Grading Teachers by the Test

Eduardo Porter of the New York Times has an ongoing economic column that at times addresses education. An issue that he took up not long ago concerns the increasingly common phenomenon of value added testing. In particular, Porter wants to highlight the potential role of “Goodhart’s Law” related to education testing and teacher evaluation.

Goodhart’s Law refers to the idea that measures designed to evaluate the effectiveness of some goal of reform often end up creating just as many problems as they solve. In other words, once everyone knows what the goal is and what the incentives associated with that goal are, they will start cutting corners to achieve the goal rather than making the sort of changes that the goal was designed to promote.

Goodhart’s Law can result in such corner-cutting as out and out cheating, schools helping only those students who can reach certain thresholds with minimal improvement, or altering teaching to focus on drilling for the tests.

Porter warns that if more states choose to increase value added test results to 50% of a teacher’s evaluation, we can expect to see more of Goodhart’s Law take effect in education.

At the same time, though, Porter affirms the importance of teacher evaluation. Tests can certainly be a part of teacher evaluation, especially if they are improved tests like those created by PARCC and Smarter Balanced and if they are one part of a multifaceted evaluation system.

We just need to be aware of the potential effects based on Goodhart’s Law and work toward implementation that avoids detrimental corner-cutting. Some ways to do this include:

  • Use multiple measures, where no single measure makes up the majority of an evaluation.
  • Allow for professional discretion of supervisors within the evaluation.
  • Think deeply about current reality, root causes, and future realities that may emerge from treating root causes. Mitigate negative effects before they occur.

For more information, please visit: