When several teachers in a department teach the same novel, students reasonably expect that a similar essay will earn a similar grade regardless of who marks it. In practice, this is rarely the case, because teachers interpret rubric language differently and apply their own standards. A shared unit on La Voix sur la montagne is an ideal opportunity to improve consistency across classrooms.

Calibration is the process of aligning how teachers read and score student work. It usually begins with a shared rubric, but the rubric alone is not enough. Teachers need to apply it to the same pieces of writing and discuss where and why their scores differ.

A department head can lead this process with relatively little time investment. Choose four or five anonymous essays that represent a range of quality, have everyone score them independently, and then meet to compare results. The discussion often reveals hidden assumptions about what counts as strong analysis or acceptable language.

Running a calibration session

The most effective sessions are short and focused. Start by reading one essay aloud or silently, score it, and share the results on a board. Where scores differ by more than a level, ask the teachers to point to the exact lines in the essay that led to their decision.

• Select anonymous essays that cover weak, average, and strong performance.
• Have each teacher score independently before any discussion.
• Compare scores and discuss the largest differences first.
• Revise unclear rubric descriptors based on the discussion.
• Save the agreed-upon examples as anchor papers for future years.

Revising rubric language

Calibration conversations often expose rubric descriptors that are too vague. Phrases like insightful analysis mean different things to different readers. Replacing them with observable descriptions, such as explains how at least two pieces of evidence support the thesis, reduces disagreement.

The department should keep a record of the revisions and the reasoning behind them. This helps new teachers understand the intent of each criterion and prevents the rubric from drifting back to vague wording over time. A short shared document is usually sufficient for this purpose.

Using technology to support consistency

AI grading tools can play a useful role in calibration by applying the same rubric language to every essay. If a department uploads its shared rubric to a platform like GraideMind, the draft feedback produced is based on one consistent reading of the criteria. Teachers can compare it to their own scores to identify where their interpretations differ.

It is important to treat this output as a reference point and not as the final word. The aim is to prompt discussion, not to override professional judgment. When teachers see a difference between the tool's reading and their own, it opens a useful conversation about the rubric.

Sustaining calibration over time

Calibration is not a one-time event. Staff change, students change, and standards can shift gradually. Revisiting the anchor papers once a year keeps everyone aligned and helps new colleagues learn the department's expectations.

Over the years, a department builds a library of calibrated examples for each major unit. This resource reduces preparation time and supports fair grading for students. It is one of the most valuable long-term investments a department can make.