Anthony Ding / anthology

Undergrad Teaching

1500 students, 70 instructors: that was a rough head count for UC Berkeley’s Spring 2020 offering of Data Structures, a lower division computer science course.

You see, Berkeley has an awesome undergrad teaching and mentoring system. My computer science classes usually feature legions of uGSIs (underGrad Student Instructors) doing vital work: writing exam questions, holding discussion sections, and building projects and their autograders. So when offered an opportunity this spring to join course staff for Data Structures, I accepted almost instantly. Having signed up for a lighter course load, I figured it would be the perfect opportunity to get some valuable teaching experience.

I can’t claim to be the best or most experienced teacher. Honestly, I don’t even know if I’m above average. But during the experience I did much more learning than teaching, and I thought I’d share some of that here.

The Feynman Technique works

Richard Feynman: Nobel laureate, physics popularizer, creator of those revolutionary subatomic particle collision diagrams I’ve heard referenced by physics-major friends. Also, a firm believer that teaching is the best way to learn. Teaching is the cornerstone of the Feynman Technique, which essentially says, to learn a concept, iteratively explain it to a layperson, or a six-year-old child, while filling in gaps in understanding.

Since it was an intro class, students had highly varying levels of exposure and understanding. Adopting the Technique’s core ethos of teaching to anyone, regardless of technical background, meant breaking down abstract concepts into concrete ones, finding analogies that still preserve nuances, and adapting to each student’s needs and thought framework.

In doing so, I found myself re-learning a great deal. To prepare for my weekly sections, I’d run through the material beforehand, many, many times. No matter how confident I felt about a topic, I could always find something to learn. And there were definitely a few times in the semester when I said, “I’m not 100% sure - I’ll make sure to confirm after class and get back to you.”

Imposter syndrome can really get in the way

But admitting that I wasn’t sure, or even just plain wrong, about a topic in front of my students was difficult at first. I wanted students to keep attending my sections. I wanted to earn their trust as a mentor. I didn’t want to be seen as incompetent, unworthy of teaching students roughly my age. Sometimes I wondered if my getting the job was a fluke, if they sent the email to the wrong person.

As a teacher, I soon learned that this only hurts your students. Your responsibility is to communicate accurate information, to help guide them through the maze of concepts, to elicit those lightbulb moments of understanding. To teach and learn with them. And so, maybe out of a sense of duty, I pushed those feelings out of my mind during section. I let my students point out mistakes, corrected myself and apologized often, and made clear the limitations of my knowledge. Honestly, they seemed to trust me more from this, which in retrospect makes sense. After all, I really disliked when high school teachers would say “I only made that mistake to see if you were paying attention.”

Imposter syndrome is a serious issue, and these days it feels like a universal experience. I can’t say I’ve found a personal solution, but at least I’ve learned to be mindful of it when teaching.

Negative reviews happen

Of course, the aforementioned imposter syndrome only heightened when I opened my course evaluations and surveys. “Course evals” are an integral part of Berkeley’s teaching system and enable students to give feedback on professors and student instructors. In addition, Data Structures also added bi-semesterly Google Forms for constructive criticism on a variety of topics, ranging from teaching pace to worksheet creation.

Before opening up the survey responses, I was admittedly freaked out about potential bad reviews. I’d seen some of the hurtful comments written about professors online, even the ones I thought were excellent instructors. In the end, placing these reviews in context really helped. A 2.5/5 in a subcategory here or there isn’t the end of the world, just another goal for self-improvement. Also, I had to remind myself that some people grade on harsher scales, whereas others would give out fives just for participation. Plus, sometimes students were venting their general frustrations at aspects of the course itself.

In aggregate, my evals turned out just about how I expected: mostly good, with a few things that I realized might need work. This isn’t to say take them with a grain of salt, and especially do be mindful if you see red flags like large swathes of negative ones. But it’s important to learn to tolerate small variations in reviews and to isolate important criticism.

Leverage technology to enhance student learning

In one course evaluation, a student noted how helpful it was to have a website dedicated to our section, with notes, worksheets, and other resources.

This all started the first week of school, when I developed a teaching portal on my personal website. Over the semester, I kept it regularly updated. We’d have a different worksheet every section, so I’d usually post those and the solutions beforehand. I even tried my hand at LaTeXing some personal notes, but stopped once I got busy with schoolwork.

After the COVID-19 pandemic swept across the US and the semester turned virtual, the teaching portal became especially useful. Students who needed a copy of the worksheet could easily find it on the website. Whiteboard notes from tablet drawing could be accessible at any time. Overall, actions that took me a couple minutes at most notably improved my students’ quality of learning during virtual instruction.

Content creation and preparation is half the battle

Weekly preparation, however, wasn’t just about maintaining my teaching portal. Preparation meant a lot more: getting ready to field questions about related topics, reviewing notes from prerequisite classes in case students needed refreshers, and even being mindful of speaking tone and pace. I didn’t always look forward to it, especially when I had three midterms and a project in a week. But it was necessary, as anyone who’s learned from an unprepared TA would tell you.

Content creation honestly was an underrated aspect of preparation. For instance, every week, three or four teachers would edit and QA the section worksheet, ensuring that information was correct, solutions were typo-free, and diagrams were aligned properly. In addition, we’d have to consider if the content was a good fit. Was this question adding value to the lesson? Did this sentence clear up misconceptions or simply create more? The extra work paid off: that week, teaching the lesson plan was a lot easier.

Yet the most important preparation to make as a teacher is not about lesson plans or textbook material. It’s really acknowledging that, as much as you prepare, you will still be blindsided. Some question will make you pause and search it up. Some student will find a more clever solution. At the end of the day, we’re all still learning.