A start/stop/continue for Fall semester
Robert's short list of changes and continuations for alt-grading in the new year
All summer long, August has been on my radar screen. I took May, June, and July to completely1 disengage2 from “work”. But August, I told myself, would be when I’d ramp back up to doing “actual work” most of the day. So unlike a lot of academics, the impending start of the school year didn’t sneak up on me. I’ve been thinking about the upcoming year all summer, and I’ve reflected a lot on what worked for me in 2022-2023 and what I’d like to change, especially with respect to grading.
Today I’m sharing that with you in start/stop/continue format like I’ve done before — one thing I am not currently doing with respect to grades and grading that I plan to start doing; one thing I am doing that I plan to stop doing; and one thing I am doing that I’ll continue to do.
For reference, this link will take you to the most recent draft of my syllabus for my Fall semester course (two sections of the same class). Be advised, between now and August 27 this will be going through lots of changes.
Start: Using a three-tiered model for assessment
Back in 2019 I wrote at my main blog about this idea I had for flipped learning design where we take the familiar pyramid for Bloom’s Taxonomy, split it into thirds (two levels each), and map each third onto a specific learning context: Before class, during class, or after class.
Under this model, class meetings in flipped learning become primarily focused on Application and Analysis, the crucial passage in the overall learning process through which students go from mere basic knowledge to mature analysis and creative problem solving. This has turned out to be a very fruitful and useful way of thinking about flipped lesson design (and not just flipped).
When I was building the class I taught last semester for engineering students, I ran into an issue: While I was placing a high value on the middle third of Bloom in class, all my assessments were targeting only the lower and upper thirds. I was lacking a real feedback loop in that crucial middle third. So I made a class of assessments called Application/Analysis (because I am terrible at naming things).
The idea was simple: Students were doing active work in groups during class, so I just flagged a subset of each day’s group work and had students complete a writeup of that work each week, which was turned in through our LMS and graded either “Success” if it was complete and more-or-less correct, and “Retry” if it wasn’t. Students got one revision if their work was marked “Retry”. (Only one, though; I figured this incentivized engaged group work.)
Application/Analysis last semester was an experiment that I kind of threw into my syllabus at the eleventh hour. It was OK but had a lot of flaws. For example, students tended to work laser-focused on the group work problems I flagged but ignored the rest; the difficulty level of the problems was all over the place; and I didn’t have a system for staying on top of the grading, and fell behind a lot.
What I am “Starting” this semester is a more intentional use of these Application/Analysis assignments with some corrections and improvements from their first introduction. For example, I won’t be flagging the problems to turn in until after class; and I’m keeping tighter control over variations in difficulty. I am keeping, with some changes and also with some exceptions, the assessments that target the lower and upper thirds. The result, on paper so far at least, is an assessment and grading framework that looks like the activities we do in the class, and which is aligned better with those activities.
I’ll keep you posted on how that goes.
Stop: Unlimited deadline extensions
For some time now (example, example) I have advocated for removing most if not all deadlines from student work, or if that takes things too far, giving students more or less limitless flexibility on deciding when they will turn in their work. Last semester, as in previous terms, My deadline policy looked like this:
(Click here to read the whole thing.) Basically, if you need extra time, take it and let me know what you’re up to.
This policy conformed with research on deadlines in private-sector jobs, made students happy, and got me retweets. It all worked fine. Until it stopped working fine. At first students used this policy sparingly and judiciously. Then, I started getting due date changes several days after the deadline (a loophole in the syllabus — there was no deadline on when to submit a deadline extension!). And several due date changes each week from some students. And due date changes placing then new due date 2-3 weeks into the future. And due date changes, followed by changes to those deadlines, and then changes to those deadlines, and so on.
I don’t mean to blame or shame students here; only a portion of the class ever used this policy at all, and most didn’t push boundaries. The problem was the push-ability of the boundaries. It might or might not have helped students to be this flexible; I couldn’t tell, because the situation with students’ graded work devolved into chaos once the deadline situation became something of a free-for-all.
So this semester, I’m ditching this policy and going back to using tokens for deadline extensions. Each student gets five tokens at the beginning of the semester; among the things that a token can buy is a single 48-hour extension to any deadline, which can be used once per assignment. There will be opportunities to earn more tokens, and there is syllabus language saying that if you run out of tokens and it’s a problem, talk to me.
I think tokens offer all the flexibility that the choose-your-due-date approach offers, while also placing useful boundaries on deadline extensions. After all, as I wrote before, deadlines in the real world are primarily there as commitment devices, and it seems counterproductive to have such things without commitment to go along with. When you spend a token, you’re making a commitment and it costs you something, even if it’s fake. I think that’s helpful for students to encounter.
Continue: Onboarding in the first week
This post at my main blog went semi-viral back in December when I proposed using a 12-week schedule to build courses in a 15-week semester. Specifically, we put all course content content into weeks 2 through 13; spend the first week of the semester doing onboarding activities to get students acclimated to the course; and use the final two weeks for review and reassessment with no new content coverage.
This semester I am definitely continuing this design pattern (although that next-to-last week is hard to clear out) by focusing the entire first week of classes on onboarding:
On Day 1 (Monday) we will do two activities. One introduces some of the main ideas of the course through a puzzle problem. The other introduces the concept of feedback loops, where I ask students to share something they are good at doing, and also how they got good at it. Stay tuned for an additional blog post about that second activity.
On Day 2 (Wednesday) we will practice with the grading system by looking at hypothetical students and their accomplishments at the end of a semester, and the students are supposed to go to the syllabus and assign the course grade based on that list of accomplishments. This never fails to be awesome — students figure out how to track and compute their grades, and there is always some great discussion about how it works.
On Day 3 (Friday) we will do some hands-on workshop style activities with some of the course technologies and take time to work on a “Getting Started” assignment that pulls all these ideas together.
Then we kick into the course content the Wednesday following Labor Day.
This “startup week” approach does a lot of great things. It gives students time to get acclimated to the course (and late additions, of which there are always several, aren’t so far behind). It improves their facility with the syllabus and grading system so we don’t have to talk about those things so much later. It also helps me, because these classes are easy to prep for and they are not heavy lifting during week 1.
So while I’ll be sad for summer to come to an end, and although I’m in no way logistically ready for the start of the new school year yet, I don’t dread or fear the start of the new year. I’m looking forward to working with a new cohort of students in a class (Discrete Structures) that’s probably my favorite to teach. I’m hopeful that I’m learning from my mistakes and misfires and turning those into something useful for students.
Well, mostly completely. I had a few meetings I had to attend and one talk I gave because it was supposed to happen in March but got postponed.
I wanted to write more about this but David said it didn’t fit the piece. Short version: I said no to almost every work-related request; taught no classes and did no research; worked on personal learning projects and writing for three hours a day; practiced bass guitar for another three hours; then did a combo of exercise, reading books, and playing with my band out on the lakeshore. In other words I practiced my retirement for three months.