Formative Feedback in Large Classes

This issue has proven remarkably resistant to exploration for me so far; I suspect that I have spent too much time attempting to frame my discussion in some kind of context, and so am attempting to do less of that here. If the question is how we might provide formative feedback to a large class, it strikes me that this requires some definition.

For the purpose of this discussion, then, a large class is 125 students enrolled in a first year English course. I assume that the average student in this large class is in their first or second semester at University, that they did reasonably well in their ELA 30-1 or equivalent course, and that they are taking a first year English course because they have to.

Ambrose defines formative feedback as feedback which “explicitly communicates to students about some specific aspects of their performance relative to specific target criteria, and [that] provides information that helps students progress toward meeting those criteria” (139). Formative feedback, then, needs to provide explicit communication about specific aspects of performance.

Generally, the feedback that I provide on written assignments consists of the following:

  1. Copy-editing of the first paragraph or so
  2. short (1-5 words) comments written in the margins
  3. a point-form conclusive comment containing
    1. a summary of the paper’s position
    2. three tips for next time

I try to reinforce what a students has done well with short comments, and in the summary portion, and perhaps also in the three tips, but also make suggestions about one or two things that the student might work on for their next assignment. This is meant to be transferable strategic knowledge.

This grading/feedback takes me approximately 15 minutes to make on a 1,000-ish word paper, so for the class sizes I am used to dealing with (25-35 students), I can expect to spend about ten hours split up into two or three sessions grading a class set of essays. Generally, the criteria on the assignments I grade for my instructors of record is reasonably specific, but Dr. X does not provide rubrics, preferring to suggest that an A paper has a different feel than a B or C paper.

Typically, students receive feedback in the form of 2-3 tests and 2-3 written assignments over the course of a semester, so if I imagine that I’m suddenly teaching 120 students instead of 40, there will be some immediate needs to adapt. (120 students likely take 30-ish hours to grade, which means in a semester I would either need TA’s or have to spend about half of every work-day grading.)

So, I think that one technique worth exploring is to “Look for Patterns of Errors in Student Work” (148). The general idea here is that a large number of students will have consistent patterns of error. Particularly if the same instructor taught the same course over a number of years, there should be enough data to pull out the major problems that students are encountering. (I imagine these to be unclear or absent theses, poor use of quotations, and a reliance on plot recapping.)

Knowing that students will experience these kinds of problems, they need to be provided with practice. Here a scaffolded assignment, asking first for the summary of a critical article, then something relatively short that forces the student to articulate their position in a page or less, then something reflective which requires the student to respond to feedback, before finally submitting the essay should have a number of benefits.

First, the student is forced to work through a writing process rather than cranking out a first draft for submission. Second, the first few stages of the assignment are relatively short, and can likely be graded on very specific criteria allowing for much quicker grading. The comments could also be directed at pointing out where a student can see immediate improvement. (A rubric would be useful in cases like this.) “For the next stage, make sure to emphasize X,” or “you can

Second, because there are a number of smaller assignments, there are a greater number of opportunities to receive feedback. This also allows the instructor to monitor for specific problems and make interventions with the class as a whole. In theory this should also cut down on plagiarism, but that’s another issue.

Providing the feedback over the course of two or three shorter assignments, and making suggestions about how best to revise their material into the next phase of the assignment seems more formative than summative, particularly if there were two staged papers in a semester. It would, perhaps, be possible to grade each of these shorter assignments in a week. Arguably a final essay of better written material should also be faster to grade (or at least more enjoyable to read).

This entry was posted in Uncategorised. Bookmark the permalink.

2 Responses to Formative Feedback in Large Classes

  1. Jon Rozhon says:

    The formative feedback over several assignments is noble and good. But daunting, too, over 125 students. You’re right, Tim, the time commitment would be pretty brutal — likely not reflecting why you got into the teaching business… I wonder if Prof X would agree to have you use a rubric for the shorter assignments? You wouldn’t even have to make many comments, just circle the appropriate areas on the rubric. Prof X needs to know what you now know about the redeeming features of rubrics!

    I like your idea of staging short assignments into a final essay — perhaps you could introduce an element of peer evaluation to complement your own evaluation. So, let’s say you decide to split the essay into three parts: 1) intro paragraph(s), 2) body, and 3) conclusion. You could have a peer-review session for the intro, followed by rewrite/review, then finally your own brief reading. You could do the same with the body of the essay, and then one more time for the conclusion. That way each student has their work looked at 6 times per semester.

    The danger here, I suppose, is possible reinforcement of the old mechanical style of essay construction they learned in high school. What about adding a fourth element, then? One last review where students are asked to look over the whole essay — change the thesis statement to reflect better the ideas developed in the body, figure out why it is sometimes necessary to “kill their darlings”, and learn other more sophisticated aspects of the writing process. Then they get their work looked at 8 times per semester and start to realize that writing is recursive. Forget the 2 to 3 tests if you can (or reduce them to 1 test) — the students just need to pass; nobody has to die.

    I know the peer-review road is not easy, either — students need to have their hands held the first time or two — one can’t have freshmen engaging in ad hominem attacks or over-scrupulous copy editing. They will learn a thing or two about the importance of decorum in class discourse, not to mention the sorts of details they should be looking for in reviewing their own essays. It should get them thinking metacognitively from the get-go.

    I don’t know. It’s tough going down the formative evaluation path for mass classes and sure would require a lot of planning at the outset. Hopefully, the Red Deer folks don’t give you 125 students and it’s a problem you’ll never have to face…

  2. Tim McNeil says:

    Peer review is something I’ve found to be largely unsuccessful in the past, but your suggestion in class about using a student developed rubric for peer evaluation might get enough people in the right kind of headspace to make it worthwhile. Thanks Jon

Leave a Reply

Your email address will not be published. Required fields are marked *