More Adventures in Ungrading

24Jan22

As my blog has become more intermittent over the past few years, one topic seems to still get lots of traffic: rethinking grading. I first started experimenting with grading (and writing about it) around five years ago, and I’m proud to say that I have not “graded” an assignment since! But the ways that I’ve practiced ungrading has changed a fair amount and warrants revisiting.

Back in 2016, my approach to ungrading was focused on specifications grading, the system of evaluating every assignment as Satisfactory or Unsatisfactory in meeting the assignment’s outlined specifications. I have continued using that system for my introductory Television & American Culture course, where a comparatively high number of students (around 35 each semester, which is “large” for Middlebury) and a large amount of content to cover makes it (mostly) effective. During COVID times, I shifted the course to include more remote instruction, including recorded lecture videos (off-set by reducing the length of in-person scheduled meetings) and numerous asynchronous “engagement activities” that students can choose from to demonstrate their ongoing engagement with the material. Even as I taught the course fully in-person in Fall 2021, I retained a lot of the online components to allow more flexibility for when students may have to miss class due to illness (which happened regularly) or otherwise had trouble keeping up with the regular workload during times of high stress and uncertainty—my guiding values in course design is to prioritize student agency, flexibility, and transparency, and this system emerged from that foundation.

However, I’m planning on rethinking some of this course’s structure the next time I teach it in Spring 2023. Some students seemed to treat the flexible micro-assignments too much as “boxes to check” on their bingo cards rather than opportunities to engage in the material in thoughtful ways—even with low-stakes Satisfactory/Unsatisfactory grading, many were still driven more by the extrinsic motivation to accumulate points rather than to dig into the topics, and I spent too much time replying to messages about points and deadlines, rather than ideas . So I’ll likely reduce the number of engagement activities and design them to be more meaningful and… “engaging.” I also need to shift the timing for the final essay so that it doesn’t fall amidst the crunch of finals, leading too many students to under-perform and sacrifice opportunities for feedback and revision. And I am going to read more about labor-based grading to see if I can adapt some of those approaches to suit this course design. But I’m certainly not going back to grading assignments!

While specifications grading has remained a mostly effective system for my introductory course, I found that it did not work in my other classes. In 2016, I wrote about redesigning one of my advanced seminars with specifications grading, and I realized that I never followed-up to reflect on how it worked. In short, it didn’t. The model of numerous small assignments grouped into tiered bundles for final grades doesn’t really work for an advanced writing-intensive seminar, and the effective pedagogy within the course around both theoretical content and writing seemed completely detached from the regimented counting of assignments and outcomes. So I knew that I needed another way to determine grades for my seminars, which is most of my teaching outside of the introductory course.

Thus the next time I taught the course in Fall 2019, I fully embraced the ungrading model that I had been reading about, largely from Jesse Stommel and later in Susan Blum’s book, as well as hearing from colleagues who had similarly abandon traditional grading. Spoiler: it was a huge success, and I’ve adopted it for every other seminar I’ve taught since. Here’s the explanation of ungrading from that first syllabus, which I’ve left mostly intact in future semesters:

This course uses an unconventional approach for assessing student learning roughly termed “ungrading.” You will not receive a “grade” for any single assignment, with only a final course grade registered into Banner [our student record system]. While Professor Mittell will register that grade, he will not assign it—you will. Such self-grading means that students are fully responsible for their own learning, and it is meant to fully sever the link between that learning and the “outcome” of grades. This grade will emerge through ongoing conversations between each student and Professor Mittell; while he reserves the right to alter the grade that a student assigns, it is a sign of mutual trust and shared responsibility for learning that he does not anticipate doing so.

Even though there will not be grades, there will be lots of feedback, evaluation, assessment, and revision—these will all hopefully be channeled toward maximizing learning. Students will create an individual learning plan, write self-reflections on their learning, meet with both peers and Professor Mittell to discuss their progress, and undertake revisions based on feedback. Since all students who pass the course will have achieved the goals for College Writing, the expectations for success are quite high. In exchange for students’ hard work, Professor Mittell agrees to take however much time is needed to ensure students understand expectations and practices, and are poised to succeed to their desired goals. His goal is to help each student achieve their learning goals, and to be transparent about expectations for learning throughout the semester.

Students were a bit skeptical of this approach, as having that much agency and flexibility was quite novel, and many thought there must be a catch—given that each of them arrived at an elite college via a system where maximizing grades is seen as the ends, not the means, this approach produced some serious culture shock. But after giving a brief overview in the first week, I stopped talking about grades and focused on the material and writing assignments (another drawback of the specifications grading approach is that since it is complicated and unconventional, you need to spend a good deal of time explaining and reviewing it, as students want to “get it right” – even as I try to decenter grades, I end up discussing them a lot). In this ungraded seminar, whenever students asked about grades, I simply said, “focus on the material and we’ll discuss your grade at the end of the semester.”

And that’s what we did. Students did a brief assignment for the first week, writing a “Statement of Learning Intentions”: they were asked to read through the syllabus and submit a short reflection “that outlines what they hope to learn through this course, and how they hope to accomplish those goals.” Then at the end of the semester, they submitted another reflection essay, looking back at their initial statement, reviewing the course learning goals, and reflecting on their own engagement throughout the course. These two documents became the basis for individual conferences that I had with each student to conclude the semester, where we discussed what they learned in depth and dived into their writing projects. Each conference concluded with me asking, “So, what grade do you think best reflects your learning?”, as guided by these learning goals from the syllabus:

The course design is based around a series of core learning goals, assembled in a hierarchy of sophistication. Students will highlight their own learning goals from this list, as well as devise their own. These are roughly grouped in tiers that correspond to expected grade levels, with each student expected to reflect their particular goals via written and conversational reflection.

All students who pass the course (C) will demonstrate the ability to:

– Describe how various theoretical approaches approach the study of popular culture

– Apply specific vocabulary and concepts to examine popular culture

– Read dense theoretical writings and summarize their core ideas

– Communicate their ideas orally and via writing with fluency and clarity, per college CW standards

– Revise their writing to improve both ideas and communication, per college CW standards

Students who achieve a higher level of accomplishment (B) will also demonstrate the ability to:

– Analyze popular culture with original insights, effective use of sources, and connections to theoretical models, different examples and cultural contexts

– Engage in serious conversation about often fraught topics with an ethos of mutual respect and generosity

Students who achieve the highest level of accomplishment (A) will also demonstrate the ability to:

– Create, substantiate, and communicate an original analytic argument that synthesizes multiple facets of popular culture, appropriate types of evidence, and theoretical approaches with sophistication

– Meet class expectations per the assigned schedule with consistency, and provide strong support to peers to facilitate our learning community

These conferences proved to be one of my favorite pedagogical activities I’ve ever done, providing real closure on the semester and a chance for substantive conversation about a student’s learning. We typically get fully engaged in the course content and student writing that when I bring up the inevitable, it feels like a disruption: students don’t really want to talk about the grade either! They typically hem and haw about needing to sum everything up with a letter that feels so reductive (welcome to the worst part of grading!), and then tentatively present a grade, or often a range of two options. We go back and forth a little, but usually I enter the grade they suggested into my spreadsheet, and they depart with a real sense of closure.

This approach to ungrading appeared to be quite successful to me and seemingly to my students as well, and I have adopted it for all of my other courses since: Videographic Film & Media Studies, Key Concepts in Film & Media Criticism, and my department’s senior tutorial that mentors senior thesis projects. It proved to work well with both in-person and teleconferenced meetings, and has helped all of these classes focus on learning rather than assessment. These closing conferences are really the highlight of the semester, as I come away impressed with most students’ learning and their own articulation of their engagement with the material. I’ve found that students are typically honest about their struggles and successes, and I can usually help emphasize the latter in a way that leaves them feeling good about the course.

Whenever I discuss this ungrading approach with colleagues, they usually want to talk about what grades students give themselves. Honestly, this is the least interesting part of the approach to me—I want to focus on how it facilitates learning, builds community, and encourages reflection. But since academia has put so much stock in the outcome of grades being the marker of “rigor” and student success, that ends up being the focus for many. So far, the grade distribution in my ungraded seminars closely matches the breakdown from my traditionally graded days, with a bit of a shift upward but still a solid range of grades from A to B– (in these classes, Cs or lower have always been quite rare, due to students not completing significant work). My seminar this past fall had significantly higher grades, but that was an indication of a great group of students who did excellent work (more about that work in a forthcoming post)—this is a clear case of “learning inflation,” not “grade inflation”!

Before trying this approach, I’d read that self-assignment leads to students often suggesting lower grades than faculty feel are appropriate, and that has been my general experience. I go into each meeting with a ballpark sense of what grade feels most appropriate to me, and sometimes our reflective conversation shifts that sense (almost always upward, as students can demonstrate deep engagement via these conversations); students’ self-grades are almost always at or slightly below my assessed level. When students undershoot my assessment, I can articulate their strengths, hopefully helping students value their own work more, and acknowledging that I recognize their efforts—since such low assessments are often from students who lack confidence or struggle with extraneous factors, such conversations feel impactful.

The ungrading literature also commonly mentions that women and students of color can undersell their accomplishments, while white guys tend to self-inflate. In my very limited sample size, I have not seen that. My most memorable moment around this was in a conference with a white female student who had done solid but unexceptional work throughout the semester, and I’d come to our meeting with a B+ in mind. In our conversation, she asserted that she thought she’d earned an A; when I offered a little hesitation, she proceeded to go through the learning goals to highlight her hard work, her growth, and her progress—and I was convinced that her learning process earned an A, even if her essays were not As in my conventional rubric. Such self-advocacy and reflection seems like an important outcome for students whose perspectives are structurally marginalized in academia, and I do what I can to acknowledge such contexts and encourage those students.

Through my ungrading practices over the past two years, I have tried to let go of the conventional rubrics I bring to these conferences. I try not to think of the grade as the culmination of my work with the student that semester, but rather a reporting requirement that my institution imposes on that work. I don’t approach these conversations to police the grades students assign themselves—if I come to a conversation with a pre-assessed grade of B, and a student makes the case for a B+, I enter the B+ into the spreadsheet. One of the byproducts of my turn to ungrading is that I’ve come to an important realization: this does not matter to me. I have absolutely no investment in keeping the class GPA below some arbitrary number, nor making sure that student grades somehow are “true” to an abstract standard. I want the grade to indicate the student’s sense of their own learning, but once it’s entered into the college’s system, that letter means nothing to me. What matters is their learning and our relationship, and I’ve tried to make these letters get out of the way of what really counts.

Thus I plan to continue this ungrading approach for all subsequent smaller courses, including a first-year seminar in Fall 2022—it should be quite interesting to see how students take to ungrading in their very first college course. I’ve thought about adopting ungrading for my large intro course, but I’m not convinced that I can do it in a way that maintains what works best—the intensive feedback, ongoing conversations, and final conferences—while ensuring that students remain engaged through the amount of material that the course covers. I have a colleague who ungrades his intro course, and I know other faculty like Susan Blum do the same for even larger courses. So maybe that will be the next step on my journey…

If you’re an academic who’s made it this far and are intrigued, I highly recommend reading more from Jesse Stommel and see how this approach might fit with your own courses. I hope to see more fellow travelers on this journey away from grading!

Filed under: Academia, Middlebury, Teaching   |  4 Comments
Tags: specifications grading, ungrading