In this excerpt from How To Ungrade used with permission of its author, Jesse Stommel provocatively suggests alternative assessment strategies that leave more room for dialogue between students and teachers than numerical grading. Stommel writes from a higher education context but the examples can be very useful in K-12 spaces. Edcampers, we suggest you check out these examples and think about how they can be applied to your classroom.
The work of teaching shouldn't be reduced to the mechanical act of grading or marking. Our talk of grading shouldn't be reduced to our complaining about the continuing necessity of it.
But, if grades are going to control so many of the conversations we have in education, at the very least can we be more creative in how we approach them?
Alternate Approaches to Assessment
As I've said, there is no single approach that will work for every teacher, with every student, or at every institution. I find that ungrading works better when a teacher feels they can fully own their pedagogical approach (which requires administrators and institutions to defend the academic freedom of teachers, especially adjuncts). There are lots of different possible paths toward ungrading, and smaller experiments can be just as fruitful as larger ones.
Grade Free Zones
Sometimes, it's hard to imagine diving right into the deep end of ungrading, so try having the first 1/3 of the term be ungraded, a sandbox for students to experiment inside before moving on to the more formal activities of a course. Or decide to grade only a few major assignments.
I've already talked at length about how I use self-assessment. What I'll add is that this work is both part of my approach to the problem of grades and also an end goal in and of itself. Ann Berthoff writes “Learning to look carefully, to see what you're looking at, is perenially acclaimed as the essential skill for both artist and scientist.” Metacognition is a practical skill that cuts across disciplines.
If you're only grading a few assignments, you may not feel like you have enough information to determine a final grade at the end of a course. I have students write process letters, describing their learning and how their work evolves over the term. This can be text, including (or linking to) representative examples of work they don't otherwise turn in. You might also ask students to take pictures of their work as it evolves, add voiceover to a screencast, or I've had students shoot video for a film documenting their learning (a sort of behind the scenes reel for the class).
In “Grading Student Writing: Making it Simpler, Fairer, Clearer,” Peter Elbow describes what he calls “minimal grading,” using a scale with only one, two, or three levels instead of giving students (admittedly bizarre) grades like 97%, 18/20, or A-/B+. Scales like these with too many gradations make it difficult for teachers to determine grades and even more difficult for students to interpret them. Elbow recommends scales with fewer gradations: [one] turned in, [two] pass/fail, [three] strong/satisfactory/weak. He also describes a “zero scale,” in which some work is assigned but not collected at all. This frees the teacher from feeling they have to respond to, evaluate, or even read every bit of work students do. And this last, moving away from student work as a thing to be “collected,” might actually prove best at creating intrinsic motivation to do the work of a course.
In my film courses, I often ask students to organize a film festival or premiere in order to share their work for the class. These usually include talk-backs with the audience. Increasingly, I don't ask students to turn any assignments into me (aside from their self-reflections). The community of the class (including me but not just me) becomes their audience. In a service learning course, this community expands even further beyond the boundaries of the class. In short, how can we create reasons more meaningful than points for students to do the work of a course?
Grading contracts convey expectations about what is required for each potential grade. In “A Grade-less Writing Course that Focuses on Labor and Assessing,” Asao B. Inoue argues for “calculating course grades by labor completed and [dispensing] almost completely with judgments of quality.” Students work toward the grade they want to achieve, and goal posts don't unexpectedly shift. These contracts can also be negotiated with the class. Either way, contract grading pushes against the relegating of people into categories, “A student,” “B student,” by keeping the focus on the work rather than the student. They can be humane in a way standardized teacher-centered rubrics usually are not. Contracts run the risk of centering grades even more than traditional grading, but at their best, the negotiating around the contract becomes a way for a group of students to collectively worry the edges of grading as a system.
Increasingly, many corporate e-portfolio platforms are walled gardens, giving students a regimented way of gathering together their work for the purposes of assessment. I prefer more authentic portfolios that have use value beyond the needs of individual, course, programatic, or institutional assessment. University of Mary Washington's Domain of One's Own project offers a space for students to do this kind of work, choosing a domain name, carving out space for themselves on the Web, and crafting a digital identity that exists outside (but also in conversation with) their coursework. It's a metacognitive space, but also one with immediate practical value (as a way to share their work with potential collaborators, employers, grad. schools, etc.).
Peer-assessment can be formal (having students evaluate each other's work) or informal (just having students actively engage with one another's work). It can be particularly useful when having students work in large groups. I frequently have students work on projects that have an entire class (of 25 or more) collaborating. When I do this, I ask every student to write a process letter that addresses their own contributions as well as the functionality and dynamic of the team they're working with. With large group projects, it's hard for me to see what and how each student contributes, but these letters help me get a view into a process I might not otherwise be able to see. If it's a project students work on across the entire term, doing multiple process letters also allows me to get the information I need to step in and help when and where I'm needed.
I'm really not a fan of rubrics. Alfie Kohn, in “The Trouble with Rubrics,” describes them as an “attempt to deny the subjectivity of human judgment.” Rubrics are often recommended as a way to make standards for evaluation transparent, but for me, a 5x5 grid filled with copious text is bewildering and inscrutable. Rubrics have never helped me make sense of grading or being graded. Peter Elbow encourages making rubrics plainer and more direct, a 3x3 or smaller grid. The rubrics I find most exciting are ones crafted by students—so that the making of the rubric becomes an act of learning itself rather than a mechanism (or set of assumptions) created entirely in advance of students arriving to a course.
Jesse Stommel is Executive Director of the Division of Teaching and Learning Technologies at University of Mary Washington. He is also Co-founder of Digital Pedagogy Lab and Hybrid Pedagogy: an open-access journal of learning, teaching, and technology. He’s online at jessestommel.com and on Twitter @Jessifer.