
Several months ago, the faculty at Harvard University proposed a new policy to address one of the institution’s most embarrassing problems. The concern? Its professors were awarding too many A grades to its students.
At the time, the institution calculated that A grades represented nearly two thirds of all course grades. So how might it address the grade inflation concern? One approach might define clear expectations of reasonable grade distributions for different types of courses.
For instance, in an introductory psychology course with 100 students, a skewed grade distribution might be reasonable. In other words, the university might expect a large amount of grades at one end of the spectrum and a tailing number of poorer grades towards the other end. Such an outcome would illustrate that, for a course with relatively straightforward introductory material, it is indeed reasonable for a majority of students to receive a very high grade.
On the other hand, in an theoretical quantum physics course with 10 students meeting in a small seminar room, a bell shaped distribution might be reasonable. In other words, the university might expect the majority of grades to be clustered in the middle of the spectrum, with tails of high and low grades falling on each side. Such an outcome would illustrate that, for a course with extremely difficult material, it is indeed reasonable for a majority of students to receive a moderate (i.e. not very high nor very low) grade.
Did Harvard adopt such a policy? No. Instead, it initially decided to cap the number of A grades in each class at 20% plus 4 of the student roster. Under this system, up to 24% of all students in our hypothetical introductory psychology course could receive A’s, while up to 60% in our hypothetical theoretical physics course could do so. This outcome would reverse the logic of our skewed vs. bell example, where introductory students are assumed to be more likely to “ace” the material than advanced students.
At a recent meeting where Harvard leaders were expected to formally ratify this new grade cap for the upcoming academic year, the institution instead decided to delay implementation for a year. However, there was no attempt to consider any revision that might address the difficulty of the course material and the ability of students to master the subject matter.