District curriculum teams often use shared novels to create common assessments, because a single text lets them compare writing across schools. A book like Blubber is attractive for this purpose since it is short, accessible, and generates varied but comparable responses. The challenge lies in making sure the data produced is trustworthy.

Common assessments are only meaningful if scoring is consistent across classrooms. If one school's teachers are generous and another's are strict, differences in the data reflect grading habits rather than student learning. Leaders should therefore invest as much in scoring consistency as in prompt design.

The prompt itself should be aligned to the standards the district wants to measure. If the goal is to assess evidence-based argument, the prompt must require a claim supported by textual evidence. A vague prompt generates vague data.

Designing the Prompt and Rubric Together

Prompt and rubric should be developed as a pair. Drafting the rubric first often reveals gaps in the prompt, such as a criterion that students are never actually asked to demonstrate. Piloting both with a small group of teachers and sample essays before district-wide release can catch problems early.

• Align the prompt directly to one or two named writing standards
• Draft the rubric and prompt together and test them on sample essays
• Provide anchor papers at each performance level
• Offer a brief scoring training for all participating teachers
• Plan a check where a second scorer reviews a sample of essays

Scoring Training and Reliability

Scoring training does not need to be lengthy, but it should include hands-on practice with real student essays. Teachers score independently, compare results, and discuss disagreements until they reach shared understanding. Documenting the decisions made creates a reference that supports scorers throughout the year.

Reliability checks provide evidence that scores can be trusted. Having a second reader score a small percentage of essays, and calculating how often scores agree, offers a measure of consistency. Districts can use that information to identify schools or teachers who may need additional calibration.

Scaling Scoring Across Many Schools

Scoring thousands of essays manually is a serious logistical burden, and it often delays the return of results. AI-assisted grading tools can apply a single rubric to every essay in the district, producing consistent first-pass scores and feedback. Human reviewers then verify samples, address outliers, and make final decisions.

This approach shortens the time between assessment and action, which makes the data useful for instruction. Teachers receive results while the unit is still fresh, and leaders can identify patterns across schools. A well-designed common assessment, supported by consistent scoring, becomes a genuine tool for improving writing instruction rather than just another reporting requirement.