The Lectern: School testing and performance measurement
We had a really interesting discussion Feb. 13 in my introductory public management class for master's of public policy students at Harvard University’s John F. Kennedy School of Government. As part of our discussion about using performance measurement to improve organizational performance in the public and nonprofit sectors, we talked about using standardized tests as a tool to improve the performance of public schools.
The discussion was interesting from several perspectives. First, I was surprised when I took a vote at the beginning of the class about whether people were "enthusiastic," "somewhat positive," "somewhat negative" or "very critical" about this idea (I stipulated that we would talk about school testing as a school improvement tool, not as an individual graduation requirement), to find that the clear balance of sentiment was to be relatively sympathetic -- 37 students put themselves in the first two categories, nine in the second two. Given that Kennedy School students are more often Democrats than Republicans, this was an interesting result, especially because of the skepticism about school testing among many Democratic political leaders, who may be reflecting the views of some (not all) teacher interest groups more than sentiment among (at least young) supporters. Just as interesting is that the class includes six students who have worked for Teach for America and taught in disadvantaged schools. The three students in the class who described themselves as "enthusiastic" about school testing were three of the Teach for America teachers (as was one student, also a veteran of the program, who had written an undergraduate thesis on school testing).
The students highlighted several things during the discussion. There was a lot of sentiment that the numbers didn't help with anything unless schools used them to diagnose problems and promising approaches, and to pinpoint areas for management priority. I agree completely -- if we just display numbers, without using them as a management tool to improve performance, they are worthless. Particularly the Teach for veterans were strong on the point that test scores created more urgency for improving the education that disadvantaged students receive. And there was, from some students, skepticism about the punitive elements of the No Child Left Behind Act, which many students worried encouraged gaming and resentment. I am also inclined to agree with this view.
This was a good discussion, and there is broader good news here for good government. Most of the next generation of people in public service -- or at least the next generation coming out of the Kennedy School -- looks ready to embrace the idea of using measurement to improve the performance of public and nonprofit organizations. I have expressed to them my hope that managing using performance measures will become as natural to their generation of people in public service as it is to their counterparts in the business world.
Posted by Steve Kelman on Feb 14, 2008 at 12:09 PM