Grades Are Stupid, but Oh So Useful!

Eben is correct: grades are stupid. This idea is not new. We have all probably thought this already to some degree. Worse, we have all probably felt this, perhaps very recently. Whether we have only now realized their stupidity by being placed lower on the curve than we had hoped, by uncomfortably sitting atop the curve, or simply by Eben informing us of their stupidity, we ought to ask: in precisely what ways are grades stupid?

First, let us attempt to air some different, but related, grievances with grades. One, they purport to measure some particular academic ability that they, in fact, do not accurately reflect for a variety of reasons. Performance on a written examination on tort law, for example, is intended to test one’s knowledge of core tort concepts. Doing well on such a test is only, if anything, sufficient for demonstrating proficiency in torts. It is possible that one was too rushed, too tired, too stressed, too unprepared, etc. Thus, doing well on a torts exam cannot be a necessary condition for being proficient in torts despite often being treated as such. It is more likely that the test determines, well, how well one takes the test. Moreover, it is graded on a curve. The grade assigned cannot be an objective evaluation of how well one knows the material because, by design, the grade reflects how ‘well’ one did solely in comparison to others taking the test. This way of grading is also troubling in that it takes one sample on which one’s entire grade is based. When reasoning inductively, the more data one uses the more solid the foundation that one’s generalization sits upon. It strains credulity near its breaking point to use this one piece of data as a measure of much at all.

Two, evaluations are arbitrary. Evaluations can be arbitrary either because they are not carried out in a systematically uniform way, or because they merely test arbitrarily chosen portions of the tested subject. Idiosyncratic preferences of the evaluator appear in both ways. Undoubtedly, we are all familiar with having an essay grade determined by how much the professor ‘liked’ it, or you, irrespective of the work’s more objective characteristics. This is a common complaint of the humanities or the ‘soft’ sciences, yet this applies to the ‘hard’ sciences as well, albeit in subtler ways. For instance, a math test may appear to be an objective measurement of one’s ability. But, why does the test only include two questions on derivations and not more? Why those derivations? Why is partial credit given for some sorts of partial work rather than others? Etc.

There are, of course, other problems with grades; yet, the good news is that each of the above problems is not inherent to grading, or evaluation, in general. The real problem lies not with the grades, but in what we do with the grades. Grades are intended to help us and others gauge our progress and abilities relative to our peers.

There are two parties concerned with grades: employers (or future admission committees) and ourselves. Each ought to realize the particular shortcomings of grades and place the appropriate amount of value on them. For our own purposes, this involves realizing that grades don’t do anything except stand in as a very rough proxy for how well one did compared to one’s peers. That means that one’s letter grades should play little to no role in affecting how one approaches academic improvement. What matters is the student’s response to careful feedback from others. For employers’ purposes, this involves realizing the near meaninglessness of fine-grained distinctions between students and that grades are but one small imperfect piece of a picture of one’s abilities. The problem cannot be that grades are inherently unjust or stupid because complaining about their inaccuracy, arbitrariness, and unfairness assumes that the grades are failing to get ‘it’ right. The ‘it’ that we are after is real and useful. The ‘it’ is one’s knowledge. Jettisoning grades altogether would not help us know who has more or less of ‘it.’ We need to make sure that grades are doing a better job at ranking people by ‘it’ and we need to realize the imperfect relation between grades and ‘it’ reflects teachers’ imperfect abilities as evaluators.

One might think that the problem with grades is one of reduction by thinking that it is a mistake to reduce something qualitative, namely knowledge, to something quantitative, namely grades. This is mistaken. Grades are not actually quantitative despite appearances to the contrary. They serve to rank students relative to their peers (even if the class is not on a curve, the grade will only be useful when the outsider knows roughly where the grade falls amongst other students at the institution, at that time). Ranking students relative to their peers requires no use of quantities. It is a rough ranking of qualities in the same way that we prioritize other non-quantifiable values in our own lives. Upholding a promise to my best friend is clearly more valuable than sleeping in until noon. Precisely how much more valuable? Who knows?

Grades are essential. They are needed to get things done. Imagine I am looking to hire someone and all I have is a pile of fresh graduates. Who am I going to pick? I could spend the time reading detailed evaluations from professors, reading all of the student’s written work, etc. All of this would be relevant information, but it is too time-consuming. I need to parse through the masses, and quickly.

Should we assuage our obsession with grades? Sure. Should we seek more careful feedback on work? Sure. But, do we need evaluations packaged in small, quickly digestible forms in order to get things done? You bet. Thankfully, we have those. They are called grades.