In this series, I have been arguing that grade inflation, or the allegation that widescale grade increases have occurred absent corresponding learning and achievement gains, is a red herring. Whether grades are going up (and there is compelling evidence, if not definitive, that they are going up overall in higher education) because teaching techniques are more effective, or standards are going down, or both (or neither), is complicated and messy. But the real issue here is that when grades go up across the board, they become compressed, thus narrowing the distribution of grades.
Compression is clearly a problem for Team Sort—i.e., people who believe education’s primary purpose is to identify the most promising and highest achieving students—because narrowed grade distributions make it harder to distinguish and rank students. But the implications of compression for Team Support—i.e., people who believe education’s primary purpose is to cultivate learning and growth for all students—is less obvious. Adam Mastroianni reminds us that teachers are not paid by the businesses, graduate programs, and scholarship committees that seek our classifications of students. Why not focus on supporting all students and leave the job of sorting to those other groups?
Over the next two posts, I will argue that compression is also a significant problem for Team Support, no matter how well-intentioned its causes. To make this case, I pick up my previous post’s thread on alternative grading, which is the primary mechanism by which Team Support operationalizes its values.
Let’s begin with a goal that unites alternative grading practitioners across their (otherwise heterogenous) movement: designing learning environments where students experience less distraction from coursework because of grades. Education should challenge students and push them past their comfort zones, but I imagine almost every alternative grading practitioner recognizes that when concerns over grades cause or exacerbate acute anxiety and debilitating distress, students have less mental bandwidth to focus on course content; this is counterproductive to learning.
Here is the problem in a nutshell: it is very hard to imagine alternative grading structures that reduce the (cognitive and affective) impacts students experience from grades without narrowing grade distributions relative to courses that employ traditional grading. But the more grade distributions narrow, the less effective alternative grading approaches will be at decentralizing grades. To put this another way, factors that contribute to grade compression within individual courses resonate with the values that motivate Team Support (this is the current post’s focus). But in the aggregate, compression ceases to align with Team Support’s values (this will be the focus of my next post).
Grading in the Age of Precarity
Reclaiming bandwidth for learning is a worthwhile goal impeded by the fact that large numbers of students and their families have come to view a college degree as less a catalyst for moving up the socioeconomic ladder than as a handgrip for not falling off the ladder entirely. Our courses reside within a highly precarious, sorting-driven educational ecology that leads many students to fear that low grades—often meaning anything lower than an A—will compromise their career prospects. This fear (whether rational or not) compels students to fetishize acquiring (and not losing) points.
So, if teachers want students to stop concentrating so much on grades, and (so many) students want to procure As, we see a basic formula taking shape, whatever one’s specific approach to alternative grading: the less time you find yourself talking (bargaining? arguing?) with students about points, and the more time you find yourself discussing their understanding of course content, the less likely students are obsessing over grades in your class (and projecting that obsession onto how they interact with you). The absence of haggling, then, is a strong signal that students trust you, which is no small feat. But if you are still spending a lot of time quibbling over points, then in practice students don’t trust you (and it might be time to reexamine your methods).
Now, how alternative graders go about earning trust can range widely, but I argue that across the alternative grading landscape, the price for decentralizing grades is to provide students clear and feasible routes to an A.
The Price of Trust
This was the logic I followed in my Failure Club courses, where I challenged students to “get an A for failing” at ambitious projects they found personally, intellectually, and creatively meaningful. I endeavored to take grades off the affective table, so to speak, by telling students they had begun the semester with an A that they needed to maintain rather than earn from scratch. Assignments were graded for completion, with students largely determining the specific content of what they created and submitted. Not everyone finished with an A, but most did: a collateral effect of cultivating exploration and play in a system poorly designed for these experiences.
Granted, this course was an unconventional, even radical, experiment in generative failure, the idea that we can learn much from traversing pathways that (ostensibly) lead to intellectual and creative “dead-ends.” So, let’s look at a less extreme approach, which (as noted in my last post) I believe is exemplified by David Clark and Robert Talbert’s four pillars of alternative grading. These pillars, which centralize clear standards, clear feedback, and multiple opportunities to demonstrate proficiency—including do-overs without penalty—represent what is arguably the prevailing ethic of alternative grading (especially in STEM fields).
Imagine a scenario in which a student “bombs” their midterm for whatever reason (maybe they didn’t prepare enough, maybe they were sick, maybe they just had a bad day…). In a traditional course employing “one-and-done” assessments, that midterm grade will hang around the student’s neck, weighing down their overall grade even if they show significant improvement on the final exam. Whereas in a class abiding by the four pillars, this student could retake the midterm—potentially more than once, depending on the course design—meaning their improved performance would show up in both the midterm and final exam grades.
Clark and Talbert would surely add that the student in the first case might be so emotionally weighed down by the midterm albatross that they decide it’s not worth preparing for the final at all, leading them to fail or withdraw from a course they might otherwise have passed. While the student in the second case, relieved of worrying that one (inherently high stakes) bad performance earlier in the term will doom their prospects for an A, has good reason to redouble their efforts. This logic underlies the theory of the case for earning trust in four pillars courses, even if specific approaches (standards-based grading, specifications grading, hybrids of the two, etc.) have their contextual idiosyncrasies.
Such courses may not produce as many A-range grades as did my Failure Club experiments, but grades should go up in general: teachers, after all, go out of their way both to make expectations clear and to provide scaffolds for students to meet these expectations. And students are not competing for an artificially limited number of As; every student who successfully meets the designated expectations will receive one.
The alternative grading movement has flourished in recent years as ever more teachers recognize flaws with traditional grades. This is not to say that alternative grading is the cause, or even the primary cause, for the compression trends that seem to be occurring in higher education, but alternative grading is surely an important contributing factor. And again: whatever its cause(s), grade compression makes it harder to sort students by achievement and aptitude, which creates serious complications for Team Sort.
In fact, there is evidence that Team Sort is already … out of sorts. But remember that Team Sort is the perpetual big dog in the sorting vs. supporting dynamic. As I address in the next post, this means Team Sort’s terrible vexation offers Team Support little reason to rejoice.
Teaching Inquiry 