In a required math sequence at the United States Air Force Academy, students take the same exams but are randomly assigned to different sections taught by different instructors. Scott Carrell and James West show that students assigned to an instructor with better teaching evaluations perform better on the course's exams. But they also show that being assigned to a popular instructor decreases students' scores in subsequent math classes. What might explain this puzzling pattern? How does it relate to the problem of failing to measure your mission?

Display keyboard shortcuts for Rich Content Editor