Against letter grades

Next semester, no quantitative grades until the end of the semester. No As, no Fs, no 83%. At least one study has shown that grades not only do not help students, they actually impede their performance (Butler, 1987). Students tend to take a horse-race approach to grading, and pay less attention to how they are doing and more to whether their grade on one assignment has gone up or down in comparison with the previous assignment, or various satisficing strategies for achieving whatever they have set as their minimum acceptable grade.

I am already using self-assessment in all of my classes. Rather than giving letter grades, I will list the strengths and weaknesses of each student’s work, and leave the assessment–in terms of grade–until the very end of the semester. Of course, I would prefer to go all Evergreen, and have narrative grades make up the final grade in the class, but I don’t think this can happen at my university.

I have toyed with another possibility, which is not assigning a grade, but rather force-ranking students and revealing to them where they land on that ranking. As a matter of practice, in large undergraduate classes, I have often force-ranked assignments, in order to make sure that all my Bs were grouped together, all Ds together, etc., and that I hadn’t somehow mis-evaluated a project. Moreover, students can estimate their ranking when they see the histogram (when I provide one, which is rare). And at some essential level, this is what we are doing when we grade: my grades at QU are essentially a comparison of work I’ve seen at QU. It would be unfair to compare them to, for example, students in a top doctoral program, or students in 9th grade.

Nonetheless, I can’t decide whether the dire knowledge that your assignment was the worst in the class (or the third worst, or whatever) would be so depressing that you would just give up, or if it would spur you to get off the bottom of the list. According to an article that appeared in the Chronicle, a couple of universities have toyed with class rank and similar measures as alternatives to the all-mighty GPA, but these have generally fallen through.

I suspect, however, that class rank on assignments frankly would not change much from assignment to assignment, even if grades varied somewhat more. I wonder whether the stability of this rank over time would lead to extreme competition among students. My experience has been that under conditions of such competition (think law school) students actually tend to band together, and that may not be a bad thing.

In any case, I’m trying a structured non-grading approach to my courses next semester. If you’ve had success or failure in doing this, I’d be interested in your feedback.

Butler, R. (1987). Task-involving and ego-involving properties of evaluation: Effects of different feedback conditions on motivational perceptions, interest and performance. Journal of Educational Psychology, 79(4), 474-482.

This entry was posted in Teaching and tagged , . Bookmark the permalink. Post a comment or leave a trackback: Trackback URL.


  1. Posted 12/22/2008 at 12:36 am | Permalink

    This probably gets at the heart of what you’ve just said, but I find it aggravating when students are more focused on grades than the material in the course. I’m a grad student in a department where undergrads have to apply to be upper-division majors, but it seems like the greatest disappointment comes from students who are far from borderline – they just want confirmation that they’re putting in the requisite amount of effort for an A.

  2. Mary Halavais
    Posted 12/22/2008 at 12:32 pm | Permalink

    University of California, Santa Cruz, of course, used narrative grading. I recently sat down with a re-entry student for advising, and so have actually seen a UCSC transcript. I also have a friend who graduated from a private university’s “experimental” college back in the good old days (um – the sixties), and then faced the task of explaining his transcripts to some bemused folks in graduate admissions at The University of Chicago.

    For spring, I will be planning fewer graded assignments (reviews, essays, research papers, examinations) and will be including “participation” as a significant component of the grade. (Actually, it’s always between 10-15% for my classes; I’m just going to make it a higher percentage in the spring.)

    So, how will I measure “participation” in a general education class of 48-60? (Hmmm – thought we were a “small university,” did you? The GE classes are telling…) I think I will learn the students’ names REALLY quickly this semester. Wish me luck-M.

  3. Posted 12/22/2008 at 3:02 pm | Permalink

    McChris: Yes, absolutely. And it’s hard–when they are so grade motivated–not to use grades as motivation. But I’m trying to get out of that cycle.

    Mom: One of the faculty in business at UW, I think, used to have students all give him flash cards with a head shot on one side and their name on the back side, and then carried them around and quizzed himself on the names. He promised an A in the course to anyone whom he couldn’t name by the end of the semester, but never had to deliver on that.

    I have to say, although I almost always have a participation component in my courses, it’s really hard to assign a grade to it. Generally, you have a mass of people who you have very little impression of–the ones you remember are those who were most active or who were *completely* silent. While exam grades are generally normally distributed (it always surprises me how they fit to that curve in large classes), participation is almost necessarily a power-law distribution of some sort: 80% are just not going to be active participants.

    And then there is the question of what “active participation” might mean. I think that most of the people who talk a lot generally have something good to say, but we’ve all had people in our classes who talk a lot and have nothing worthwhile to say. And then there are those who only say three things during the whole semester, but those three things are absolutely brilliant. I’m clear in my syllabus that “participation” means having something worthwhile and informed to say, but it’s still really difficult to convey that idea.

    And some people–extroverts, men, etc.–seem to be disproportionately represented in participation. That’s kind of OK for me, since for many of my courses, a student will do very well in the blogging, but not in class, or the other way around. Multiple learning styles and all that.

    You might have them bring in questions that they plan to ask and have them turn them in on slips of paper. Then, even if they don’t get the chance to get their question in, they can at least get credit for thinking of one.

    In any case, I don’t think I could as easily abandon traditional grades in a 50-person class, though I might turn to forced ranks as an experiment there.

  4. Ed Crowder
    Posted 12/28/2008 at 10:42 pm | Permalink

    I was at Hampshire College way back when and they did the Evergreen thing, too. (Markup test here.) Transferring was fun, because admissions simply couldn’t make heads or tails out of the Hampshire evaluations. Having experienced both systems, I far prefer grading. It is a transparent and unambiguous system. What’s more, I’ll go a step further: Colleges should go back to the old system where they POST students’ grades in public locations. It is a great way to harness the power of peer pressure (which these days leads kids to beer bongs rather than better grades). And, frankly, I think it would help cull some of the dead wood.

  5. alex
    Posted 12/28/2008 at 10:57 pm | Permalink

    Well, students will still be getting letter grades for the class. I don’t know. They just don’t signify much to me. But as long as we are giving them, I am not opposed to giving them publicly. Unfortunately, federal law prohibits doing so.

  6. Posted 12/29/2008 at 1:29 pm | Permalink

    Feh. Like anyone obeys federal law these days.

Post a Comment

Your email is never published nor shared. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>