Beyond the Princeton 'A' Cap
Most disagreements surrounding the issue of grade inflation can be reduced to a divergent understanding of the purpose of grading, namely, whether grades exist to communicate the raw or relative levels of student achievement in a course. At some institutions, grading can do both, for there will be enough students who do both well and poorly that the differences in students’ raw accomplishment will be significant enough to communicate relative performance as well. Princeton’s decision to cap A-range grades at 35 percent in order to restore grades’ role as a communicator of relative performance jars with students expectations precisely because students persist in viewing grades as a measure of raw accomplishment. Harvard’s decision to limit the percentage of students awarded Latin honors received a similar response because most Harvard students feel that they are capable of performing at the level of an “honors student.”
At a school such as Harvard, however, where a vast majority of students are capable, in terms of raw measures, of A-level work, grades based on merit alone fail to distinguish most students. But ironically, these ubiquitous A grades hardly communicate even raw performance, much like a perfect score on an intelligence test cannot accurately measure intelligence for failure to challenge the test taker.
This problem would be solved if Princeton—and Harvard—would instead focus on combating the inflation of grades as a signal of raw accomplishment and committed itself to a much bolder change. The notion that a C denotes “average” performance in a course seems positively quaint today, but why not revive it? This is not to say that courses should be graded on a strict curve with a C as the mean; that would reinstate grades exclusively as a relative communicator of performance and have similar effects as the Princeton grading cap. Instead, why not rescale the expectations of achievement in a course so that a C would indicate a fairly strong level of accomplishment? Professors could challenge their classes with interesting and difficult exam problems without worrying that most students would be unable to answer them. Papers could be graded much more critically; even students would admit that the papers that we think are good are probably not the best pieces of academic scholarship. Once in a blue moon a student would meet these challenges and earn an A, and this would not demean the accomplishment of the B’s, C’s, and D’s earned by his or her peers.
The desirable consequences of drastic grade rescaling would extend to other aspects of academic life. For one, there would no longer exist pressure to earn straight A’s, for this would become an unreasonable goal. As a result, students may be more inclined to take academic risks, perhaps enrolling in challenging courses in unfamiliar disciplines, without worrying that this exploration would damage a perfect GPA. Without any “easy A’s,” students may be inspired to work harder in all of their classes, not just the most difficult ones.
Naturally, challenges exist with any grading system, and this one is no exception. The familiar complaint that science students typically earn lower grades than students of the humanities (perhaps because it is easier to assign a poor grade on a problem set than on a paper) would remain. It is not obvious how one would standardize grades across departments or even if one would wish to do so. But this problem is not as serious as it seems. It is quite possible that Harvard attracts more students capable of exceptional work in one field than in another, and any raw indicator of performance would necessarily reflect this. Latin and English honors could continue to signal relative performance, as they do currently, and the College could consider normalizing the GPA calculation among different fields of study if departmental grades end up particularly unbalanced.
As Princeton is currently aware, the adjustment to any new grading system is bound to be rocky. Unlike Princeton’s gradual model, a drastic rescaling of grades would make it impossible to calculate a cumulative GPA for students spending part of their time under each system. Fortunately, fears that grade deflation place students’ futures at risk, in terms of graduate school admissions and job hunting, are largely unfounded, because Harvard’s spot in the limelight would make certain that any dramatic grading change, like the cap of A-grades at Princeton, would be widely publicized. With this in mind, students and educators may find that a dramatic attack on the perennial problem of grade inflation is well worth the risk.
Emily E. Riehl ’06 is a mathematics concentrator in Adams House. She is a member of the Educational Policy Committee of the Harvard College Curricular Review.