Identifying the Incidence of "Grading on a Curve": A Within-Student Across-Subject Approach

Theoretical work shows that grading on a curve, i.e., teachers assessing students relativeto their classmates, can negatively affect students' learning effort. However, little isknown about its empirical incidence. To overcome bias from non-random sorting andomitted variables like teachers' grading standards, we exploit within-student acrosssubjectvariation observing both teacher-assigned grades and test scores of German4th-graders in reading and math. We find that having classmates with one standarddeviation higher test scores lowers a student's grade by about 10 percent of a standarddeviation. Importantly, only female teachers grade on a curve and there is no associationbetween students' learning effort and relative grading.