I’m taking a short break from grading team assignments in my current text analysis methods class to consider how I might revise the way I evaluate student work in the future. Grading has always been an unsatisfactory but institutionally necessary practice in my teaching. Of course students need feedback, but much of their educational experience has taught them that this is an opaque and punitive experience rather than a valuable, enlightening, and supportive one. Although I approach assessment and assignments from a continuous improvement mindset, it is very hard to convince students that their experience with evaluation will be different in my class when the mechanism remains the same. This post is a reflection on my past and present grading habits and shares ideas for restructuring future classes to create a better assessment experience for everyone.
I know I’m not alone, as many instructors have written on the evils, or shall we say, negative consequences, of traditional grading and have posed alternatives. What works in and for one type of class at one type of institution, however, does not necessarily work for all. Our own individual needs, desires, and constraints as instructors must also be taken into consideration. Given my own time constraints at the moment and the necessity of finishing assessment tasks today, I will leave the literature review for another day and simply reflect on my own experiences and ideas.
Due to my own workload, I have had little time to ponder or change my grading habits, and my classes tend to be so large that “ungrading” and the exchange of ideas between instructor and student that I’d like to employ is just not feasible on a regular basis with 35-40 students in my seminars and 120 students in the much larger introductory class, especially with the fast pace of 10-week quarters. This leaves me with seemingly few choices, especially at a research university where my research and publications must take precedence to retain my position.
Out of desperation after getting quite behind in grading due to illness and other professional obligations, I graded one assignment in consultation with each project team, taking class time to meet with each in turn, discuss the assignment, provide verbal feedback and written notes, and assign a grade to the work. This is the way I would prefer to assess student work and offer suggestions for improvement (preferably, without attaching a grade), but in a three-hour class, it took two hours to meet with everyone in a nearly 40-student class, and is thus unsustainable. However, it saved me about six hours of additional time I would have spent providing written feedback alone on all of the assignments.
Every class I teach asks students to conduct original research with the methods we explore. In terms of authentic assessment, rigorous learning objectives, and deep engagement with course content, this is great. This also means that every project is unique, students are working on writing skills in addition to learning new computational research methods, understanding the context of their topic, and, especially in the introductory class, experiencing the ambiguous and often befuddling nature of research for the first time. Although I assess some of these learning objectives separately on smaller individual scaffolded assignments (which take even more time to grade), the projects themselves also require a great deal of time and energy to evaluate, even with rubrics.
One way I would to provide more frequent feedback on low-stakes assessments is to create weekly quizzes for few points that are auto-graded and provide detailed information on the solutions/answers to each of the questions. These take time to design well. Now that my courses are largely built and will remain consistent over the next few years, though, it seems like it would be well worth the effort.
Nevertheless, I’m still left with the conundrum of providing helpful feedback on individual and team written assignments in a timely manner without the task becoming a black hole that drains all of my time and energy. A grader is one solution, but an unsatisfactory one based on past experience. Even when the grading has proceeded smoothly, I’m farther removed from what students are working on so it becomes more difficult to provide detailed guidance on their projects since each is unique.
Ultimately, the solution would appear to lie in capping the classes at the original 25-seat maximum, forcing the institution to hire more instructors for the program, and creating a better learning experience for those in my classes. With fewer students, I can spend more time with each individual and team while also reducing my workload to a manageable level. With fewer students, I could implement the continuous improvement model I prefer, rather than a more rigid grading structure. In the desired model, students would receive feedback on whether their work meets the standards, goes beyond, or requires improvement and have the opportunity to revise and resubmit. I already employ a version of this for students who are struggling, but it would be nice to offer this chance to everyone. An additional advantage of the continuous improvement model is that it is distinct from the more familiar grading model, and in this way would more clearly communicate my own stance on assessment and firm belief that every student can succeed.