New study on teacher merit pay questions over outcomes

A recent and widely publicized new study by Vanderbilt University suggests that teacher merit pay has a negligible impact on student achievement. The study’s findings are well summarized in a recent Washington Post article (pasted below), and an analysis in Education Week. The study is a three-year, randomized experiment involving 300 teachers in Nashville, Tennessee, and considered to be methodologically solid.

The articles point out, however, that the study does not consider two important aspects of merit pay. The first is that merit pay systems may not appeal to many current teachers, but may attract more talented people to the teaching profession, who respond to merit pay and can do a better job of improving learning. The second is that merit pay may work best when combined with other elements, like professional development strategies and teacher performance evaluation systems, none of which were part of the Nashville study.

In his own critical reaction to the study, AEI scholar Frederick Hess makes similar arguments, but goes further. He suggests that the Nashville study may confuse the merit pay policy implications given its rigorous design but disconnect from “what we actually care about” when it comes to evaluating merit pay.  

The growing experiments with and evaluations of merit pay in the US reflect a widespread unease about basing teacher pay on training and seniority, and the search for a better approach.