An anecdote about authentic assessment

At last month’s CAPS meeting, our guiding coalition was asked to reflect on the extent to which our team’s gradebooks were an accurate reflection of student learning. That is to say—as a team—do we avoid things like grading on compliance or completion? Can a student with an A in a class honestly say that they’ve mastered grade-level standards for that class? By extension, can we as teachers honestly say that a student who has a D or an F in a class has not grasped grade level standards? It’s in my nature to be a reflective (read: overthinking) person. So naturally, I thought about my own evolution as a teacher. 

My first four years of teaching were at a school where we were required to input three grades per week—our evaluations depended on it. When I came to Hart in 2017, I was in the habit of blindly putting in X/10 in my gradebook for things like whether or not my students were in uniform each day that week. Wild, right? These were English classes. Can the color of a student’s sneakers reflect how coherently they can defend a thesis? Spoiler: it can’t. I quickly adjusted to the practices my Hart English department colleagues had on the books, and in no time, I felt I could honestly give a bird’s eye look at my gradebook and say that, yes, if a student wasn’t doing well in my class, it was simply because they were not mastering the standards of the course. 

Fast forward to last month’s CAPS meeting, when our guiding coalition colleague, Zach Koebel shared that his Reach students often feel frustrated by the arcane nature of the weighted grading system. They’d come to him feeling baffled when an assignment worth 100 points barely affected their grade (10% weighted category) while an assignment worth 20 points dropped their course grade by a significant margin (40% weighted category). Cue me, feeling implicated—I have a talent for being able to take everything personally. I had a guilty conscience for a specific grading practice of my own that I knew fit the description of what Zach was discussing. I’ve known for a while that I needed to change one assignment in particular. Be it pride or be it conviction, I never changed. I made moderate adjustments over the years, yes, but I never truly fixed the problem. Please note that I am in no way about to make a case against a weighted grading system. I love a weighted grading system. 

In addition to being an over-thinker, it’s also in my nature to be completely candid with people regardless of how it makes me look, so here I go, getting vulnerable. 

I have this assignment in my AP Lang classes that I’m particularly fond of: my podcast challenge. I require students to select a different podcast every quarter and log six hours of listening over the course of ten weeks. I also attach a series of Socratic seminar questions that students respond to via typed responses and then hash out together at the end of each quarter. The seminar questions correspond to the skills we are working on each grading period. The thing is: the assignment is a lot of work, especially when they leave it until the last minute, which high school students are wont to do. And since ChatGPT hit the scene, I’ve had to rethink the way I assess it since so many of the students simply turn to generative AI that they’ve used to do the heavy lifting of the assignment. I ended up making the assignment worth more points, but I entered it into a weighted category that was not worth as much. I guess I truly was trying to trick the students when in 2023 I re-designed the way I graded the assignment: 100 points in the classwork category (10% weighted). I wanted it to feel like the assignment held weight when in reality, I knew it wouldn’t cause a student’s grade to plummet when I inevitably found them to have used ChatGPT to either fill out the log or respond to the seminar questions, or both. 

This change left students who use the assignment as a learning experience in an unfair position. What if a student struggled all quarter and then felt something click for this assessment? Their grade wouldn’t move an inch even if they earned an A on the assignment. I’ve refused to abandon the podcasts simply because the students who do the project in earnest get so much out of it. Selfishly, I also REALLY enjoy talking to the students about the podcasts they are listening to if they are actually listening to their podcasts. But for every student who gets so much out of the project there is a student who turns in nonsense, or worse, simply cheats. This podcast assignment is a textbook example of me “refusing to kill my darlings,” and now, I find myself looking at not only the way I input the assignment into my gradebook, but also the things I even ask them to do for the assignment. Why was I asking them to answer so many Socratic seminar questions when I really only needed to monitor one or two skills?

Here is what I have come up with: rather than giving the students all quarter to complete the assignment, I do periodic check-ins. Each check-in is worth a small portion of the overall grade. I recently even began asking the students to complete some of the seminar responses in class. Each quarter, I have six Socratic seminar questions, and not every question entails an essential standard for that quarter. The non-essential standard questions are the ones I now ask them to complete in class. For example, asking students to characterize the rhetorical situation or identify instances of bias. These questions are important, yes, but at this point, they are review, and they do not measure the essential components of what I’ve been teaching. These responses will go into the gradebook at 10%. I’ve decided to make only one of the six questions go into the gradebook at 50%: this quarter it is the question that asks them to identify and respond to a claim made in the podcast. This is the crux of what we’ve been working on. This will allow me to read their responses more deeply. This will allow me to more accurately measure learning based on what I’ve been teaching. 

Over the years, I’ve been tempted to throw the baby out with the bath water and just go with that which makes my job easier. If so many students complain about the assignment, if it’s also SO MUCH to grade, why not just scrap the assignment completely? Because my podcast challenge is my BABY. It’s an assignment I conceptualized myself and I love it so much. A few years ago I taught the sibling of a student I taught prior to Covid and the mom cornered me during open house and raved that she and her daughter still listen to NYT’s The Daily together even 5+ years after my class, and they get so much out of it. Why would I want to get rid of something that has clearly benefited students so deeply? I do not want to kill my darlings (out of pride) and I also do not want to throw the baby out with the bath water (because of the students who try and therefore benefit), so why not re-evaluate my own grading practices and make the assignment more equitable?

I realize the solution to my podcast assignment doesn’t fully solve the problem. There will always be capable students who perform poorly in a class because they neglect to turn in assignments. Conversely, we will always have those work horse students whose skills are lacking, but manage to pull off a B in a class. However, taking time to really consider the value of the work I put in front of my students allows me to cut what’s not needed so my students can get more out of the essentials. Further, carefully crafting assignment point totals and category weights so that they reflect the difficulty or workload of any given assignment is simply an equitable grading practice.