I had a little break from teaching CS222 last semester as I wrapped up work on STEM Career Paths. I have not blogged much about that project, but you can read all about it in my 2024 Meaningful Play paper, which I understand will be published soon. In any case, here I want to capture a few of the highlights and setbacks from the Fall 2024 class, and I promise, I'm trying not to rant about Canvas more than I have to.
Regular readers may recall that I tried a different evaluation scheme this semester, which I wrote about back in July. In September, I wrote a detailed post about some of my initial frustrations with the system as well as a shorter one about how I felt my attention being pecked away. I don't want to bury the lede, so I'll just mention here that to compute final grades, I went back to my 2022 approach, the tried and true, the elegant and clean system that I learned from Bill Rapaport at UB: triage grading. Between my failed experiment this semester and the similarly failed EMRF experiment from last year or so, I feel like I'm looking for a silver bullet that doesn't exist. It reinforces to me, yet again, that I should really be running some kind of workshops for local people here to learn about what makes triage grading superior.
I still want to track some of the specific problems of the semester, though, so that readers (including future self) won't walk into them. First, I tried to set up a simple labeling system in Canvas such that I could mark work as being satisfactory, needing a minor revision, or needing a new attempt. I made no headway here in part because of Canvas' intolerable insistence that courses are made up of points. I talked with a respected colleague who is willing to toil over Canvas more than I about his approach, and he mentioned that he encodes this information into orders of magnitude, something like 10 points for satisfactory, 1 point for minor revisions, and 0.1 points for new attempt required. Combining these together, students get a weird combination of numeric and symbolic feedback. He acknowledged that it wasn't perfect.
What I tried to do instead was to use Canvas' built-in support for grading as "complete/incomplete." Because that was all I cared about, I set the assignments to be worth zero points. When I used SpeedGrader, sure enough, the work was labeled properly. It wasn't until midsemester that I downloaded all the grades as a spreadsheet and saw that it only gave me the zero points. That is, whether the work was complete or incomplete was stripped from the exported data set. There wasn't so much data that I couldn't eyeball it to give students midsemester grades, which was facilitated by my recent transition to only giving A, C, or D midsemester grades (which are epistemologically vacuous anyway).
It wasn't until weeks later that it dawned on me that my students almost certainly had the same problem: Canvas was showing them zeroes instead of statuses. Of course, all my policies for the course were laid out in the course plan, and I do not have any qualms about considering those to be the responsibility of my students. However, when the university's mandated "learning management system" actively disrupts their ability to think about the course, it becomes more of a shared responsibility. About two weeks ago, I went in and re-graded all of the work to use triage grading instead, which allowed me to distinguish not only between complete and incomplete, but also between things that were submitted-but-incorrect and things that were not even attempted.
One positive change that I made this semester was counting achievements as regular assignments. This made processing them simpler for me, and I suspect it made thinking about them easier for the students too. While they have a different shape than the other assignments, they are "assigned" in the sense that I expect people to do them to demonstrate knowledge. I also set specific deadlines for them, spaced out through the semester. This reduced stress from the students by providing clear guidelines, since they could still miss one and resubmit it later by the usual one-resubmission-per-week policy. It also helped me communicate to them that the intention behind the achievements is that they give you a little side quest during the project-oriented portion of the course.
I had a really fun group of students this semester, as I mentioned in yesterday's post. There were still some mysteries around participation, though. I had several students withdraw a few weeks into the semester without ever having talked to me. It is not clear to me if they decided the course was not for them or if they were simply scared. By contrast, I know I had at least one student who was likewise scared early on, but who stuck with it, and ended up learning a lot. It is not clear to me if there is more I can do to help the timid students lean toward that mindset. Also, despite excellent in-meeting participation, I had many students who just didn't do a lot of the assigned work. I have some glimmers of insight here, but it still puzzles me: how many times do I need to say, "Remember to resubmit incomplete work?" I hope that some of the simplifications I have made to the course will help streamline students' imagination about it, but more than that, I am thinking about the role of the creative imagination. I am sure that a lot of students come into this required sophomore-level class without a good sense of what it means to study, to work, or to learn. My friends in the Biology department recently took their required senior-level professionalism course, in which students do things like make resumes, and made it a sophomore-level course. I wonder if we can do something similar to help the many students we have who are not well formed.
No comments:
Post a Comment