Character analysis is the most common assignment attached to Slam, and it is also one of the hardest to grade consistently. Sam is a sixteen-year-old skateboarder who narrates his own life in a chatty, self-deprecating voice, so students often confuse retelling his story with analyzing who he is. A teacher reading sixty of these essays in a weekend will notice that half of them summarize plot while only a few explain how Sam changes. Setting up a grading approach before the stack arrives makes the difference between useful feedback and exhausted skimming.

The first decision is what the essay is actually supposed to prove. A prompt such as "Describe Sam" invites a list of traits, while a prompt such as "Explain how Sam's idea of responsibility shifts after Alicia tells him she is pregnant" demands a claim that can be argued and tested against the text. Teachers who write the stronger version of the prompt find that grading becomes easier because every essay can be measured against the same question. The prompt itself is the first piece of the rubric.

Once the prompt is settled, decide what a strong thesis looks like before reading any student work. In a Slam essay, a strong thesis names a specific change in Sam, such as his move from treating problems as something he can skate away from to accepting that other people now depend on him. A weak thesis says only that Sam is a typical teenager who learns a lesson. Writing one sample strong thesis and one weak thesis on a note card gives you a fast reference point during grading.

Build a Rubric Around Change, Not Traits

Character essays tend to collapse into trait lists because students find it easier to say Sam is immature or funny than to show how he develops. A rubric that rewards movement over time pushes students to compare early and late scenes, such as how Sam jokes through his first conversation about the pregnancy and how he handles later moments differently. This also helps the grader, because the question becomes whether the essay traces a shift rather than whether it sounds thoughtful. Weighting the change criterion more heavily than mechanics signals what the unit values.

• Thesis names a specific change in Sam rather than a general trait
• At least two pieces of textual evidence from different parts of the novel
• Explanation connects each quote to Sam's thinking or choices
• Attention to Sam's narration style and how it shapes the reader's view of him
• Organization that follows Sam's development rather than the plot order alone

Score Evidence and Explanation Separately

Students often drop in a quote and assume the work is finished, which means a single combined score for evidence hides the real weakness. Splitting the criteria lets you reward a student who found a great moment from the novel while still pointing out that the explanation never connects it to the argument. For example, a student might quote Sam addressing his Tony Hawk poster and then say only that this shows he is sad. A stronger response would explain what the habit of talking to a poster reveals about his need for guidance.

This separation also produces better feedback language. Instead of writing "needs more analysis" in the margin, you can write that the quote is well chosen but the next sentence should explain what it reveals about how Sam handles pressure. Students tend to act on that kind of comment because it points to a single concrete fix. Over a whole class, the pattern of low explanation scores also tells you what to reteach before the next essay.

Where AI Feedback Fits Into the Process

A rubric built around thesis, evidence, explanation, and organization is exactly the kind of structure an AI grading tool can apply consistently across a large stack. The teacher defines the criteria and the point values, and the tool produces draft scores and comments tied to those criteria for every essay. That frees the teacher to spend time on the judgment calls that matter, such as an unusual interpretation of Sam's choices that deserves credit even if it departs from the class discussion. Consistency on the routine criteria is where most grading time is saved.

Teachers should still read a sample of the essays and adjust anything that does not match their expectations. Calibrating on five or six papers before relying on the draft feedback catches problems early, such as a rubric that accidentally rewards length. After that calibration, the same rubric can be reused for every section of the course. The result is a Slam unit where students get specific feedback faster and teachers get their weekends back.