Grade the 60th paper like you graded the first

You know the moment. It's late, you're grading the thirty-eighth submission, and you catch yourself re-reading a paragraph you'd have waved through two hours ago, or going back through one you've flagged this morning. The work hasn't changed. Your standard has.

Grading drift is one of the quietly uncomfortable facts of the job. Fatigue, mood, the order papers happen to land in, the strong one you read right before a weak one... etc. all of it could influence the marks. Two students hand in genuinely comparable work and, through no bad intent, walk away with different grades. We rarely talk about it, because the honest version sounds like an admission. It isn't. It's what happens when consistency is asked to live entirely inside one tired human's head across sixty judgments.

The vague rubric is where the drift gets in

Most of the drift traces back to the rubric — or the lack of one. "Content: 40%. Analysis: 30%. Presentation: 30%." That isn't a scale, it's four labels, and everything that actually decides the grade happens in the gap between them, in the moment, differently each time. On paper one you're generous about what counts as "strong analysis". By paper forty, "strong" means something stricter. Nobody wrote that shift down, because there was nothing precise enough to shift against.

And it costs you twice. Once in fairness — the same work getting different marks. And once when a student comes to your office to contest a grade. With four vague labels, the conversation is your word against theirs. With a rubric that says, concretely, what "Excellent" versus "Satisfactory" work looks like for this specific task, the mark defends itself. You're not defending a feeling; you're pointing at a line the student can read too.

Same work, same grade — whether it's paper 1 or 60

The fix isn't to grade faster or care harder. It's to move the standard out of your head and onto the page before you start: criteria drawn from what this assessment actually asks students to demonstrate — not generic "Quality" and "Content" — with a short, observable description of what each level looks like. Anchor the levels to your institution's own passing mark, so "pass" means the same thing on the last script as on the first. Then the sixtieth paper meets exactly the standard the first one did, because the standard stopped depending on what time it is.

The catch, of course, is that building that rubric (specific, aligned to your outcomes, level by level) is itself an evening's work you don't have. So the vague four-label version wins, and the drift comes back. That's the gap we built the rubric tool for. You give it the assessment you're actually grading and your course outcomes; it builds a rubric with concrete criteria, the achievement levels you choose, and score bands anchored to your passing grade — in minutes. You read it like a capable TA's first pass and adjust until it's yours. The judgment stays with you; the drift doesn't.

Your course. Your style. Always aligned.

Grade every paper by the same standard: build the rubric for one of your assessments and see the difference on the sixtieth script. Get early access →