doing their homework — before then spending the truly big money.

Take the Bill & Melinda Gates Foundation's work on teacher effectiveness. A few years back, the foundation came to suspect that good teachers mattered more than anything else for student success — a humbling realization for a foundation that had famously spent a fortune pushing small schools, among other ideas that didn't have much impact. (See Gates Foundation: Grants for K-12 Education)

But saying that good teachers are all-important and knowing what makes a good teacher good are two very different things. And the Bill & Melinda Gates Foundation rightly saw that nobody actually knew—empirically speaking—what made a teacher good. Existing evaluation systems failed to truly measure teacher quality, a daunting task given the many different aspects of teaching and the difficulty of isolating the effect of teaching from other variables.

Even a foundation with the deepest pockets can't promote good teaching if it doesn't know what that means.

So, in 2009, Gates launched a huge project, Measures of Effective Teaching, to learn more. The scope of this project is pretty amazing, involving 3,000 teachers who have volunteered to participate from half a dozen large school districts around the country. And it's been a financial boon to researchers. As the home page for the MET project notes,

Lead researchers involved in the project are affiliated with Dartmouth College, Harvard University, the University of Michigan, the University of Virginia and Stanford University. Participating non-profits and education companies include RAND Corporation, Educational Testing Service, Teachscape, The Danielson Group, The New Teacher Center, National Math & Science Initiative, and Westat.

It's hard to say exactly how much money Gates has spent on the MET project—possibly tens of millions of dollars. All to answer questions about teaching that may, at some level, be unanswerable given the number of variables at play. The long-time education expert Diane Ravitch, now among the Gates Foundation's leading critics, has called this effort a "quixotic search to guarantee that every single classroom has a teacher that knows how to raise test scores." (Read director of College Readiness, Vicki Phillips' IP profile).

But, of course, that's not really fair, as Ravitch surely knows and anybody can see by looking at the measures of teacher effectiveness that the Bill & Melinda Gates Foundation is analyzing — for instance, the Tripod Student Perception Survey, which looks at whether students think their teachers are doing a good job of helping them learn.

What really make critics uneasy about the Bill & Melinda Gates Foundation's work on teaching is that it's part of a larger drive in education to quantify everything and apply business management principles to public schools, historically one of the most democratic spheres of U.S. society. Among other things, critics worry that teachers will face far greater physical surveillance and be subjected to stricter and narrower forms of accountability that will diminish their freedom and creativity as professionals. Even the language Gates uses to describe teachers makes critics queasy — phrases like a "teacher’s value-added score."

It doesn't help that the group of lead research partners for the MET project doesn't include anybody from teachers' unions, but does include representatives from ETS, the RAND Corporation, and several for-profit consulting groups. 

In any case, what has the Bill & Melinda Gates Foundation learned from its huge investment in studying teaching?

What the foundation has shown — or so it says — is that it is possibly to rigorously evaluate teacher effectiveness. A paper published in January shares insights and guidance on how to observe and evaluate teachers. The paper claims that the MET project has adopted the right methodology for answering its core question of what makes teachers good, and that researchers have succeeded in isolating the "added value" of teachers (there's that phrase again).

Still, critics should be reassured by a key message of that preliminary paper: Measuring teachers is not simple cookie-cooker work. It's nuanced and you need to be careful, using well-trained observers who look repeatedly at teachers (not just once) and use data from standardized tests need in conjunction with other forms of information, like what students are saying about teachers.

The MET project don't believe they've found a silver bullet, nor do they think they have a devised a scientific system for measuring teachers. Rather, the whole point of this exercise from the start has been that teaching is really important and we can and should have better tools for understanding what good teaching looks like. In that sense, a penultimate finding of the MET project so far is quite modest:

Combining new approaches to measuring effective teaching—while not perfect—significantly outperforms traditional measures. Providing better evidence should lead to better decisions.

What's so bad about that?