Evaluations Are Better Than No Evaluations
Eric "Rick" Hanushek continues a blog conversation with Deborah Meier this week.
In our conversations about accountability, we have skirted around the issue that I think drives the most heated debate—namely, that accountability involves evaluation of teachers and administrators. And teachers and administrators are "agin it," period.
Can't we pare through some of the smoke and move the discussion forward to a better place?
It is curious, for example, that there is a sudden uproar about high-stakes testing with school accountability, but that there were essentially no complaints when only students were subject to high-stakes testing. The standardized SAT test—to say nothing of the entrance exams for medical, law, and business schools—have been fine.
Moreover, there is an inordinate amount of distortion in the discussion. While citing a litany of potential flaws in testing, the opponents of teacher and school accountability often create a straw man that bears little relationship to the real policy discussion.
- Nobody suggests that the level of student achievement should be used to judge teachers. Instead, when any discussion of a test-based component comes up, it revolves around value-added scores that are designed to parse the input of the teacher and school from other influences on test levels. Simply arguing that schools serving concentrations of disadvantaged students invalidate any use of tests misses the point completely.
- Nobody suggests that just value-added scores should be used in evaluations. All policy discussion recognizes that a minority of teachers are found in tested grades and subjects and have enough students to produce valid value-added scores. For example, only a fifth of teachers in Washington, D.C.—where value-added scores are one component of evaluations—have valid value-added scores. Thus, the real discussions center on improving the observational components of evaluations, i.e., eliminating the "widget effect."
- Nobody believes that all teachers are equally effective and that the extreme differences in learning growth across classrooms and schools are all due to flawed tests. Simply stating potential flaws in testing or value-added estimation does not obviate the large, systematic, and true differences among teachers in their classroom effectiveness.
At this point, I recognize that a large number of the first-responders like to repeat the prior points—just more emphatically. But my purpose is to see if we cannot move to a better place.
Specifically, I think that the "no way, not now, not ever" reaction to evaluation is not in the interest of teachers and administrators. I think that the vast majority of teachers and administrators would be better off by abandoning the scorched-earth approach to evaluations and by joining into the constructive development of better evaluation approaches.
They would be better off for a number of reasons. To begin with, since virtually nobody in this country believes that all teachers are equally effective, it is hard to argue that policies should not take effectiveness into account. Simply saying that it is all about due process rights and not about protecting weak teachers does not fly with the public. As a result, a growing number of state legislatures are saying that they will dictate the evaluation system if the participants cannot come to a better position. This approach is clearly not in the interest of teachers and administrators (nor do I think in the interest of getting the best results), but I believe that it will sweep into more and more states in the absence of serious discussions of evaluations.
Finally, while I do not want to get embroiled in a salary discussion here, I firmly believe that the failure to recognize differences among teachers has worked to hold teacher salaries down. Teacher salaries are politically determined—not market determined. As such, it is exceedingly difficult for the political system to grant substantial salary increases to all teachers when everybody recognizes that some of the teachers should not be in the schools at all.
Deborah, here is where your wisdom would help. You obviously have paid a lot of attention to the evaluation of teachers over your career. Is there a way to move to a better system—one that would benefit not only teachers and administrators, but also students?