Students Are More Than Data Points
Students Are More Than Data Points
On student assessment, Spotify, and surveillance.
To end my final lecture of the Fall 2022 semester, I thought I would humor my students with some data I’d been collecting on them. Specifically, I parodied the then-relevant Spotify Wrapped to show them how they did on their homework assignments this year.
Complete with slick, zoomer-friendly slide transitions, I gave the class an overview of their performance for the semester, with an emphasis on positivity and vulnerability. After all, it was my first time teaching this course, so their problems taught me a lot too!
The next slide got the most laughs of any slide: a grand total of all the extension requests from the entire semester. As a rule, I allow students to request a no-questions-asked two-day extension on any assignment as long as they email me before the deadline. 62 students across 6 assignments asked for 97 extensions, with their fifth assignment (which coincided with a major deliverable in another one of their classes) having 38/62 students request an extension. I made it very clear that this was not to judge them; in fact, I thanked them for their willingness to ask for help when they needed it!
For this next slide, I wanted to demonstrate how far they’ve come. At the beginning of the semester, I polled them about numerous course-related items, from how well they knew other people in the class to what learning tools they were familiar with. Somehow, this particular group of Seniors weren’t that familiar with MATLAB, a tool that I had been planning to use in the course. Normally, UMass students learn MATLAB very early on in their Engineering degree program. But thanks to COVID-19, this group wasn’t attending UMass during the semester they were taking their MATLAB course; they were attending Zoom University, which had affected their retention of these skills.
In the same poll, 33% of students responded to the open-ended question “What are you most worried about in this course?” saying something about how the course was going to use MATLAB. So, rather than putting them through the gauntlet of coding their own S-functions, I made my MATLAB problems a bit simpler, with my most challenging problems having them modify pre-written code to do something new (which I also showed them how to do using in-class problems).
In the end, they did pretty well at these problems, so I wanted to end the semester with some positive encouragement. “See? You did the thing you were super worried about! You can do ANYTHING you set your minds to!!”
I finished up with two summaries in the actual 2022 Wrapped style. I used the “Top Genres” format to rank the hardest homework problem of the semester (of the ones graded up to that point). Throughout the semester, I have my TAs track the point distributions across individual parts of individual problems so that in future semesters I can know which problems to change (or scrap entirely). I thanked my students yet again for their perseverance, letting them know that they walked so that future students could run. On the right, I made some jokes about how long they must have spent in MATLAB, as well as their “Top Excuses” for asking for homework extensions (there were just made up by me, and not quantitative at all).
I hope all of this can serve as an example of how instructors can use pre-class surveys to fine-tune certain aspects of their course! (And an example of how humor can be used to students’ benefit). That said, it wouldn’t be an Anna Marie newsletter if I didn’t bring it back around to warn teachers about the dangers of surveillance as well as the quantification of education.
Spotify, the company being parodied in this final lecture, is notoriously evil. They underpay their artists. They have terrifyingly-enormous amounts of data on us, far more than what they share in our annual Wrapped montages. The CEO supports the military-industrial complex. They are the embodiment of the industrialization of music. In this way, they make the perfect parallel for the industrialization of education.
As engineers, our instinct is “more data is always better”. “Of course we should collect as much data as possible, that way we can make the student experience even better!!” To an extent, this is true; we should always be checking in with students about what is and is not working so we can sharpen our teaching skills and provide a better learning experience week over week, year over year.
That said, certain things are beyond quantification. My big pre-semester survey was simply a rough approximation of what I actually wanted to do: get to know every student in my class on a personal level. I’m glad I had a Google Sheet with all of my students’ self-reported skillsets, limitations, questions, and concerns (mostly because I have terrible memory). I’m also glad I kept track of grades so that halfway through the semester I could see who was struggling and reach out as needed. But this is no replacement for relationship-building. This is no replacement for chatting with students before class, starting class with memes and a breathing exercise to get us all centered, and establishing through my actions that I’m a person they can talk to if they’re struggling. (I will once again take this opportunity to remind readers of the horrors of student surveillance.)
About 3/4 of the way through the semester, a student in my class emailed me apologizing for their late submission, sharing that they had a 6-hour panic attack that night. Midterms had been rough and they were going through some familial struggles. I told them they were welcome to submit at a later time or drop this particular homework, and that they were welcome to come to my office if they needed someone to talk to.
Later that morning, they came into my office. I made them some tea, and I gave them a stim toy to play with while they told me about what was going on. I let them stay and co-work with me for a bit. Before they left, we did a breathing exercise together, and I even gave them a mini bisexual pride flag to remind them that even though family can be difficult, there is a whole community of people cheering them on.
As strange as it is to say, this was probably the highlight of my entire semester. I obviously wish that this student didn’t have a panic attack, but since they did, I’m glad they felt safe enough to come to my office and chat.
Our interactions with students are more than mere “data”, so trying to collapse them into mere data will always be an incomplete measure of who they are and how they’re doing. Our students are full, complex human beings, and so are we; this deserves to be honored. Next semester, my course will be entirely discussion-based, and will be entirely un-graded, trading in quantitative measures of success with qualitative measures of success. I greatly look forward to a mostly-data-free environment where I can form genuine connections with my students.
Cheers to a good first semester as a faculty member! 🥂
Currently Reading
More proof of the need for peer mentorship in STEM education! I have long been an advocate for formalized mentorship programs, at both the student and faculty levels.
Speaking of surveillance by tech companies, check out “The algorithm knows I’m Black” by Daniel Meyerend, or my past podcast episode about being a trans content creator on TikTok.
This incredible op-ed about the dark side of the NYC Christmas tree industry. Please, please read this one, it is horrifying yet hilarious. Merry Christmas, by the way!!
Watch History
My advice for students applying to grad school has always been that you need three things: 1) a project you like, 2) a PI that’s a good fit for you, and 3) a strong support network. UC Berkeley grad Dr. Fatima expands greatly on the latter two in this incredible new video, which is also partially about the still-ongoing UC Strike (quick plug for my new podcast about labor strikes!!) The video also references Dr. Chanda Prescod-Weinstein’s essay about abuse by professors, which is also a must-read for those starting out in grad school. Send both of these to your undergrads!!
I’ve recently been enjoying this ongoing video series about making authentic NYC pizza at home. It not only explains nerdy engineering things (like how dough hydration and CO2 bubble formation impacts the shape and texture of the cooked crust), but it has that special quality of “oh my gosh I can’t believe how much work you’re putting into something so niche” that I always love from an Internet show.
Bops, Vibes, & Jams
My favorite rapper of all time, Little Simz, just blessed us with a new album, “NO THANK YOU”. Go listen to the GOAT immediately!!!
If you’re hosting a holiday party these coming weeks, might I recommend my personal Christmas playlist or a Holiday Soul mix?
And now, your weekly Koko.
That’s all for now! See you next week with more sweet, sweet content.
In solidarity,
-Anna