To make the GPA useful, standardize it

As the semester ends, we oppressed teaching fellows get a chance to pull ourselves up out of proletarian squalor and participate in the most dysfunctional seasonal act in all higher education: grading. Under the current system, whatever mark your professors give is factored into an archaic index of student performance — the grade-point average — that is no longer meaningful. Invented as a way to sort students according to their academic ability and drive, it has largely become a measure of how well one can avoid math, science and rigorous humanities in favor of courses in other disciplines where high grades are earned by simply showing up.

The first part of the problem comes from the fact that different departments give radically different average grades. And so it comes as no surprise that an above-average math major can wind up with a significantly below-average GPA or that a middling Women’s and Gender Studies major can graduate with a significantly above-average GPA.

Being more or less rational, students choose their courses accordingly. Why work hard for a B or a B- in “Typology” when you could take “The Homoerotic Diacritics of Late Aztec Pottery” for a guaranteed A? As statistician Valen Johnson notes, students who want to get into selective graduate schools and Wall Street firms that put considerable value on the GPA are undoubtedly aware, and swayed, by these facts.

And how to find these easy classes? Equally easy: course evaluations. While not all highly ranked courses are easy, nearly all easy courses are highly ranked. Altogether unsurprisingly, professors looking to maintain course numbers and advance their careers have the motive and the ability to give undeservedly high marks. And they do.

Proposed solutions to this problem are based on the unremarkable idea that a comparative index like the GPA should be fair among disciplines. The pioneering Johnson, Brian Nagle and others have suggested indices of performance that rank students on the basis of their achievement relative to the performance of other students in the classes they take as opposed to those in courses with radically different average outcomes. In such a way, class rank could become a real measure of relative performance; it would show who is above average and who is below — something the current GPA decidedly does not.

For example, a student who gets a B (3.0) in a class where the average grade is a C (2.0) has performed at 1.5 times the average. In a class where the average grade is a B, however, getting an A (4.0) results in a relative performance of 1.3. And, finally, in a class where everyone gets an A, the relative performance of all students is the average 1. Class rank would be calculated by averaging the sum of relative performances over one’s academic career. This system most rewards those who take the hardest classes and do the best and most punishes those who take the easiest classes and do the worst, which seems much more fair than the current system that rewards slackers in useless classes.

Of course, this isn’t the only way to improve fairness in measures of academic achievement, but it would be a good start. One of the most positive early effects of adopting such a system would be to remove a major institutional disincentive students contemplating an undergraduate career in science now face. It would also promote experimentation in class choice for students, since getting good grades would be less dependent on the subject taken and more related to the amount of applied effort.

Regardless of the specific intervention, reforming the egregiously unfair GPA system is an essential task for colleges and universities — particularly if U.S. science and industry hope to keep their international edge.



Matthew Gillum is a first-year graduate student in molecular and cellular physiology. His column regularly appears on alternate Fridays.

Comments