When three teachers assign the same essay on Synnøve Solbakken, a student's grade can vary significantly depending on which section they are in. One teacher may reward bold interpretation while another prioritizes structure. These differences feel arbitrary to students and can undermine confidence in the department.

Calibration is the process of aligning how different graders interpret the same rubric. It does not require identical opinions, only a shared understanding of what each score level means. Done well, it improves fairness and makes grades more defensible.

The process is best done before grading begins, when teachers can still adjust their expectations. A single hour of calibration can prevent weeks of inconsistent scoring. It also builds collegiality, since teachers learn from one another's reasoning.

Run an Anchor Paper Session

Select several essays that represent different score levels and have each teacher grade them independently. Then discuss the differences, focusing on why each person scored as they did. The goal is to reach a consensus on what evidence of each level looks like in practice.

• Choose anchor essays across the score range
• Grade independently before discussing
• Compare scores and identify disagreements
• Discuss reasoning and revise rubric descriptors
• Record the agreed anchors for future use

Check Consistency During Grading

Calibration should not end when the session does. Have teachers exchange a small sample of graded essays midway through and compare scores. Drift is common, particularly as fatigue sets in, and an early check can correct it before it affects many students.

Keep a short record of the borderline cases and how they were resolved. These examples become a valuable reference for future grading and for training new teachers. They also support responses to grade appeals.

Use AI as a Consistency Check

An AI grader applies the same rubric to every essay, which makes it useful as a reference point. Comparing its scores to each teacher's can reveal systematic differences, such as one section scoring consistently higher. This is not about overriding teachers but about spotting patterns worth discussing.

Treat discrepancies as prompts for conversation rather than proof of error. Sometimes the tool is wrong, and sometimes a teacher's reading reveals something the rubric missed. Either way, the discussion improves shared understanding.

Make Calibration a Routine

Departments that calibrate regularly see more consistent grading and fewer disputes. Schedule a session at the start of each major assignment and keep the anchor papers on file. Over time, the process becomes quick and routine.

Share the outcomes with students in general terms, explaining that teachers work together to ensure fair grading. This transparency reassures students and families. It also reflects a professional culture of shared standards.