Working rubric examples for two common use cases: writing placement (determining course level) and program assessment (measuring learning outcomes for accreditation). Designed to be adapted, not copied wholesale — your institution's standards and student population should shape the final version.
Behavioral descriptors outperform evaluative ones. Fewer criteria scored carefully outperform many criteria scored hastily. Performance levels should be meaningfully distinct. Criteria weights should reflect actual course priorities.
Five criteria across four placement levels (Developmental, ENG 101, ENG 102, Exempt/Advanced): Thesis and Argument (25%), Organization (20%), Evidence and Support (25%), Grammar and Mechanics (15%), Style and Voice (15%). Each criterion includes behavioral descriptors at each level describing observable features rather than quality judgments.
Three learning outcomes across four performance levels (Exemplary, Proficient, Developing, Beginning): Critical Thinking and Argumentation, Written Communication, and Use of Evidence and Sources. Benchmark for meeting an outcome is typically Proficient or above.
Run a norming session with faculty to calibrate shared understanding of descriptors, especially at adjacent-level boundaries. For AI-assisted scoring, provide 2–3 representative essays at each level with written explanations. Align criteria weights with your actual course outcomes.
See our guides on AI essay scoring for program assessment and setting up a writing placement test.