Ask any teacher who has graded a full class set and they will tell you the same thing: the hundredth essay does not get the same read as the first. Fatigue is real, and it shows up in scores. An Ender's Game essay set makes a good example because dozens of students write on the same prompt, which invites comparison and drift. Knowing the patterns helps you guard against them.

Research on scoring has long shown that graders drift over a session. Standards tighten or loosen without the grader noticing. The paper you read after three strong ones may score lower than it deserves, and the one after three weak ones may score higher. That is called a contrast effect.

The halo effect is another trap. A student who wrote a great first paragraph may get the benefit of the doubt for the rest, while a messy opening can sour your reading. Knowing a student's past performance adds another layer. You may find yourself scoring the student instead of the paper.

Fatigue also affects comments. Early papers get thoughtful notes. Later ones get short, blunt ones. Students do not see the fatigue, only the difference in feedback.

Where bias creeps in

Handwriting, formatting, and length can sway scores. A tidy, typed paper looks stronger than a scrawled one, even if the ideas are the same. Long essays can feel more thorough than they are. A clear rubric that names the qualities you care about reduces this pull.

• Grade one criterion at a time across all papers
• Shuffle the stack so the order changes between sessions
• Cover names or use anonymous grading where possible
• Re-score two or three early papers at the end to check for drift
• Take a real break every 45 to 60 minutes

Simple habits that keep scores steady

Keep anchor papers next to you as you grade. When a paper lands in a gray zone, compare it to the anchor for that score. This works best when the anchors come from your own class and reflect your rubric.

Re-check a sample at the end. Pull five papers you scored early and re-read them. If your scores would change, you have found drift, and you can adjust the batch.

Using a second reader

A second reader offers a useful check. It can be a colleague, a teaching assistant, or software. The point is to compare scores on a sample and discuss differences. Large gaps signal a problem in the rubric or in the reader.

An AI tool such as GraideMind applies the same rubric to every paper without getting tired, which makes it a steady reference. It is not perfect, and it can carry its own biases, so compare its drafts against your judgment. Where the two differ, look closer.

Building fairness into your process

Write down your grading process and follow it each time. A written routine reduces the temptation to make exceptions when tired. It also helps you explain your choices if a student or parent asks.

Invite students to ask about scores using a specific process, such as a short written request pointing to a rubric line. That keeps conversations focused and gives you a chance to catch real mistakes. Fair grading depends on being open to correction.