Student learning gains ought to be a component of teacher evaluations. Measures such as value added are a useful and important complement to classroom observations. But not all models are created equal, as illustrated by a new lawsuit in Tennessee that reveals a rather preposterous policy.
Last week, the Volunteer State’s largest teacher union sued the state in federal court over a law that ties student test scores to evaluations of educators who teach such non-core subjects as art, French, and gym. Teachers in Tennessee receive annual scores between one and five, with five being best. Those scores determine all manner of high-stakes administrative decisions affecting teachers, including bonuses, termination, and tenure. Approximately half of the metric is based on classroom observations, the rest on student test scores. For a teacher in a core subject such as math, and in a grade in which students are tested, this model makes sense. The bulk of the test-based portion of her rating is based on how well her students do on the math portions of the state’s standardized tests. That’s rational. A smaller portion, 15 percent, is based on “school-wide” performance—how well all the schools’ students do in all subjects tested. That also makes at least some sense as a strategy for encouraging teacher collaboration.
Yet for non-core instructors—the focus of the lawsuit—the law becomes rather absurd. Aside from a few questionable alternative assessments that aren’t widely used in Tennessee, no standardized assessment data exist for the subjects and pupils that these people teach. Nevertheless, the law mandates that at least 40 percent of their evaluations be based on test scores. So school systems use the only metric they have—school-wide performance. For over 50 percent of the state’s teaching force, in other words, nearly half of their annual performance review can be based on pupil scores in subjects they don’t teach, often earned by students they haven’t even met.
The bases for the lawsuit—the harms the plaintiffs complain of—are equal protection and due process violations under the U.S. Constitution. I wouldn’t hold my breath on the first of these, but the latter cause of action might have teeth. Due process balances three factors: (1) the private interest at stake; (2) the risk of erroneous deprivation of that interest; and (3) the government’s interests, such as avoiding lengthy and costly administrative proceedings.
The first prong here would be things like employment and government benefits. The second prong asks whether the procedures are fair and reliable. The final one would probably be the state’s interest in providing its children with quality teachers and a quality education. The balance appears to be weighted heavily in the non-core teachers’ favor. Their private interests are strong. The evaluation model, by attaching major employment decisions to things these teachers have virtually no control over, is unfair and unreliable. And there’s very little about this policy that improves education.
It’s always hard to know what courts are going to do in cases like this. Regardless of the outcome of the lawsuit, however, there’s value in pondering the issues and challenges that it brings to light. Value-added measures—when done right and sensibly—are worthwhile tools. Tennessee’s model for its core teachers makes sense, and nothing about this case should change that. But in standing by an ill-founded policy, the state is making itself look bad; worse, it’s drawing negative attention to test-based teacher evaluation models at a time when testing is under broad national attack.
Going forward, the state’s policy leaders ought to do one of two things. The first option would be to significantly reduce the evaluative weight given to school-wide performance for non-core teachers, instead basing the vast majority of their reviews on observation. This is probably the best path. Alternatively, if the state can develop rigorous, proven alternative assessments that produce useful data for students in a few courses like world languages—no small undertaking—then it could use those for the test-based portion of evaluations. There is no third option. Wise up, Tennessee.
For the plaintiff union’s part, it must be challenged to confine its grievance—and the implications of a court decision in its favor—to the non-core teachers with a legitimate beef, not use it (as they have in the past) to tear at the fabric of learning-based teacher evaluations.
***
Correction: February 12, 2015
A previous version of this article said that for over 50 percent of Tennessee's teachers, nearly half of their annual reviews were based on pupil scores in subjects they don’t teach, often earned by students they haven’t even met. While this can be the case—and is true for many—it isn't necessarily true for the entire cohort.
