Not part of your permanent record!

This is another post where I explore and struggle with the notion of grading.

This week I reviewed the first substantive assignment I asked my intro students to write this semester. It was actually assignment number three, where the first two took effort but not much thought. This one asked students to complete a sophisticated task: the apply an economic theory to a real world situation and, after identifying the facts revealed by the theory, to draw some normative conclusions about the situation. I have used this assignment for years and know what to look for in their answers. As I read the essays, I wrote substantive comments for how the essays could be improved, raising questions that would help students take their thinking to a higher level. Only after my reading and writing of comments on all the papers did I start to think about assigning grades.

A not-insignificant number of students failed this assignment. That is, they approached this as a non-economist would, going with their intuition rather than letting the theory reveal the answer to them. Should I give these students an ‘F’ on their essays?

I think it depends on the purpose of the grade. Is the purpose of the grade to make a summative judgment about what the student has achieved on this assignment? Does an F imply earning less than 60% of the total credit on the assignment? I don’t think it should. For one thing, this assignment isn’t about learning content, where learning 59% of the facts isn’t a passing grade, but learning 61% is. Rather, the assignment, applying a theory to derive insights about an issue or problem, is the fundamental goal of undergraduate economics education. It is what economists do. It is far from trivial, and students in the intermediate theory courses regularly show they haven’t grasped it yet. So the objective is not to score the degree to which students fully demonstrate this task, but to help them learn how to competently go about the task.

I have another concern. There’s something about a formal grade and the way that students usually interpret grades that gets in the way of learning. I think this is especially the case for first year students who often have always excelled grade-wise. Getting even a C on an assignment can be a rude awakening to college. I think that the feedback teachers provide on assignments, particularly early in the term, should provide operational information about how students can improve. Traditional grades don’t necessarily do that.

We need to provide opportunities for students to take intellectual risks, but where the consequences of failure are minimal. We need to encourage students who have taken the right approach, a very sophisticated approach, but one which they haven’t executed perfectly, to continue to work in that direction.

What I propose for homework assignments is focusing on the approach students take and (largely) ignoring the details of their reasoning. What this implies is affirming the student who’s taken the right approach but whose product might not be refined enough to earn an A. It also implies providing a signal to the student who is fundamentally taking the wrong tack, but who may have provided enough ‘content’ to avoid an F. In other words, it’s not about assessment of where they are, but the direction in which they are heading.

In practice, I give students credit for a genuine effort at completing the assignment. (On rare occasions I give no credit if it’s clear that the student didn’t put in real effort.) I evaluate their effort using a three level scale which is easily assessed: On the right track, Not on the right track, or Sufficiently well done to be nearly perfect. When computing final grades for homework, I come up with a holistic judgment based on the pattern I see in their assignments graded this way, from which I determine a traditional letter grade.

This entry was posted in Assessment & Grading. Bookmark the permalink.

4 Responses to Not part of your permanent record!

  1. Jenny says:

    “There’s something about a formal grade and the way that students usually interpret grades that get in the way of learning.” Our societal focus on grades has pushed us farther and farther from a focus on learning. Our best institutions of learning are often our pre-K education and our graduate schools. Many, many of those do not have grades. That should tell us something.

  2. Dispersemos says:

    I appreciate these comments on grading and the underlying concern for student learning. To my mind, grades are only useful to the extent that they communicate to students their progress toward learning objectives. Letter grades don’t do this well. Comments on student assignments and one-to-one conferences do this work much better.

    That said, summative assessment plays an important role in learning, and it will be a long time before we abandon the letter grade system at the college level. I see the need, then, to make sure that students understand what constitutes A-level work, B-level work, etc. For summative assessment, I pass out rubrics that link letter grades to clear language about criteria for assignments. When students receive their graded assignment, the rubric shows them how their work measures up to the criteria.

    On the topic of grading, I learned much from Ken O’Connor’s book _How to Grade for Learning_.

  3. Gardner says:

    I’ve been thinking a lot about grading lately, inspired by the presentation I did for the Baylor religion department. I’ve got a few tentative conclusions and many, many questions. One tentative conclusion is that grading is a technology that reveals chronic problems (what do we mean by understanding? by deep learning?) after it appears to solve acute problems (how can we assign a portable, context-free evaluation of student work that plugs neatly into the whole GPA and credit system we’ve evolved?) The GPA-and-credit system evolved for a reason, sure, but it’s now so normalized that it begins to be an end in itself and the supporting technology (like grading) looks like necessary infrastructure rather than a set of choices that have a history and might be done differently. Radical thinking, I know, but there you go.

    I ran across this site yesterday while googling on “deep learning”: http://www.learningandteaching.info/learning/deepsurf.htm. I thought it had some interesting things to say about extrinsic motivation and the ways in which surface learners stubbornly redirect all encouragement to learn deeply because of years and years of successful surface (or even “strategic”) learning. For what it’s worth, I think grading on aptness of approach is a great idea. Content memorization without the mastery that comes from deep learning pretty much guarantees mediocre thinking and rules out insight and thought-leadership, in my view. That’s not to say that one shouldn’t memorize things. Of course one should. But one doesn’t memorize the alphabet so that one can then memorize a phrase book. One memorizes the alphabet to give one freedom to utter one’s mind and become a knowledge co-creator.

    Looks like some good stuff here as well: http://www.engsc.ac.uk/er/theory/learning.asp

  4. Gardner says:

    An afterthought: I’m not sure it’s really a risk if the consequences are minimal, but I need to think about this some more. I agree completely that we need to encourage students to pursue sophisticated strategies even when the immediate results aren’t quite on target.

    I do think that we need to send clear, unambiguous messages about the need for deep learning instead of surface learning, and I really do like the assessment approach that asks the learner about their own behaviors (part of the APGAR idea) as a way of encouraging them to begin to behave in ways that will, if they have the stamina and ability, result in deeper learning.

Leave a Reply

Your email address will not be published. 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=""> <strike> <strong>