Every large humanities course has the same complaint sooner or later: my TA grades harder than yours. Students compare their scores across sections, and small differences feel like injustice. The complaints are often right, since graders drift without a shared standard.

Plato essays make calibration especially important. The papers are interpretive, and two thoughtful readers can reasonably disagree about a paragraph. Without clear guidance, TAs fall back on personal preferences.

Calibration is not complicated, but it does need to be deliberate. A single meeting before grading begins prevents most of the disputes that would otherwise arrive later.

Here is a step-by-step approach that fits within a busy schedule.

Run a Norming Session

Choose four or five sample essays that represent a range of quality. Have every TA grade them independently, using the rubric, and bring their scores to the meeting. The point is not to reach agreement immediately but to see where the differences are.

• Collect all scores before any discussion starts.
• Look at where the biggest gaps appear and ask each grader to explain.
• Agree on how to interpret unclear rubric language.
• Rescore the samples after the discussion to check alignment.
• Save the annotated samples as a reference for the rest of the term.

Give TAs a Comment Bank

A shared set of feedback phrases for common issues improves consistency in tone as well as content. Students receive comparable guidance regardless of who grades them. TAs also work faster because they are not reinventing explanations.

Encourage TAs to personalize the comments so they do not feel canned. The bank is a starting point, not a script.

Monitor During the Grading Period

Check in halfway through with a spot review. Ask each TA to send three graded essays, and compare them. If scoring has slipped, a short conversation is easier to fix now than after grades are released.

Look at score distributions too. If one TA's average is far from the others, investigate whether it reflects the papers or a difference in standards.

Add Software as a Consistency Check

AI grading tools can score every essay against the same rubric, giving you a reference point for comparing TAs. Papers where a TA's score differs sharply from the tool's are natural candidates for review.

This is not about replacing TAs. It is about giving instructors better visibility into how grading is going, so they can help TAs grow and give students a fairer experience.