running a grade-norming session
Students often suspect that their writing will be graded subjectively; they assume that each instructor will approach a stack of drafts with a different menu of preferences. Given this, student writers reason that they will do well on a paper if (and only if) they can divine an individual instructor's fetishes before turning a paper over to her or him. Little do these students suspect that their instructors often worry that they do in fact grade subjectively. Setting out explicit criteria in grading rubrics that can be distributed with assignments can be extremely useful in keeping grading practices tied to an assignment's stated objectives, and in keeping fatigued graders from responding unevenly (subjectively) to the work. Still, even armed with grading rubrics, instructors might be uncomfortable, worrying that their grading practices are different from those of other instructors who teach the same or similar courses. In these situations, taking an hour or two to sit down as a group, grade a sampling of student writing, and then discuss the similarities and differences in these grades, can be helpful. I am happy to run these sessions with you. If you choose to organize and run one yourself, here are some procedural suggestions:
- Select a specific assignment that all participants are familiar with (or can quickly become familiar with). Make enough copies of the assignment to distribute to all participating instructors.
- If a grading rubric or guide is not included with the assignment, either create one in advance, or plan to negotiate one in the group.
- Procure at least two sample final drafts. Delete all identifying information about the student, instructor, and course, and make enough sets to distribute to each participant.
- Take a few minutes to ensure that the model assignment and rubric are understood. Note clarification questions, as they might be shared by students!
- Participants take time to grade and comment on the samples, either one at a time, or all at once. Allow 10 minutes or so per sample.
- Before launching into a discussion, survey all participants about the grades they have assigned so that an overall impression about the similarities and differences can be achieved prior to deliberations.
- A desired outcome of these discussions is that all instructors come to a sense of agreement about the correlation between the grading grid and the student writing. If the graders are all over the map, it could be that they are not sticking to the rubric, and it could be that the rubric is not adequately clear. Often students and instructors might understand an assignment's expectations differently, and it is important to determine the causes of these differences. Perhaps the assignment's description lacked sufficient clarity? Perhaps the grading rubric contained generalized objectives, such as "clear, insightful prose…"? Perhaps participating instructors are swayed by preferences that are not outlined on the assignment?