Monday, February 22, 2010

Merit pay. (Again.)

In my eyes, we should pay better teachers more because it’s right. But that is not why merit pay is being written into policy. Ostensibly, the idea is that tying pay to student scores will make teachers and teaching improve. What we need to talk about is: how, exactly?

If you believe that teachers do not perform as well as they could because they lack the proper incentives, then this policy shift makes sense. But my years spent observing classrooms tells me that that ineffectiveness has as much to do with ability as with motivation. An awful lot of teachers simply do not know how to teach more effectively than they do now. It is not like they have these reserves of greatness they are withholding from children simply because they don’t have good enough reason to share it.

I would love to see far more reporting, as we enter the new world of merit pay, that plumbs teachers’ points of view. Not just a passing quote, but get into people’s classrooms and brains a bit, challenge them, get a sense of what kind of impact performance pay might have, or might not.

And there is no better time than now to address potential impracticalities, so they can be addressed during implementation. For example: In the Houston Chronicle this weekend, Joel Klein called value-added “a leveling of the playing field that allows us to isolate teacher impact.” In the high-poverty schools you cover, how much of a child’s reading instruction comes from one teacher? Where the student is taught by only one teacher all day, it is easier to isolate his or her impact. But what about the children I spent a year with at Tyler Heights Elementary in Annapolis? Every day, the students were shuffled into Title I-funded interventions that meant that two-thirds of the children in Ms. Johnson’s third grade classroom spent at least half of their reading time away from her. And that doesn’t count the afterschool tutoring most of them were in every day, which she didn’t teach. But for paperwork’s sake, she was their teacher, and in a value-added system their improvement or lack thereof would be attributed to her.

Or would it?

Journalists in districts that have been using performance pay for some time owe it to everyone to go far more in-depth on the topic, from the classroom, so as new plans are put in place all relevant considerations are on the table. Is there any great journalism on the topic I might be missing?


  1. I too would like to hear of some great journalism on this topic.

    I'd also like to point people to research by Jesse Rothstein on the topic. He looked at the fourth grade tests scores of fifth grade teachers' students in North Carolina to see if the teachers had any effect on their students' performance. They did. They shouldn't have, obviously. So that has huge implications as more and more places implement merit pay, especially based on value-added testing.

    Unfortunately, Jesse is not available for the national seminar since he is working for the Council of Economic Advisers this year.

  2. Amen. I've visited a number of schools--most recently Viers Mill Elementary in Silver Spring, MD--where any number of teachers and resource staff will share responsibility for a single student.

    While it's true that the quality of instruction has a lot to do with a teacher's ability, so do the conditions in which they teach. Teaching conditions don't often come up in discussions of pay for performance.