Mastery grading, revisited

As you know, we use a “mastery grading” approach in this course. Since not everyone completely understands it, SamR wrote a message to some colleagues who were working with students in the course. Here’s that message. Perhaps it will be useful to some of the students in the course, too.

Since 151 takes an odd approach to grading, I thought I’d give you a quick summary.

CSC-151 primarily uses a “mastery grading” approach. That is, we care more that students master the material by the end of the semester than when they master material. Hence, they usually have a chance to redo the primary assessments for the course. There are two kinds of primary assessments: small “learning assessments” (exam problems / quiz problems) and larger “mini projects” (homework assignments). This spring, we plan 24 learning assessments (LAs) that correspond to 24 learning objectives as well as 9 mini-projects (MPs).

We give two kinds of LAs. Each Friday, students take an LA on paper during class. We also call these “quizzes”. Every three weeks or so, students take a set of learning assessments (SoLA) on the computer outside of class. In most cases, if a student misses an LA, they take another LA on the same topic at the next opportunity. If they miss a quiz, they try the topic on the next SoLA. If the miss a problem on a SoLA, they also try it on the next SoLA. There are a few topics that we only give on paper. For those, students get new opportunities each Friday, typically before or after class.

For mini-projects, students get another opportunity after they receive the graded work. In my sections of the course, there are fixed dates for those redos (and second redos and, in rare cases, third redos).

Those are the basics. Then there are some complicating factors.

Complicating factor 1: Tokens. We don’t want students to keep redoing mini-projects again and again and again. And we want to limit late work. We give students some ownership over these areas by issuing them virtual “tokens” that they can spend for redos and late work. (In essence, tokens are similar to the common policy of “you can have up to N late days for homework assignments”, but spread out a bit.) We don’t charge tokens for redoing LAs or for the first redo on any mini-project (provided students have made a reasonable attempt the first time). Students can earn extra tokens in various ways, most commonly by attending a designated activity (e.g., department talks, convocation; in my case, wellness activities and peer performances also count).

Complicating factor 2: Timely work. There are also some kinds of work that students need to be more timely about. This work includes the nearly-daily readings (and reading responses), lab writeups, and metacognitive pre-reflections and post-reflections for both mini-projects and SoLAs. We grade this work on a satisfactory/unsatisfactory binary, with any serious attempt counting a satisfactory, even if it’s wrong. In essence, these things help ensure that students try to keep up to date on the basic ideas of the course, even if they haven’t completely mastered them.

Complicating factor 3: Grading. How do we turn all of this information into a letter grade? In essence, students must reach a certain level of achievement in each category. In the past, we counted MPs, LAs, and all of the timely work. These days, the base grade is based only on MPs and LAs; other work may affect the base grade. For LAs, getting at least 22 topics correct is A level, 20 is a B level, and 18 is C level. MPs are a bit more complicated, but they also have A, B, and C level depending on how well they’ve done on the various mini-projects.

Complicating factor 4: Involving the timely work in the grade. We allow them to miss a certain amount of timely work. Right now, they can miss up to six total readings, lab writeups, and metacognitive reflections without affecting their grade. After that, missed timely work starts to drop their grade. Since the timely work is what they should be doing, this makes no difference to the vast majority of students.

While the policies are a bit strange, I think they are better (more equitable) than traditional percentage grading. Students definitely appreciate the clear mechanisms for improving their scores. And I remain convinced that a student who is able to show mastery of the course material on the LAs (small problems) and MPs (synthesis of multiple ideas) deserves a high grade, even if it takes them some time to show that mastery.

Let me know if you have questions!