by Robert Rue—
We live in the era of Big Data. One can hardly make a point about anything these days without being asked to back it up with numbers.
So it seems like a short leap when students and their parents ask about the data behind students’ end-of-course grades. What’s the formula, they want to know. How much does each assignment count? How about class participation? Teachers who make all of this transparent—to the point where students can calculate their course grades at virtually any point in the year—can avoid a lot of arguments. No surprises at the end of the marking period for Johnny and his parents. The whole approach radiates fairness and objectivity, and this aura is particularly useful for humanities teachers who are more susceptible than math and science instructors to complaints about “subjective grading.”
But here’s the thing: An open grade book is a bad idea.
It’s not objective, it prevents teachers from giving the fairest grades, and it creates perverse incentives for teachers and students.
First, to the claim of objectivity: Grading is never totally objective. The sooner we can put to rest the false notion that it is, the sooner we can stop casting aspersions on subjectivity, which is in fact necessary for generating a fair grade.
People often believe that grades in math and science are objective, while evaluation in history and English is subjective. This distinction is misleading. First of all, any teacher in any subject can evaluate students on questions that have factual answers. This is what most people think of as “objective grading.” But in reality this grading almost certainly depends upon subjective decisions made by the teacher.
Let’s say a student gets the answer to a math problem wrong. Well, the teacher has a grading choice to make. She can give partial credit or just mark the whole problem wrong. If the teacher gives partial credit, a whole series of subjective choices ensues. How does one break down the various steps of a math solution and assign point values to each? Should she take off more points for a calculation error or more for reversing a sign? What if the student used the wrong formula from the very beginning? These are rather obviously questions that require subjective choices. Ah, yes, you say, but that’s why real math teachers just stick to right and wrong. Either you got the answer or you didn’t. None of this partial credit stuff!
Guess what? Giving two students the same grade—0 points—when one has made a tiny error and the other does not even understand what formula to apply (both got the answer wrong, after all) is a subjective decision, and in my view a bad one. And whether grades in this teacher’s course are “transparent” throughout the year or not, these students’ final marks will be deeply impacted by hidden subjective decisions.
The truth is, of course, that almost all grading entails some objectivity and some subjectivity. Some courses and some assessments tilt more one way than the other, but pure objectivity is a myth.
Fine, you say. Grading is inevitably subjective. But this is why transparency matters so much. If teachers are going to use subjective criteria—and there’s really no way not to—then at the very least they owe their students clarity and consistency in what all the decisions will add up to at the end of a marking period. Students should know, for example, that tests count for 70% of their grade while homework is 15%, quizzes are 10% and class participation is 5%. At the more micro-level, teachers should be responsible for explaining ahead of time how many points will be taken off for each kind of error on assignments and tests.
But even if this kind of transparency were possible—and I don’t think it is in any course that involves analysis in writing—this reasoning is flawed. Furthermore, I think most teachers know it’s flawed. I’d be surprised if there were a thoughtful teacher anywhere in the country who has not had the following experience with “transparent” grading formulas:
So you do everything by the book. You tell the students and parents right at the beginning of the year exactly how much every assessment is worth. At the end of the marking period, you plug in all the numbers and get your results. There are your grades—fair because they’re transparent, the same standard applied to all.
But then you see it. Susie got a lower grade than Mitch, but you know Susie is a better student. You know she understood the material more deeply. She made some mistakes, sure, but she got the harder stuff right and was always able to understand her errors better than Mitch did.
If you stick virtuously with your formula, you give the better student the worse grade. If you fudge the formula, you’ve given up what you thought was so valuable all along—the transparency. Different teachers will understandably make different choices when faced with this dilemma. Some will stick with the formula in the name of objectivity. I’m guessing that some will keep their formula and simply boost the grade on the final assessment or on class participation until it allows them to give the mark they want to give to their Susie. But no matter what we do, we’re right back in the realm of subjectivity, and either way, we’re being dishonest.
Obviously, I’m not in favor of a grading free-for-all in which teachers hide their methods from their students and then at the end of the year consult their deepest feelings to generate grades. We can certainly tell students what our standards are; we can give them models of great work; and we can let them know their grade-average from time to time. But we’ve got to stop demonizing the subjective. “Subjective” does not mean “based on nothing.” It means that judgment is involved, which, remember, we’ve already demonstrated is part of grading no matter what we do. And judgment is not just inevitable. It’s valuable. It allows us to focus on what actually matters—learning.
When “subjectivity” is demonized, teachers are inclined to avoid assessments that will be “difficult to grade.” “Difficult to grade” means that the judgment teachers must employ to evaluate the assignment will be impossible to justify in purely objective terms. So those assignments get jettisoned. But what’s so sad about this choice is that the things that are most important to evaluate in learning are things that are “difficult to grade”—things like insight, originality, curiosity, the ability to revise, the ability to be tenacious in the face of not understanding, the willingness to take risks, just to name a few. What a perverse environment our schools foster if we are incentivized not to include these things in a course for fear that we will be accused of being subjective!
And speaking of perverse incentives, being able to calculate their grades to the decimal point at any given moment in the year—one of the “perks” of the open grade book—does more harm than good for students. It encourages them to treat their learning like the air pressure in a bicycle tire. Grades too low? Time to ride on into the station for a few pounds of extra effort. Grades above the score you need to get an acceptance to Elite U? Time to cruise! Sure the tire might pop, but more likely it’ll just deflate to acceptably normal.
And besides the objectivity myth, the “transparent grade book” rests upon another false assumption—that it is fair for a teacher to decide ahead of time the precise formula for determining all students’ grades. In other words, apparently a teacher should know that 70% of the learning in a course will always—and for all students—appear on tests, 15% of it on quizzes, etc. etc. But what if one student demonstrates deep insight in class discussions and another on homework? Shouldn’t a teacher be allowed to exercise her judgment and give the grade that most reflects the student’s actual learning? Why would we trust our prefabricated formula over what we’ve perceived with our eyes and ears for many weeks? Because our formula is “objective”?
Instead of playing the objectivity game, why shouldn’t a school simply announce to all parents and students something that is obviously true: grading requires a teacher’s judgment. I don’t pretend that controversial discussions won’t ensue. I don’t pretend that some teachers will not feel unmoored when they have no formula to point to in these controversial discussions. But we can all get better at this. We can train ourselves to better articulate our criteria for learning. We can train ourselves to better observe the actual learning a student does in our class, to better articulate it, and we can weight the components of each student’s final grade accordingly.
In other words, we can make our grades more accurate by making them more subjective.
If I’m right about any of this, what’s truly unsettling—maybe for you and definitely for me—is not just that the open grade book is a bad idea, but that the “transparency” it offers is actually just a way of obscuring both the bad practices and the inevitabilities of grading in general. Instead of grading learning, most of us are overcommitted to grading that which is easier to evaluate and easier to justify. Instead of embracing subjectivity, we shroud it in “objective transparency.”
Some people will no doubt argue that grades themselves are the villains of this story. As I’ve laid out elsewhere (“What Are We Grading?”), I think that’s wrong, and in fact, I think it’s yet another way of avoiding the responsibilities of subjectivity. Grades, carefully conceived, based on the right criteria and accompanied by teachers’ comments, will be more clarifying for students than comments alone.
Let’s face it. We are in a difficult profession. Judging learning is, well… difficult. But we must judge, and in order to do it effectively, we must, it seems to me, be truly transparent about what it requires.
