Dan Goldhaber serves as the principal investigator of CALDER’s Washington State team and as a member of CALDER’s Strategic Planning Group. He is a research professor at the University of Washington’s Center on Reinventing Public Education, an affiliated scholar of the Urban Institute’s Education Policy Center, and senior nonresident fellow of Education Sector.

Read more about Dan Goldhaber»

Plain Talk Archives

In recent years, education reformers have zeroed in on teachers and teacher quality. Teachers are getting more federal attention too—even the economic stimulus plan includes money to improve the quality and distribution of teachers. Why are we seeing this shift?

I think it’s the direction that research has taken us. Research has shown that teachers are the most important schooling factor in improving student achievement. Compare that to the 1990s, for example, when the focus and funds went to lowering class size. The advent of longitudinal databases in the mid- and late 1990s in a few states allowed us to compare the benefits of class size reduction with the benefits of more effective teachers—and teacher quality beat out class size. So if you’re trying to leverage what schools can do for students, the message is, upgrade the quality of teachers.

The problem is that teacher quality varies greatly and the attributes of effective teachers are hard to identify. Years of experience and advanced degrees determine pay increases, but aren’t strongly predictive of teacher effectiveness. We’re finding that more teacher experience beyond the first three to five years of teaching does not appear to improve student learning. And there’s not much evidence that advanced degrees in general have much impact on student achievement. Teacher licensure scores and other measures of academic proficiency are a little better at predicting whether a teacher will be effective, but they sure don’t provide guarantees.

So how do we get quality teachers into classrooms? Is performance pay a promising option?

The policy options range from alternative routes into the classroom to changes designed to make teaching a more desirable profession. Compensation is a key workforce policy lever. The traditional district salary schedule has been largely sacrosanct and very difficult to change, but it doesn’t seem to be working particularly well. We don’t know exactly what pay reform should look like, but we ought to be experimenting to figure out what works. There’s a lot of promise in what schools are trying, particularly through the Teacher Incentive Fund (a federal initiative that supports performance pay programs in high-needs schools).

We’re seeing some experimentation with pay for performance. The early evidence, still thin so far, suggests performance pay may help improve student achievement, possibly by attracting different kinds of people into teaching. You might think it would prod people to work harder, but most such incentives are so small that I’m a bit skeptical of that. More promising is the long-run effect. The notion here is that people who think they’ll be successful will be attracted to a profession that rewards success. That’s generally the way that the broader labor market works, so many highly productive people gravitate toward better financial opportunities outside of public schools. Pay for performance could change that.

All that said, I don’t think performance pay is enough on its own. Systemic change requires making many simultaneous and aligned moves. If all you do is plunk down a pay for performance model and it’s not implemented well and you don’t have data systems in place to figure out who the strong performers are and you don’t have mechanisms for teachers to improve, then why would that model work or even survive? We need to think about systems of reform, not just one kind of reform, if we want to dramatically change teacher quality. Policymakers are always looking for silver bullets, but there don’t appear to be any.

Critics of performance pay have qualms about overrelying on student test scores to measure teacher effectiveness. What’s the best way to measure performance?

I don’t think we know the best way to measure performance yet. Subjective evaluations typically don’t fly in education, but there’s no consensus on the right statistical approach. The profession hasn’t done a very good job of evaluating teachers and using those evaluations to make substantive decisions about hiring, tenure, and dismissals.

Policymakers are looking to value-added measures—measures that isolate a teacher’s contribution to student improvement—as an objective, more politically safe way of judging teacher quality. But they often don’t know the complexities associated with value-added analyses and probably oversell the ability of these measures to gauge an individual teacher’s effectiveness.

The potential beauty of CALDER and state longitudinal databases is the opportunity to learn about what’s working and what’s not. Too often, schools and districts have made investments and changed policies with no capacity to figure out whether the changes were working. More often than not, politics drove policy. Thus, new superintendents and newly elected school boards would change policies just to put their stamp on the system—without evaluating prior policies’ effects. The potential of CALDER is to turn school systems into learning organizations so decisions aren’t based on reform fads or political whims, but on empirical evidence about what impacts student achievement.

How might the economic crisis affect teacher policies and teacher quality?

Public schools will likely to have to make some painful cuts. The question is how they’ll structure layoffs. Will they base them on seniority? Less experienced teachers make less money, so you’ll have to cut more of them to keep higher-paid teachers with more experience. But, as I mentioned, more experience beyond the first three to five years doesn’t necessarily make teachers better. The crux here is the trade-off between the size of the cuts and the quality of teachers laid off. This decision could have long-run effects on the quality of the teacher workforce.

From a student achievement perspective, the least effective teachers should be let go if layoffs are necessary. If something other than teacher performance determines who is laid off, schools may jettison some very good teachers. Then, even if economic conditions improve, allowing for re-hires, some of those let go may be lost to the profession forever.

                                                                                                                                         Read other commentaries in the Plain Talk Archives