Finding middle ground

Going to extremes with alternative grading isn't always best for students -- or instructors

Photo by Luke van Zyl on Unsplash

When you first learn about alternative grading, you might be so excited that you want to burn everything down and rebuild your courses from scratch. That can be exhilarating, and it might work out – or it might burn you out and convince you that alternative grading doesn’t work (see our origin stories for some examples that cut both ways).

Going all-in on alternative grading is kind of like cannonballing into the deep end of the pool. It can be great fun, but it also creates chaos that can swamp you as you try to come up for air. You’ll be changing a lot about a class: your grading procedures, how final grades are determined, and likely you’ll also need to change your course schedule, write new assignments, create reassessments... that’s a lot of work. It can be hard, if not impossible, to predict how all of those changes will work together. You’re likely to introduce unexpected consequences or incentives that you won’t see until you’re in the midst of teaching the class. Having to rework multiple parts of your course halfway through a semester is not a recipe for success.

If you are on fire with great ideas and want to change everything, I’m not here to stop you. For some, burning it all down is the only way to make changes. But as I wrote in Small Alternative Grading, it’s possible to work with just one or two of the four pillars and still make a big difference in your classes.

Making small changes in this way helps you learn what works for you and see how alternative grading fits in your context. That’s much better than discovering that your brand new syllabus needs major changes in the 3rd week of the semester.

Today, I hope to bring a message of moderation: When rethinking your grading practices, you don’t have to choose between all or nothing. There is a middle ground between making no changes and blowing up everything and starting from scratch, and often that middle ground is better for everyone involved.¹

Put limits on reassessments 

A key pillar of alternative grading is reassessment without penalty. Reassessments are a good place to start when rethinking your grading practices: Let students show you what they’ve learned, without penalizing them if they need to engage in the very human process of learning through a feedback loop.

But it’s easy to take this idea too far. The thought process is: Students should be able to show what they know, when they know it – and artificially limiting reassessments gets in the way of that process. So why limit reassessments? Let’s have a joyful, boundless celebration of learning!

As it turns out, unlimited reassessments are a recipe for disaster. For the instructor, more reassessments mean more assessments that need to be written, and much more grading. The “reassessment avalanche” is one of the most common experiences for new alternative graders.

Unlimited reassessments aren’t good for students, either, because they encourage students to take each attempt less seriously. This can lead to students making changes without reflecting meaningfully on previous attempts: That approach didn’t work out? No problem – I’ll make a little change and try again! … and again! … and again. This isn’t engagement with a feedback loop; it’s throwing ideas at the wall to see what sticks, and the result is low quality, unreflective attempts that don’t show meaningful learning.

My first few attempts at alternative grading had exactly this problem, and after a few frustrating days of grading reassessments that didn’t show meaningful growth, I realized the key issue. Unlimited reassessments work against metacognition, a critical part of a feedback loop. Students need time to think about their learning, to reflect on it, and to seriously engage with feedback.

So, I recommend a middle ground: Find ways to structure reassessments that encourage, rather than discourage, metacognition. Ensure that students have the time and space to really engage with feedback, by intentionally making room in the course schedule for reassessments. This might mean cutting out some other assessments to make room! Require relevant reflection or practice before submitting a reassessment to encourage metacognition. Limit the frequency of reassessments so that students aren’t trying to focus on too many things at once. You can find detailed advice in my post on reasonable reassessments.

Reassessments don’t have to be all-or-nothing: Students can get more benefit from a carefully structured reassessment policy than one with no limits.

Keep deadlines around 

While due dates for assignments aren’t the same thing as grades, they’re closely tied. What happens if a student can’t, or at least doesn’t, submit work when you want it? In traditional grading, there’s usually a grade penalty: Subtract a certain number of points per day late, refuse to grade late work without an approved excuse (effectively grading it “0”), and so on. Even in alternative grading, if we’ve put up barriers to “late” work, then we don’t get to see evidence of learning.

Part of the spirit of alternative grading is to let students show us what they know, rather than enforcing compliance with arbitrary rules – and due dates are absolutely arbitrary rules. 

Some instructors embrace this idea by removing deadlines entirely, allowing students to hand in work at any time.

Much like unlimited reassessments, having no deadlines can lead to some major problems. Structure is a key part of helping novices learn new ideas, and deadlines are a form of structure: They are a message from the instructor about what matters right now. Deadlines help students organize and prioritize their time. Going completely deadlineless removes waypoints that help students know what to focus on, and when to focus on it. Remember that many students are not only novices at our course content, they are also just beginning to learn how to manage their own time and prioritize many competing demands on it.²

Another common issue with the no-deadlines approach is that it can encourage procrastination, which works against deep learning. Without deadlines to provide guideposts, some students put work off as long as possible and then complete and hand it all in at once. Work done in this way can’t involve engagement with a feedback loop, because there’s no time to do so! Plus, this makes a giant pile of grading for you, right at the end of the semester. Again, the lack of structure hurts learning.

I’m not arguing for strict, unforgiving deadlines. That, too, hurts learning and doesn’t let students show you what they know. So here are two key pieces of advice for how to find a middle ground:

1. Don’t remove deadlines entirely. Keep deadlines as a way to communicate what matters, and when it is most important to work on it. This is why many instructors are starting to use the term “best by dates”.

2. Allow structured flexibility for those deadlines. One way is to use tokens to allow excuse-free extensions. Another is to allow limited or even unlimited deadline extensions, but only for one item at a time.

Whatever you do, if a student seems to be falling behind and if late work is piling up, have a conversation with them sooner than later. Sometimes they need a bit of help organizing their time and priorities, but you might discover that there are other accommodations that they need.

I’ve written a lot more about the importance of both flexibility and structure. Don’t lose sight of the fact that humans need guideposts, and there are many ways to provide those while remaining flexible and sensitive to student needs.

Set high standards but don’t demand perfection

Clearly defined standards are important parts of any assessment, and in alternative grading, the “grade” that is assigned for work – if you assign a grade at all! – is a mark that indicates progress towards meeting those standards. Even if you don’t assign a grade, you still communicate your thoughts about the quality of the work, relative to the standards, in your feedback. That affects a student’s understanding of their progress.

The key question that alternative graders then have to face is: What does it mean for work to “meet” a standard (which can also be called specifications, criteria, etc.)?

One reason to use clearly defined standards is because those standards tell students and instructors what matters on an assignment. Like the two sides of one coin, a clear statement of “what matters” necessarily implies that there are other things that don’t matter. Are spelling and grammar essential in a lab report? Likely not, as long as it’s understandable. But in an introductory foreign language class, spelling and grammar might be all that matters on some assignments. If you’re assessing whether a student can apply an advanced physical or statistical formula, arithmetic errors are probably irrelevant unless they show a misconception about the underlying concept. But if you’re assessing algebraic skills, then arithmetic might be an essential part of that. Context matters!

Here’s where the danger arrives: New alternative graders sometimes end up requiring perfection in order to meet a standard. Nothing short of expert-level work is acceptable. Requiring perfection treats all aspects of an assignment as equally important. It makes them all essential, because any error, no matter how trivial or irrelevant to the assignment’s purpose, leads to no credit. This makes it impossible for students to understand what actually matters. In turn, this means that students aren’t able to focus on learning key ideas, because everything seems to be a key idea.

The middle road comes from being reflective about your goals: What really matters, and what doesn’t? Practice writing clear standards that include what matters – and nothing more. Be willing to explicitly decide that certain things do not matter when assessing a standard. And then, hold yourself to using only those standards when assessing student work, and not penalizing them for minor or unstated requirements. You can always leave feedback on items that don’t fall under your standards – but don’t let that overwhelm the central message of whether the student is actually meeting the standard or not.

It doesn’t have to be all-or-nothing

Resist the temptation towards black-and-white thinking: There is not one “best” or “correct” way to use alternative grading, and there’s no need to blow everything up and start from scratch. Indeed, you’re more likely to be successful – and to make a bigger difference for more students, long-term – if you take a middle road.

Along the way, engage in your own personal feedback loop – you’re learning something new, just like students! – and make changes for the next time you teach that course. Implement a few more aspects of alternative grading in your next course. Eventually, you’ll find a collection of practices that are balanced and tailored to you, your students, and your situation.

Taking time to find what works for you, in small steps, will ensure a sustainable and successful path into alternative grading.

1. In this post, I’ll focus on how going all-or-nothing with alternative grading can overwhelm new practitioners and hurt student learning. But there’s another aspect of this kind of black-or-white thinking, too. All too often, people like me – who work to promote alternative grading and its benefits – talk about alternative grading in an all-or-nothing, “set it all on fire” fashion. That gets some people very excited. But it’s just as likely to scare away others who might be interested in dipping their toes into the alternative grading pool. That’s both because they think they have to fully commit to changing everything — an intimidating possibility — and also because they’re turned off by intensity and lack of empathy in the message. As a result, we end up turning away many people who could do so much good for students. But that’s a post for another day.

2. Faculty, too, benefit from deadlines to help prioritize their time. Just like with students, faculty deadlines are often best when they have some built-in flexibility.