Character analysis is the assignment most teachers reach for with Blubber, and it is also the one most likely to be graded inconsistently. Jill is a narrator who changes, Wendy is a leader who manipulates the group, and Linda is a target whose inner life the book mostly leaves unexplored. Without a rubric tuned to those realities, two teachers can read the same essay and award very different scores.

A strong rubric starts by defining what analysis means at this grade level. For many fifth and sixth graders, it means moving from describing what a character did to explaining why the character did it and what that choice reveals about the story. That small shift, from action to motive, is where most of the learning happens, so the rubric should reward it explicitly.

Evidence is the second major dimension. Students should be able to point to a specific moment, such as Jill laughing along with the class before she becomes a target herself, and explain how it supports their claim. Vague references to the book as a whole should score lower than a precise scene paired with a clear explanation.

Four Criteria That Keep Scoring Manageable

Four criteria are usually enough for a novel-unit essay: claim, evidence, explanation, and conventions. Adding more tends to slow grading without improving the feedback, because students cannot act on ten separate scores at once. Each criterion should have three or four performance levels written in plain language that a ten-year-old could understand if you read them aloud.

• Claim: states a debatable idea about a character instead of a plot fact
• Evidence: uses an accurate scene or quotation tied directly to the claim
• Explanation: shows how the evidence reveals motive, change, or conflict
• Organization: groups related ideas and moves logically between paragraphs
• Conventions: keeps errors from getting in the way of meaning

Describing Levels Without Vague Language

Phrases like good analysis or strong support mean little to students. Replace them with observable descriptions, such as the essay explains why Jill joins the group's teasing and connects that choice to her later experience of being excluded. A student reading that description knows exactly what the top level looks like and what is missing from their own draft.

It also helps to attach one sample sentence to each level. A sentence that merely states Wendy is mean sits at the bottom, while one that explains how Wendy uses jokes to keep classmates afraid of becoming the next target sits near the top. Concrete examples reduce arguments about scores and make revision far easier.

Applying the Rubric Across Many Essays

Consistency matters most at the margins, where an essay sits between two levels. Re-reading the first five essays after you have scored twenty is a simple way to check that your standards did not drift during a long grading session. Many teachers find that early papers are scored more harshly or more generously than later ones.

AI-assisted grading can help hold the line here because it applies the same written criteria to the first essay and the hundredth. You still review the results and override scores when a student's reasoning is unusual but valid, which happens often with a book as open-ended as Blubber. The combination of a clear rubric and consistent application gives students a fair, understandable picture of their writing.