Reframing Grading Forum 2025

Traditional grading systems have their roots in promoting disparity and continue to do so today. If the only purpose of a grade is sorting of students by prior learning and ability to learn quickly, traditional grading works well. When pressed, though, most instructors suggest their grades are meant to show a student's facility with a course's knowledge or skills. For that purpose, traditional grading systems conflate multiple abilities under a single and uninterpretable measure, prohibit students from showing delayed learning, and make a actual student's learning in the course opaque to the instructor, other stakeholders, and worst of all, the students themselves.

Instructors often use numbers as categorical data that reflect learning: Student performance is translated to a numeric score. For example, a 7 or an 8 out of 10 is probably meant to signal "good." Instructors then take those categorical data and calculate with them (think averages, standard deviations, and more) because these categorical data are now in the form of numbers. The practice isn't an instructor's fault; the conflation is easy because the representation is ambiguous. The result, however, is a longstanding and fundamental misapplication of statistics to numerically-represented categorical data. Ask yourself: What is the average of "good," "well done!" and "this needs work?" across three different content areas and tasks?

A couple examples highlighted the fact that averages serve as a poor indicator of performance. (Look for the parachute packer example in her talk!) If what we want to know is "Can the student do this part of my course?" an average will penalize late learning and obscure the answer. While there are no perfect systems, there are better ways to answer this question, including (but not limited to) using a student's last attempt or best attempt.

What does a grade do, anyway? Several participants valued this challenge and used a template to answer this very question:

The purpose of my grades is to ___(1)___  ___(2)___ to ___(3)___.

1. What action are your grades meant to perform? Choose a strong verb.
2. What academic or non-academic factors receive this action? Choose an object the verb acts upon.
3. Who is the intended audience? Choose a specific group.

The keynote includes an example. Try this yourself, and answer: What does a grade from you do?

A few faculty prized the PPP system described during the talk. PPP is preparation / participation / practice. In Sharona's system, PPP stands alongside the course standards as its own "bucket." Students accumulate points in a way familiar to most faculty, and the points are awarded to students based on—you guessed it—the student's preparation, participation, and practice throughout the semester. The standard is met when students accumulate 750-ish of 1200-ish points, and that met standard can contribute to their final course grade. In this way, the PPP standard for the course might be considered an aggregated standard for soft skills—students learning "how to student" through the type of learning actions instructors value.

Participants felt reassured that they could, like the keynote speaker, develop alternatively graded courses within the Canvas architecture available to us. Obviously, this can help streamline the implementation of alternative grading systems. Note, though, that there may be a learning curve up front. See below for more details.

Most instructors have, at this point, had some reckoning with learning outcomes or other declarations of our values for student learning. During the Forum, one participant made a particularly salient point: observability and measurability are not the same (at least in the context of our discussion.) When it comes to learning, observability is a necessary but insufficient condition for measurability. We might broadly observe students in the process of learning, but need an artifact of that learning behavior (e.g., a product) to measure. Our learning standards must be measurable in order to take on real meaning.

Such a small shift with such a big effect: Start your goals for students (outcomes, standards, specifications, etc.) with two words. "I can." Students immediately know that these goals are for them, not for a theoretical audience. Upon achievement of the goal, students get a boost, and the affirming language can motivate subsequent learning in the course.

Several faculty felt relief when they realized that "alternative grading" is not a single, monolithic practice. Instead, "alternative grading" is any practice that reduces aggregated points-based notions of learning in one-off attempts in favor of increasing disaggregated and qualitative notions of learning progress over time and across effort—a long way to say it's "The Wild West" of grading. That said, four guiding principles—the "four pillars" model of Clark and Talbert—have emerged from the milieu of work in this area and can be used by faculty when making adjustments to (or even redesigning) a course.

A significant drawback to traditional grading systems is the one-and-done demonstration of learning. There are a couple facets to this:

• If a student can't recover (mathematically, in terms of their course grade) from early failures in a course, why would they continue putting effort toward learning in the course?
• Persistence is a "soft skill" worth developing in today's students and can be codified as a learning outcome in a course. However, this only works if structures are in place to allow for persistence on a task.

One of the pillars of alternative grading is to allow students repeated attempts on a task without penalty. This helps students develop persistence and to continue to the end of a course.

Faculty appreciated that alternative grading—and specifically its "four pillars" (clear standards, meaningful feedback, marks for progress only, and reattempts without penalty)—can help all students. And we might be sympathetic to this form of learning. It's what most of us have come to rely upon after school and in our adult lives. It honors the fact that none of us learn without struggle or failure accompanied by the opportunity to improve or correct. Alternative grading whendone right reveres authentic learning processes.