Grading for Learning: How to Make Your Grades Mean Something

Grades are supposed to communicate what students know and can do. In practice, many grading systems communicate something much murkier: a combination of compliance, effort, behavior, and academic performance, weighted in ways that vary by teacher and obscure actual learning. Here's how to grade in ways that are honest, informative, and motivating.

What Grades Actually Measure (Often)

Traditional grading practices accumulate a range of factors that aren't related to academic learning:

Completion and compliance: turning in work on time, following directions, participating in class. These behaviors matter for other reasons, but they're not evidence of learning. A student who turns in every assignment but doesn't understand the content and a student who misses half the assignments but understands everything should not receive the same grade.

Extra credit and bonus points: these inflate grades above what the core assessments show, making it harder to interpret what a grade actually means. A student with a 90% who earned 15 points of extra credit understands something different than a student with a 90% who earned no extra credit.

Averaging zeros: a single missing assignment averaged in as zero can mathematically destroy a grade that otherwise reflects competent performance. A student who earns 90, 95, 88, 92, and 0 (for a missing assignment) has an average of 73 — which communicates something false about what they know.

Early practice counted equally with final performance: averaging a student's first attempts at a skill with their polished, end-of-unit performance treats learning as a fixed characteristic rather than a developing one. A student who struggled in week one but mastered the skill by week four has learned. Averaging those scores suggests they've only half-learned.

Standards-Based Grading

Standards-based grading (SBG) addresses these problems by separating academic performance from compliance and behavior, reporting performance on each standard separately, and allowing the most recent evidence of performance to replace earlier evidence.

In SBG, a student's grade reflects their current level of mastery of specific skills, not a cumulative average of everything they ever attempted. This produces more accurate grades (they reflect what students can actually do now), more useful grades (they tell you what to reteach), and more motivating grades (improving scores are possible at any time, so improvement is always worth attempting).

The objections to SBG are usually administrative (it requires different grade books and reports) and cultural (parents and students are habituated to traditional grades). Neither objection is trivial, but neither is insurmountable. Many schools have implemented SBG successfully.

Minimum Grading

If full SBG isn't possible in your context, minimum grading is a simpler reform that addresses the zero problem. Setting a minimum score of 50% (or whatever score would allow mathematical recovery with subsequent work) prevents single-missing-assignment disasters without abandoning the traditional grading structure.

This is not "giving students points they didn't earn" — it's recognizing that a grading scale that runs from 0% to 100% treats the range from 0 to 60 as equivalent to the range from 60 to 100, which is mathematically nonsensical. A student who earns 50% has done something; a student who earns 0% may have done nothing or may have simply failed to submit. Treating them identically makes the grade less informative, not more.

Separating Academic and Behavioral Grades

Mixing behavioral grades (late work, participation, effort) with academic grades produces a composite that accurately represents neither. The cleaner approach: report academic performance and behavioral habits separately.

This doesn't mean behavior doesn't matter. It means that "Angela turns in work late but has mastered the content" and "Marcus turns in work on time but hasn't mastered the content" should produce different academic grades, even if their behavioral profiles produce similar behavioral ratings.

Retakes and Revision Policies

Allowing retakes and revisions treats grades as communication rather than punishment. If the purpose of a grade is to communicate current level of mastery, then updated evidence of mastery should update the grade.

The objection that "students will just wait for the retake instead of trying" is worth taking seriously — but the response is in the structure, not in prohibiting retakes. Requiring evidence that the student has done additional study before retaking, requiring that students correct their original work before retaking, and limiting retakes to within a window of time all address the gaming concern without eliminating the retake option.

Feedback vs. Grades

One of the most consistent findings in assessment research: detailed written feedback on student work improves learning more than grades alone — but when both feedback and grades are given together, students tend to read the grade and ignore the feedback. The grade short-circuits the learning signal.

Some teachers address this by separating feedback from grades: give detailed comments first, let students respond to them, then record a grade. Others use ungraded practice with feedback before formal assessment. The goal is to ensure that students actually engage with feedback rather than treating it as decoration on the grade.

Your Next Step

Audit your current grade book: what percentage of a student's grade in your class reflects academic learning versus compliance and behavior? If the answer is "less than I thought," identify one specific change — a different late work policy, separating extra credit from grades, allowing one retake — and implement it for the next unit.