Introductory literature courses at large universities can enroll hundreds of students, and every one of them is expected to write analytical papers. Professors and teaching assistants need assignments that teach close reading while remaining feasible to grade. Roald Dahl's short stories work well because they are short enough to read in one sitting and rich enough to support serious analysis. They also appeal to students who may not think of themselves as literature people.

A good paper assignment in this context is narrow, specific, and clearly tied to course skills. Asking students to analyze how a single scene builds suspense, or how a particular object accumulates meaning across a story, keeps the scope manageable. It also produces papers that are easier to compare, since all students are working from the same textual base. Broad prompts tend to generate sprawling papers that are slow to read and difficult to score fairly.

Course design matters as much as prompt design. If the semester includes two or three papers, each can emphasize a different skill, such as close reading, comparison, and argument. This progression gives students a chance to build competence step by step. It also lets the grading team calibrate their expectations for each skill in turn.

Calibrating Teaching Assistants

In large courses, teaching assistants do much of the grading, and consistency among them is a persistent concern. A calibration meeting, in which everyone scores the same sample papers and discusses discrepancies, is one of the most effective tools available. The goal is not identical scores but a shared understanding of what each level of performance looks like. Following up with a second calibration mid-semester keeps standards aligned.

• Distribute three anchor papers representing strong, average, and weak work
• Have every grader score them independently before discussion
• Resolve disagreements by pointing to specific rubric language
• Keep a running document of decisions about edge cases
• Spot-check a sample of each grader's papers during the grading period

Writing Feedback That Scales

With hundreds of papers, comments must be efficient and meaningful. Many instructors use a combination of a summary comment at the end and a small number of marginal notes that illustrate key points. Focusing on the thesis and the quality of analysis, rather than correcting every sentence, gives students the most useful guidance. A consistent structure for comments also helps students interpret them.

AI-assisted tools can support this process by drafting initial comments aligned with the rubric, which graders then review and refine. This is particularly useful for routine observations about structure and evidence, freeing graders to spend time on the interpretive work that requires expertise. It also reduces the fatigue that leads to inconsistent comments late in a grading session. The human reviewer remains responsible for the final judgment.

Addressing Academic Integrity

Large courses are especially vulnerable to concerns about plagiarism and unauthorized AI use. Short, well-known stories like Dahl's are easy to find summaries of, so assignments need to require original interpretation tied to specific passages. Process-based components, such as an outline or a short in-class paragraph, make it easier to confirm that students are doing their own work. Clear policies stated in the syllabus reduce misunderstandings.

Instructors can also design prompts that ask students to connect the story to a lecture or discussion unique to the course. This makes generic outside answers less useful and rewards attendance and engagement. It is a more constructive strategy than relying solely on detection. Students who feel the assignment is personally relevant are also less likely to cut corners.

Using Results to Improve the Course

Grading data offers valuable information about what the course is teaching effectively. If a large share of students struggle with thesis construction, that suggests a need for more instruction before the next paper. Patterns in rubric scores can guide where to focus lecture time and section activities. Instructors who treat grading as a source of insight improve their courses year after year.

Sharing summary results with teaching assistants also strengthens the teaching team. It helps everyone understand the course's goals and encourages conversations about strategy. Over time, the course develops a body of shared practice that new graders can learn from. That institutional memory is one of the most valuable assets a large program can build.