Grade inflation has run rampant since the pandemic, but people are finally starting to notice. Compared to 2022, fewer parents today believe their children are performing above grade level. Policymakers from California to South Carolina have pushed back on so-called “equitable” grading reforms, which lower academic standards in the name of “equity.”
Adopted in many schools in the Covid era, equity grading policies quickly became a major driver of grade inflation. One “equity” policy that continues to rear its ugly head is the grading floor (sometimes called “minimum grading” or “no zeros”), which requires teachers to award a minimum score—typically 50 percent—even when a student had earned a lower score or hadn’t turned anything in at all.
A persistent argument raised by some equitable grading reformers is that awarding a student a zero (or another low score) on a 0–100 scale is “mathematically inaccurate.” For example, responding to South Carolina's new law banning grading floors, grading expert Thomas Guskey told Education Week in August that the practice makes sense “from a mathematics perspective,” even if the policy “drives teachers nuts.”
Call me a cynic, but when I encounter the “mathematical accuracy” argument, I see a technocratic veneer covering a bleeding-heart impulse to lower standards and inflate grades. When grading reformers say that their grading scales are “more oriented towards success,” I have trouble believing that anyone is buying such an obvious euphemism for grade inflation. Yet the argument has marched on, essentially unrebutted.
So allow me this small contribution to the discourse.
The argument is laid out succinctly by Scott Marion in a 2022 blog post for the Center for Assessment. Marion argues that, on a traditional 100-point scale, roughly 60 percent of possible scores receive an F, while each of the other letter grades occupies only about 10 percent. “How does allocating nearly two-thirds of the number line for a grade of F make any sense?” he asks.
He then offers a thought experiment. Two equally weighted assignments receiving scores of 0 and 100 average to 50, an F. Yet if we converted those scores to letter grades first, Marion suggests, most people would say that an A and an F should average to a C. He concludes that assigning zeroes is simply "bad math." (Marion also recently wrote that “[grading] floors recognize the reality of a simple number line. This is not a woke argument. It is a mathematical one.”)
Yet these arguments are not about math.
The first is an appeal to intuition. Marion believes two-thirds of the number line should not represent failure, but there is nothing mathematical about that belief. A teacher or school requiring students to earn at least 60 percent of available points to pass is establishing a performance standard, not making a mathematical claim.
The second argument commits a more fundamental error: It incorrectly assumes that averaging numerical scores and averaging the letter grades assigned to those scores should always produce equivalent results. It is true that both the 0–100 scale and the 0–4 GPA scale are accepted for many purposes and that the GPA scale is more forgiving. Yet the acceptance of one scale does not invalidate another. Grading scales, whether 0–4, 0–100, pass/fail, check marks, percentage scores, or anything else, are social conventions, not mathematical truths.
Indeed, teachers routinely employ various grading systems, and each involves judgments about what different levels of performance mean. A teacher might award 40 percent on a five-question quiz because a student answered two questions correctly. That teacher might give scores from 0 to 100 on an essay based on a grading rubric. And she might employ a “check” scale on a low-stakes homework assignment. Neither mathematics nor measurement theory can dictate whether a given score merits “passing” or exactly how results ought to be averaged into a final grade. It is the job of teachers, administrators, and policymakers to balance all the factors that play into how grading can be calibrated to preserve transparency about performance and motivate students.
(And what could be more demotivating than earning a 60 after trying your best, only to find that your classmate received a 50 for doing nothing?)
Of course, if we want to debate mathematical accuracy in grading, it is the grading floors themselves that do violence to basic math. I once talked to a teacher who opened his digital grade book the day after their district adopted these policies and found that all grades that had been lower than a 50 had automatically been updated (i.e., inflated) to 50 percent. The idea that a kid who gets one question out of 10 correct on her homework deserves a 50 percent is the one that truly defies mathematical sense, which is probably why so many educators say grading floors harm academic engagement.
The problem with the grading reforms is not that they wish to make grading more standardized and sensible; it is that they ignore the many ways that their policies can lower standards and expectations for students. Especially in our era of grade inflation, these trade-offs must be met with open eyes—not couched in misleading language about mathematical accuracy.