Why Are We Still Grading?

There is absolutely no way to take a student’s work in any class and put a number or a letter to it in a way that couldn’t be done in another equally reasonable way, argues Dan Houck.

September 20, 2018

Next spring, I’ll be teaching my first course. While I’m eagerly planning the course sequence, designing assignments and considering various instructional methods, I keep coming back to the same question: How will I grade?

Or maybe the question is: Will I grade? You see, I know that I will have full authority over grading. The only requirement will be the final, end-of-semester grade, so I don’t necessarily have to grade anything prior to that. But should I?

Now when I say “grading,” I mean any quantified assessment of student work. I’ll be teaching a freshman writing course, so I plan to provide lots of feedback, but I want to focus on qualitative feedback. I’m even wary to go so far as saying that an assignment is above, below or meets expectations in any explicit terms such as those.

Why am I so wary? Well, the research on grading is conclusive: It’s bad for everyone. A few of the biggest drawbacks to grading include: causing students to look for the easiest way to get the best grade, reducing their interest in actual learning and lowering the quality of their work. And that just scratches the surface of research that shows the negative impacts of grading.

Are there any upsides to grading? Grades allow us to (attempt to) compare standards among different institutions, which is more or less the reason we have grades to begin with. No one had any idea what it would do to students who now focus on the grade rather than learning. And because professors know they have to assign grades, they often create assignments and write tests that will be easy to grade. Then they teach to those assignments and tests so they can give the students the grades and oftentimes forsake actual feedback. The grades are then used in high-stakes arenas like graduate school admissions and fellowships, which gives everyone more and more reason to worry about these always and inescapably subjective numbers and letters.

And they are subjective to the point of being arbitrary. There is absolutely no way to take a student’s work in any class and put a number or a letter to it in a way that couldn’t be done in another equally reasonable way or with which another qualified grader would necessarily agree. Can anyone really tell the difference between two similar assignments on a 100-point scale? Or even a 13-point scale (A-F with pluses and minuses)? I want to avoid all that trouble for me and my students. I want my students to concentrate on improving their writing, not their grades. I want to focus on helping them rather than justifying the grades I give.

Breaking the Cycle

So what can I do? If I had my druthers, I’d give no number, letter, symbol or other measure that could be inferred as a grade at all, but I can’t. And since I have to give a final grade, is it fair to leave students’ work unquantified until the end of the semester? They can’t be altogether blamed for being concerned with their grades with our system such as it is. To paraphrase the Declaration of Independence, students are more disposed to suffer while grades are sufferable than to right themselves by abolishing the forms to which they are accustomed.

And grades are apparently still very sufferable. But here’s the irony: according to the research, by not grading I’m helping them focus on more substantive improvements that will result in higher levels of accomplishment. When translated into grades, they would get higher grades than if they had been graded all along -- assuming my grading scheme isn’t so arbitrary it misses the difference.

A potential happy medium is contract grading. There are various shades of this, but the basic premise is to tell students what they have to do to get a certain grade and perhaps work with them to write the contract. For example, if they complete all assignments to my satisfaction (which may be less than their best), then they could be sure to get a B in the course over all. Some faculty members create a requirement for each letter grade and use the pluses and minuses to capture the finer shades of grading. If the professor is very comfortable with and clear about the requirements, it alleviates much of the pressure of grading, as students are automatically bracketed. That’s a great compromise, but I worry it already sets students up to see the course as just a series of steps to complete or boxes to be checked. Sure, for some a B will not be enough, but why give them any reason to think the grade they’ll get is a good indication of the work they’ll do?

Thoughts on rubrics are mixed, since they are often a precursor or proxy to grading. Providing rubrics with assignments is a way of informing students of your expectations, at least in broad strokes. But their typically bracketed layout with its various levels of quality for each learning objective is the source of some objections. You are literally putting the assignment (and so the students) in boxes before you’ve seen what they produce. And each level is often associated with a number or even an ambiguity such as “meets expectations,” which is where the grade proxy comes in.

One solution is to reduce the number of increments even so far as pass/fail or sufficient/revise to counter the association with grades. Yet that again poses the course and its assignments as something to be completed. It runs the risk of telling students that improving yourself -- i.e., revising a work that met expectations -- has no intrinsic value. Here again, I think thoughtful written feedback is most effective for us all.

Perhaps the best thing I could teach my students as writers is how to recognize when their writing needs work and what kind of work it needs. To this end, they will be given ample opportunities to assess themselves and their peers. With each assignment, I’ll ask them to provide a short written assessment when they turn it in. Drafts will be reviewed by peers, which helps students develop skills necessary for self-critique by seeing their strengths and weaknesses in others’ work. And examples of good writing will be provided and discussed in class as models to work toward.

So what about those final grades? As it is detrimental to learning to grade anything that a student is still working on, I’m leaning toward the portfolio approach. At the end of the semester, students will turn in all of their writing assignments, all of them having been reviewed and revised at least once, for a final grade. They’ll also write an assessment of their portfolio detailing what they’ve learned, where they’ve succeeded and what still needs improvement. Here, and only here, I’ll also ask them what grade they think they should get.

But how will I arrive at a letter for each? Ultimately, I have to have a scheme or a formula, even if it’s implicit. If I’m not ranking, I’m at least comparing. Or maybe I wait till the end of the semester to quantify the quality of each piece in the portfolio, assign a weighting and add it all up. Or shall I touch each portfolio to my head, awaiting a clear message from the gods of holistic grading, and then declare, “B-plus”? There’s no avoiding this, and I don’t know what’s right or even just best. What is the most meaningful way to assign the final grade?

I don’t know the answer to that, but I think it’s time to start breaking the cycle. Maybe, just maybe, I can teach my students that they know better what they’ve learned than any grade can ever indicate. Even if they don’t feel they suffer by grading and they object to my approach, we don’t always hurt when we’re sick, and the best medicine doesn’t always taste good.


Dan Houck is a Ph.D. candidate at Cornell University, where he studies fluid dynamics and wind energy. He will be looking for a postdoc position in about a year if anyone was wondering.


Back to Top