Put the same Heart of Darkness essay in front of four English teachers and you may get four different scores. The differences rarely come from carelessness. They come from different assumptions about what counts as strong analysis, how much to reward style, and how to weigh a good idea against a messy draft.

For a department head, that variation is a fairness problem. Students in one section may receive noticeably different scores from students in another for equivalent work. Parents notice, and so do students.

Calibration is the process of narrowing that gap. It does not mean eliminating professional judgment. It means agreeing on a shared understanding of the standard, so that judgment operates within a range.

The core tool is the anchor paper. An anchor is a real student essay, scored and annotated, that illustrates what a particular level of performance looks like. A set of four or five anchors gives every teacher a reference point.

Running a norming session

A useful norming session takes about an hour. Teachers score a small set of papers independently, then compare scores and discuss the differences. The talking matters more than the numbers, because it surfaces the assumptions behind them.

• Select five to six papers that span the range of quality
• Have every teacher score independently before any discussion
• Compare scores and focus on the papers with the widest spread
• Record the reasoning behind agreed scores as brief annotations
• Save the annotated papers as anchors for the rest of the unit

Where teachers most often disagree

On Conrad essays, disagreements usually concentrate in a few places. One is the treatment of context: some teachers reward historical detail generously, while others want it tied to the text. Another is the treatment of ambitious but flawed arguments, where a bold claim is poorly supported.

Name these fault lines in advance and decide as a group how to handle them. Writing the decision into the rubric prevents the same argument from returning next semester.

Checking for drift over time

Calibration fades. After a few weeks, teachers drift back toward their habits. A short mid-unit check, where everyone scores one shared paper and compares, restores the alignment at little cost.

Tracking score distributions by section can also reveal problems. If one section's average is a full level above the others, it is worth a conversation, not an accusation.

How tools can support consistency

Software cannot replace a norming session, but it can extend its effects. When an essay grading tool such as GraideMind applies the department's rubric the same way to every paper, teachers start from a common draft and adjust from there. The consistency of the first pass narrows the range of final scores.

Departments can also use those drafts as material for calibration. Comparing a tool's draft comments with a teacher's own often exposes which rubric descriptors are unclear, and that is a useful thing to know.