I was in a panel today at The Grading Conference (twitter @Grading4Growth) whose title was “Rethinking Assumptions.” We went mostly in the direction of assumptions about ungrading, so I wanted to write down a few other thoughts about assumptions about grading writ large.

This post was going to be a twitter thread but then I remembered that people like Robert Talbert are always telling me that most twitter threads should actually be blog posts.

Thesis statement: There are a lot of widely-held tacit beliefs that I think discourage many people from using alternative grading systems in their classes; fortunately, many of them are wrong.

You probably have more room to change things than you think

This is maybe the main hurdle I hear when I tell people about alternative grading systems. Many institutions have various policies that constrain instructors' choices around assessment. For instance, virtually every university in the US requires instructors to post final letter grades at the end of the term. At some institutions, there's a requirement to also provide midterm letter grades, and at others, there's a policy that x% of the final grade must be based on a timed final exam. The constraints these policies impose are real and I'm not trying to say otherwise. I will suggest, though, that there's room to operate within them. Ask yourself: what is actually required, and what requirements am I just inferring?

Asking such questions often reveals that nobody cares what you're doing in your class on a day-to-day level. So even if there's a requirement that you give a common final exam and grade it with points, that requirement doesn't say anything about your weekly homework assignments, or your daily prep work, or your final project. You're free to alternatively grade those however you like.

Asking such questions also often reveals spaces for creative insubordination. What if instead of writing a traditional final exam, you wrote down a bunch of learning objectives and then developed questions targeting each one, and then your final exam was just a stack of all of those problems stapled together, and you told students that they only have to answer the ones that they hadn't already demonstrated their understanding of, and you graded each problem yes/no, and students could reattempt similar problems during the exam period if they didn't get them right, and you then just reported the percentage of successfully-attempted objectives at the end? Congrats, you've fulfilled the policy, but you've just done standards-based grading inside it.

A key response to this challenge is academic freedom. This notion has been used in a variety of problematic ways by a variety of problematic people, but here's a place where we get to use it for good. If we truly have academic freedom and that means that we have a say about what topics belong in our courses, then it must also mean that we have a say in how our courses are assessed, because assessment is a proper part of teaching and learning. Similarly, this is a place where people who hold tenure can use that power for good (for once).

As we leave this section I will note that being a white man makes it easier to find space for disruption or creative insubordination. People from marginalized groups are certainly going to have a harder time of this, and that's a real constraint. But I think it's still possible to make meaningful change within the structures that we must inhabit.

Points are labels, not actually numbers

Hey, did you know that the average zip code of the state of California is about 93517.78?

A tacit assumption that we make about the points in points-based grading systems is that they're ratio data when they are at best ordinal. We like to make this assumption because it allows us to do math with our labels, but since it's an incorrect assumption, such math is actually nonsensical.

The funny labels used in alternative grading systems I think often scare people away who are used to thinking about points. But as soon as we accept that points are just another kind of funny label, then it becomes much less scary to get rid of “10/10” and replace it with “Excellent”.

The grading system we have now is not natural

By “natural” here I mean in the sense of “natural law.” Points-based grading isn't intrinsic or intensive to educational systems. It was not handed down to us upon tablets of stone. In fact, it's a fairly recent development: it's not really until about WW2 that the A-F system translated to the 4-point GPA scale becomes prevalent in the United States.

People who are thinking about adopting an alternative grading system often fret about various (good! important!) questions. “How do you determine the appropriate amount of work that corresponds to a final letter grade?” “How can I make sure I'm not biased against students from minoritized backgrounds?” “Shouldn't we worry about how this system impacts students with jobs?”

The secret is that all of these questions are also applicable to points-based grading systems. We just don't think so because points-based systems have become the unmarked category in which the answers to these questions are “obvious.” And in my opinion, points-based systems provide categorically worse answers to each such question.

Points-based systems are the result of a great many choices that were made by a great many people over a great many years. We get to interrogate those choices – and thinking about alternatives is an excellent way to do this.

You probably actually care about feedback loops

You probably don't care about points, deep down. It's easy to accidentally conflate the measurement of the thing with the thing itself, but I think deep down you know that the finger pointing to the moon is not the moon. What you care about is learning, and you probably know deep down that accrual of points is not the same thing.

Humans learn through feedback loops. I mean both of those words all the way down. What you care about when you are making marks on a student's paper is the feedback that you are giving them: you want to help them see what went right and what went wrong, and find a pathway to improve. And what you'd really like is a loop: you'd love it if there were some reason for students to actually care about your feedback, take your suggestions into consideration, and demonstrate improvement (instead of just looking at a number at the top of a page and cramming said page back in their backpack forever).

Every alternative grading system, at its core, is a means to provide feedback loops. In my opinion, any move toward feedback loops is a worthwhile one, because it's a move that will promote learning. And that's what we really care about.

You can get started

If you've made it to the end of this post, maybe I have talked you into trying out some kind of alternative grading system in some way in your course. Here's some tips for getting started.

A lot of people think that they have to jump in with both feet, and if you want to do that, then that's legit. But if you want to stick a toe in first, that's legit too. Try converting just one assignment to alternative grading. Maybe you'll like it, and maybe you'll see ways you can implement alternative grading in other parts of your course.