assessment as proof of learning or as learning itself?

Recently, an idea has been bubbling in my head that’s the culmination of months—even years—of thinking about how I assess in my courses. I’ve typically taken the pretty-standard approach that assessment is the process of students’ proving that they’ve learned something. What if, though, assessment is itself the proof of the process of students’ learning something. That is, what if we doled out points for students’ proving that they appropriately participated in learning activities and then trusted the learning to happen on its own?

In some ways, this rubs me the wrong way. As I’ve written before, I don’t like grading attendance or participation in class. Yet, surely, that’s the logic behind attaching points to attendance or participation—an assumption that attending and participating in class leads to learning and that we should therefore police it. Furthermore, this is the logic I use when I grade Hypothesis annotations in my online classes: I don’t grade the content of the annotations, I just use the act of annotating as proof of completing the readings and discussing them with classmates.

This semester, I’ve also been struggling with how to measure proof of learning. Has a student learned nothing from an activity with the command line just because they don’t make the same observations that I would? Does a student not know how to adapt games for a library just because they don’t draw from the same readings that I do? In both the classes that I’m teaching, I’ve been inclined to grade students’ based more on whether they’ve followed the instructions and less what the specific outcome of those instructions are. Part of that is surely grading fatigue during a busy semester, but I’ve also been thinking hard about the idea that if the instructions are well thought-out and if a student has followed them to the letter, there ought to be some learning going on.

This last bit of thinking feels especially important in the age of ChatGPT and its ilk. I’m not excited about these technologies on a grand, social scale, but as an instructor, I agree that the best solution is generally to adapt our assessment, not to police or surveil its use. With the advent of generative AI, “product” seems less and less trustworthy as a demonstration of learning, and “process” seems more and more important. At the risk of exaggeration (though not much—this example was true of cheating before OpenAI was ever founded), the same essay could be either the natural result of a rigorous learning process or the result of phoning it in and having someone or something else do the work.

So, even though I’m still not keen on the idea of grading attendance (perhaps because I’m not convinced that attending class is itself demonstrative of learning), I’m starting to think of attendance less as the final outcome of a process of learning and more as the process of learning itself. These are initial, messy ideas, but as I start to think about my Fall classes, I’m sure they’ll continue to bounce around in my head—and hopefully refine themselves along the way.

• macro
• Work
• teaching
• assessment
• generative AI
• Hypothesis
• ICT 202”,”LIS 618
• grading