This will be a pretty short post, because I’m (still!) recovering from my first bout with Covid… 3 weeks later.1 The past few weeks have been a vivid illustration of a little-discussed benefit of alternative grading: its resilience in the face of the unexpected.
Illness, unplanned class cancellations, going slower than expected, discovering that an assessment plan is unworkable… Alternative grading has helped me recover gracefully from a number of unplanned problems, and today I’ll tell you about some of them.
Prioritizing my goals
I’m teaching Calculus 2 this semester, where I’m assessing a list of 15 standards (which I call “learning targets”). My bout with Covid hit fairly late in our semester – week 10 of 14 – when we were in the middle of the last few targets on the list. While recovering, I had to cancel one 2-hour class block, and cut another one short – so I lost 3 hours in the last few weeks of class. There wasn’t enough time left to cover all of the remaining learning targets. How could I decide which content to cut, and which to prioritize?
Luckily, the structure of my list of standards helped me out here. Before the semester, I’d identified 5 core targets, the 5 most central ideas in the class. These also happen to correspond to the main topics identified in our syllabus of record. The other 10 targets are important, but not core – they’re extensions of, or deeper dives into, some of those main topics.
So when I was faced with having to cut 3 hours of class content, I knew to prioritize the remaining core learning targets and de-emphasize the non-core targets. It’s a choice I would rather not have to make, but at least the choice was clear.
I’ve made similar choices in the past: The Great Pivot of March 2020 was chaotic and unpleasant, but my lists of standards made it easier to decide what content to prioritize.
Others have noticed this as well: When I interviewed Joshua Bowman of Pepperdine University about his alternative grading system, he described how pre-prioritizing his standards (he uses 3 levels) has helped him make decisions in more than one unexpected situation. Even before March 2020, Joshua had to make similar decisions in the face of California wildfires that forced their own online pivot. He said: “At times of crisis, having a list of standards or tasks makes it much easier to sort out what must be kept and what can be trimmed for the sake of guiding students to the essential elements of the course.”
Changing how I assess a standard
I ended up deciding that only one target needed to get cut. Here’s where I made another choice, enabled by alternative grading: I changed how I assessed the target.
Normally, I assess learning targets on regularly scheduled quizzes. These are sometimes in-class, sometimes take-home, but the targets are announced well in advance. I also ensure that each target appears on 3 quizzes, so that students have multiple attempts to complete it.
When I realized that I didn’t have enough time to include one target on a sufficient number of quizzes, I stepped back and thought about my goals. I want to see evidence of student learning for that target — but that doesn’t have to be done on a regularly scheduled quiz. So instead, I set aside one hour of class time for us to work through the topic together, on a guided worksheet. It was a heavily condensed treatment, but enough to get across some key ideas.
During that time, I circled around class and talked with each student at least briefly about their understanding. That was it: A sufficient assessment of their understanding of this non-core target. I recorded a Successful mark for each student present and had them record the same on their grade trackers.2
Reducing assessments
Much earlier in the semester, I had to make another unexpected change to my Calculus 2 class. In addition to the learning targets assessed on quizzes, I also have written “advanced homework” problems. These focus both on integrating multiple topics and on written communication, making them somewhat more advanced “regular” practice homework. These homeworks are graded holistically using a list of specifications. The entire homework assignment earns one mark: Successful or Not yet, and students could submit revisions in future weeks.
I carefully planned out the workload for both me and students: Quizzes and advanced homeworks would alternate weeks, so that students were only ever focusing on one assignment at a time, and I would only have to grade one at a time.
Well… not quite. I had planned to assign two advanced homework problems every other week, and I even pre-planned what those assignments would be. I felt very smug about having a whole semester’s worth of homeworks pre-written before the semester began.
Then the first two homework assignments were submitted, and I was suddenly completely underwater. I’d badly underestimated how much time it would take to grade these longer, more involved written assignments (in addition to having quite a bit of additional grading in what has turned out to be an overly full semester). I waded through the first batch and realized I couldn’t possibly grade that many papers again. I decided to cut back to just one advanced homework every other week.
Here I had to make some bigger changes to my plans. That’s because my syllabus specifies the number of Successful marks that students had to earn, across all advanced homeworks, in order to earn each letter grade. With half as many assignments possible, those numbers had to change too.
So, I edited the syllabus, updating the numbers for homework requirements, highlighting them in red. I announced the changes in class and posted about them on our LMS. It was particularly easy to make these changes for students, because I had already handed out a grade tracker where they could record their progress on each assignment, by checking off boxes next to the various requirements. I simply had students check off the boxes for the now-canceled assignments. To earn each letter grade, they need only check the remaining boxes.
I assure you that nobody complained, and my grading workload has been much more reasonable. A phrase I’ve heard is this: The syllabus is there to serve us — instructors and students — not the other way around. Don’t be scared to make changes when they’re in the service of health and learning.
… and more options
As you can see, it’s been a bit of a rough semester. Nonetheless, it could have been worse – my assessment system actually helped make changes feasible.
In the past, I’ve had other classes where I had to make unexpected changes. Here are other methods I’ve used:
When grading holistically with specifications, remove individual specifications that turn out to be problematic – especially specifications that turned out to be irrelevant or focused students in unproductive directions.
When grading multi-step projects that involved, for example, a draft, peer review, and a revision, I eliminated individual steps (such as the peer review) and removed those requirements from the project’s specifications.
In many classes, I alternate between “assignment weeks” and “revision weeks”. Assignment weeks involve new assignments, while revision weeks are chances to revise and resubmit any previous assignment. Students have sometimes struggled more than expected on assignments, leading to a backlog of assignments that require revision (I usually allow revising only one assignment per week, to keep my grading load reasonable). I replaced one assignment with an extra revision week, giving everyone a bit of breathing room.
In some classes, I require students to complete each standard more than once. This can be a problem for standards that we only study near the end of the semester. If the class goes slower than expected, there might not be enough assessments available for students to complete each standard enough times, much less wiggle room for reassessments. In this case, I simply reduce the number of completions required for only those last few standards. I’d rather see that students know what they’re doing once rather than removing the standards entirely.
I’m also teaching other classes this semester, including one upper level class filled with students whom I’ve known for several years. That one was even easier to change when I first came down with Covid: I removed one homework set and replaced it with a revision week, and then posted some specific instructions for what they (with pre-made teams) should do during class. It worked wonderfully; they even sent me a video of everyone in class waving and saying “Hi, Dr. Clark!” Nothing helps more in a crisis than having established, trusting relationships with students.
Finally, I always take detailed notes in a file titled “Notes for next time.” Next time I teach Calculus 2, I’ll start with a better plan, one informed by what worked and what didn’t this semester. I’ve already noted that maybe that one target that I cut back didn’t need to be there in the first place. It’s a topic we can still see during a class period, but doesn’t need to have a formal assessment. Iterative improvements like this lead to simpler grading systems with more room for students to show what they know, and fewer unnecessary hoops to jump through.
Update: Tested negative, finally, over the weekend.
The few students who missed class that day can earn credit for the target by completing the worksheet – it’s posted on our LMS – and bringing it to office hours for a brief conversation, which includes some tutoring from me if needed. I didn’t want to penalize students for missing a day of class, especially when I’d just canceled one for health reasons!
Hi David, Sorry about your dance with COVID. I’ve read several posts in the STEM domain about the usefulness of target outcomes, success-not yet, and so on. So far a few points raise questions.
1. In every case the math teacher makes an individual list of targets, and it seems like the course is tightly organized and coherent. Must these targets be individually authored? Is there some way teachers can collaborate so that the work is shared? My experience at the university is as a professor of language and literacy—teachers used to collaborate to do this sort of planning with multiple benefits. I don’t see it happening much. Maybe it is and I just don’t know about it.
2. There is a lot of what some have called “I-ness” in the posts of STEM teachers involved in ungrading. I did this, I thought that, etc. don’t get me wrong. I’m not trying to offend. The same is especially true among English teachers, my wheelhouse. The effect is sharing “my” experience, which makes me think an effort inside the institution to build a structure inside say Center for Teaching and Learning where all this individual work can become fingertip tools for everyone.
3. Do you involve students in self-assessment before you assess their work? Do they have any peer assessment and feedback? In this ungraded perspective which is highly individualized, what is the role of student collaboration?