If you’ll recall, this semester for my New Semester’s Resolution, I was trying a more collaborative approach to grading that requires students to set goals by modifying my provided rubric and then evaluate their own work according to that rubric, so that they self-grade their assignments. At this point, students have submitted their first self-evaluation and I can give a preliminary report on how that’s going.
Long story short, it’s going great!
The slightly longer, more nuanced version of the story is still overwhelmingly positive, but acknowledges that there has been some confusion and, as expected, a somewhat increased workload.
First, students were confused by the goal-setting stage of the assignment. Students are so used to external authority evaluating them that the very idea of the goal-setting stage seemed strange and suspicious. As I had not used this method before, I didn’t have a good student example of the goal-setting assignment with its rubric, and as a result the initial documents were all over hte place, and many students failed to make clear rubric requirements for themselves. This resulted in confusion later when they had to use the rubric to grade themselves, since having an unclear rubric makes generating a grade harder. However, I think the later assignments will go much smoother having had this struggle once already.
I haven’t graded all of the self-assessments yet, but generally I’m finding that students are doing honest assessments. Some reported that they don’t love the grade they gave themselves, but that they felt it was the fair grade for the work they did. This kind of honest self-evaluation is partly facilitated by my assurance that the grade they give themselves isn’t necessarily their final grade on the project, but that I use it as guidance to assign that final grade. In most cases, I’m either using their suggested grade or bumping it up a little where I feel they’ve been too harsh on themselves.
That second layer of review seems to be important to building the trust necessary for this self-grading to work. I’m not saying I dn’t trust their assessments; I’m saying that I’m checking their work and advocating for them.
Likewise, the fact that they have to explain the grades they give themselves matters a lot. Although the depth of reflection of course varies, I have not seen a case where a student was unable to provide persuasive reasons for the grades they assigned themseslves. While most of the grades are B or A level grades, that’s not unusual for this course and workload, so I don’t think it’s producing any artificial inflation. Instead, it is making me more confident that the grade I assign will seem fair to the student, rather than arbitrary or punitive, since we have effectively negotiated it in advance. Indeed, I haven’t seen a single student assign a 100% grade to themselves yet.
I’ve been able to keep the grading queue managable this semester, despite having far too many students this semester, and I think this strategy for grading is contributing to that success. A large part of the problem I’ve had with grading was a mental burden, a conditioned response that clicking a grade might produce outrage in a student because they don’t understand the grade the same way I do, no matter how much I explain and contextualize in comments or in class. But here I’m mostly rubber-stamping what the students have already said about their work and gently redirecting as needed.
In short, this has made the process of evaluating student work much less painful. If I continue to use this method in future semesters, there will be a lot of changes in how I scaffold it and present the tasks to the students, but overall the process is working. So what is the process?
- Students assess where they’re at in the course assignment sequence and add custom criteria to the rubric I provide that includes the basic criteria for the assignment
- I give some feedback on that document, recommending revisions to the rubric if necessary
- Students complete the project
- I give discursive feedback on the project, but do not grade it
- Students evaluate their own project by engaging with my feedback and then using the rubric to assign themselves a grade. They write a brief reflection justifying that grade
- I read their reflection and assign the whole project sequence a grade based on the grade they give themselves and my understanding of the rubric.
That last step is still a little vague to me and the whole thing is rather messy. I would love some revisions and suggestions to make it clearer, especially from a student perspective.
Still, despite the mess, it seems to be working, so I’m pretty pleased.