Contract grading, Part 1

I first came across the phrase “contract grading” in some Twitter feed or another (I really should write down which feeds I get things from, so I can give proper credit!). I couldn’t for the life of me figure out what contract grading could be–surely it didn’t mean that one would sign a contract with a student, promising to give them a certain grade, right?

Wrong. That’s exactly what it means. Though, of course, the student has to hold up their end of the bargain too.

I’ve looked into the idea a bit more, and I’m still ambivalent about it, though intrigued.

The Bok Blog (Harvard) starts off a post on contract grading this way:

The central feature of contract grading is the contract: a clear and detailed set of guidelines that stipulate exactly what a student needs to do in order to earn each possible grade.

Surely that’s not all there is to it, of course, since that’s what a syllabus gives (this point is also made in a post by Billie Hara on contract grading over at The ProfHacker blog, from The Chronicle of Higher Education). The first versions of contract grading I have come across, though, do resemble sections of syllabi more than contracts.

Adding up points

The Bok Blog explains one version of contract grading (from Gerald Herman, a professor in History at Northeastern University) that does set the idea apart from what many instructors already do: students can choose how many assignments to complete out of a range of different assignments. Each one is worth a certain number of points, and to get an “A” students have to get a total that could be reached by doing very well on a few assignments, or less well on more assignments. I don’t see anything in the blog post about students and professor agreeing on some contract, though.

The main differences between this version and what I normally do seem to be:

  1. Students could earn an “A” in this sort of contract by only completing a couple of assignments, potentially, and then not finishing the rest of the course (though according to the Bok Blog post, Prof. Herman says many students do so even if they don’t have to).
  2. Students in my courses who did “B” work all the way through would get a B, but students in this version of contract grading might be able to add up their points at the end to earn an “A.”
  3. Students could do quite badly on one assignment and it wouldn’t necessarily hurt them if they did quite well on others, because what matters is the total points and they could make those up if there were enough assignments to do so.

One would still need to mark essays and exams for the quality of work to decide how many points to assign, but one wouldn’t be assigning letter grades to those (unless the prof and/or students were thinking of the points for each as being associated with letter marks).

I like #3, in that I hate to see students end up with a lowered mark for one thing that they did badly, especially if the rest of their work is quite good. I can live with #1, since I suppose one could argue that if students are already able to do well what you are trying to help them learn how to do, then it might not be worth it for them to continue doing more of it.

I’m still thinking about #2, which somehow still bothers me. And I have to think further about what I think a course grade is supposed to show, what it’s purpose is, before I can decide whether it should bother me. And to complicate matters, there are a number of people arguing for getting rid of grades altogether, an idea that I think might have merit, but I need to look into it more (which I will do here in the blog, later).

#2 gets to the heart of contract grading from what I’ve read so far: it seems that in some versions anyway, contract grading focuses more on what you’ve done rather than how well you’ve done it.

Earning a B

Peter Elbow and Jane Danielewicz put this point a bit differently, in “A Unilateral Grading Contract to Improve Learning and Teaching,” College Composition and Communication 61(2) (December 2009): 244-268:

The basic principle in contract grading is simple but radical: what counts (“counts,” literally, for the grade) is going through the motions. That is, contract grading focuses wholeheartedly on processes whereas conventional grading focuses much more on products, outcomes, or results. (260)

Billie Hara gives an overview of this article in the ProfHacker post noted above (the article itself is available on Elbow’s web site). In it, Elbow and Danielewicz discuss a grading contract for writing courses, in which students can earn a grade of “B” on the basis of completing a list of tasks, but if they want to earn an “A” then the quality of their writing comes into play. They provide a sample grading contract, listing all the things students must do to earn a B–such as attending a certain number of classes, completing writing assignments according to criteria, providing substantive revisions, participating in class and group activities/discussion, attending conferences with the teacher, and more (these are all listed on the ProfHacker post noted above). After providing this list, the sample contract given in the article states:

Thus you earn the grade of B entirely on the basis of what you do—on your conscientious effort and participation. The grade of B does not derive from my judgment about the quality of your writing. Grades higher than B, however, do rest on my judgment of writing quality. To earn higher grades you must produce writing—particularly for your final portfolio—that I judge to be of exceptionally high quality. (246)

Billie Hara from ProfHacker notes that some of the things on the list needed for a B are not simply a matter of completion, but have qualitative elements (“give thoughtful peer feedback during class workshops and work faithfully with your group on other collaborative tasks (e.g., sharing papers, commenting on drafts, peer editing, on-line discussion boards, answering peer questions)”). Elbow and Danielewicz argue that even with these “fuzzy criteria” as they call them, they aren’t making fine-grained distinctions. They only say students have violated them if the performance obviously falls short (e.g., a draft that isn’t long enough), and if students disagree with their judgment, they’ll “take the student’s word for it” (251).

What Elbow and Danielewicz emphasize as valuable about their system is decoupling grades from evaluative feedback, and from the process of writing and rewriting. Giving grades can be problematic from both the instructor’s and student’s perspective. For instructors, we can agonize over the difference between a B and B-, while recognizing that single grades don’t adequately express the quality of various pieces of writing anyway. For students, receiving grades lower than one wants can be debilitating, and the emotional impact can take away from actually taking the feedback seriously and using it to help improve later work. Elbow and Danielewicz also note that when feedback is connected to a grade, students can see the feedback as more authoritative and just take it as something they must use later to earn the grade they want, instead of deciding what they agree with and what they think isn’t quite right (254-255).

Elbow and Danielewicz say they use just three marks throughout the term:

not satisfactory for B, satisfactory for B, and better than B. We don’t distinguish among grades higher than B until the end of the semester, when we have student portfolios in hand. (246)

With this system, they say, they “ignore quality of writing for grades up to B—but focus explicitly on writing quality for higher grades” (246). Of course, they do have to take quality into account just to determine these three marks, but yes, it does mean one doesn’t have to spend time wondering if a paper should be given a B or a B+ (or worse, with numerical systems, whether the mark should be 76 or 77 or 78).

I find this idea very intriguing. The part I dislike the most about evaluating student work is giving a mark. I am happy to give feedback, but when it comes to giving a mark I get anxious, wondering if I’m being as fair as possible, wondering whether the mark will negatively affect the student and discourage them (but recognizing I must give the mark I think the work deserves anyway), etc.

What is really different from “traditional” marking in this system?

  • Students receive only three marks throughout the term: not at B-level, at B-level, and above B-level. But when you think about it, is this terribly different from: C or D, B, and A? Of course, at least one doesn’t have to go into gradations within those, and doesn’t have to decide between C or D. And maybe the emotional effect of “not satisfactory for B” is less than the emotional effect of “C.”
  • Quality of writing isn’t taken into account until considering grades higher than B. Still, that isn’t fully the case, since one has the three gradations noted above. For things like attendance and participation in group discussion, one could conceivably think of these as mostly a matter of completion, but not for the writing assignments if the above three marks are given out.
  • The focus is more on feedback than on grades–if this really works for students, I think it’s a great benefit.
  • It could send the message that an A grade is reserved for excellent, exceptional work, rather than being a matter of simply doing everything required (Elbow and Danielewicz, 251). I like this aspect too, as this sort of message is something I’m continually trying to emphasize in my own courses.

Contract? grading

In both of these versions, however, I’m not seeing a full sense of a “contract”–they look mostly like just different ways of writing a syllabus. Rather than saying one’s marks will be added up into a total (with an A not being possible unless one completes all the assignments) or that each mark will be counted as a certain percentage to reach 100% as the final grade, one gives students an explanation of how they can add up points to achieve different grades, or gives a list of things one needs to do earn a “B.” Elbow and Danielewicz, in their 2009 article, note that they call their system a “contract” because:

… we want to give students written evidence that we contract ourselves to keep this unusual promise to award a B for doing things rather than for writing quality. (247)

But isn’t that what a syllabus is supposed to be? I certainly think of it as a contract between myself and the students, saying what I have agreed to do and what they need to do to get a certain grade. So in a way, I’m contracting with them that if they do certain things, and if their work is of a certain quality, they are guaranteed a certain grade for the course. The difference seems to be in how those grades are calculated, and how much weight is put on the quality of work vs. completing certain tasks at some satisfactory level.


Other contract grading schemes seem to be more like contracts. I’ll address a couple of those in the next post.


One comment

Comments are closed.