Co-Construct Rubrics¶

Strategy · All strategies
Evidence · 2 claims (2 for) · 5 studies (4 quant-synthesis, 1 review), q2–q4 · 4 of 5 report an effect size

Description¶

Co-constructing rubrics means involving learners in building the assessment instrument itself: examining samples of work, identifying what distinguishes strong from weak performance, and drafting the criteria and level descriptors together with the instructor. The finished rubric then guides both the learners' production and the evaluation of it. The activity converts assessment criteria from an external mystery into knowledge learners articulate themselves.

Design Implications¶

Co-construction operationalizes Assessment for Learning by making success criteria explicit and owned by learners, which strengthens self-regulation and the accuracy of self-evaluation Involving learners in assessment practices improves achievement. [+M]. Because quality judgments are a form of task-level information, the discussion during co-construction functions as feedback delivered before the work rather than after Feedback is most effective at the task and process levels. [+S]. The negotiation itself — arguing about what counts as "thorough" versus "superficial" analysis — is where most of the learning happens, so the process matters more than the artifact.

Context¶

Requirements¶

• Authentic samples of work at multiple quality levels (student work, Non-Examples, or published exemplars) for learners to analyze
• Instructor willingness to share assessment authority and to accept learner-proposed criteria that are defensible
• Time for structured discussion — typically a full session, not a handout at the end of class
• A draft rubric or skeleton the group can revise, rather than starting from a blank page

Constraints¶

• Co-construction without exemplars to anchor discussion produces vague, unmeasurable criteria ("creativity," "effort") [-M]
• If the instructor later grades against a different, hidden rubric, the exercise damages trust and yields no benefit [-M]
• Groups without prior experience judging work tend to propose criteria about surface features (length, neatness) rather than quality of reasoning [~M]; the instructor must push past these
• Time cost is real; for low-stakes or highly standardized tasks, a pre-made rubric may be the better trade-off [~W]

Implementation Variability¶

• Analyze-then-draft: learners sort exemplars into quality ranks, then articulate the criteria that distinguish them (most robust)
• Revise-the-draft: instructor provides a draft rubric; learners critique and rewrite descriptors in their own words
• Student-generated criteria, instructor-formatted: learners propose criteria; instructor consolidates into a usable instrument
• Ongoing co-construction: the rubric is revisited and amended after each major task as understanding of quality deepens

Target Learners¶

• Learners new to a genre or task type who do not yet know what quality looks like in that domain [+M]
• Adolescents and adults with enough metacognitive development to discuss quality abstractly; younger learners benefit more from simplified criteria co-creation [~W]
• Less valuable for expert learners who already internalize domain quality standards [~M]

Target Learning Goals¶

• Internalizing quality criteria for complex, open-ended products (essays, projects, portfolios, designs)
• Self-regulated learning: building the evaluative judgment needed for self-assessment and revision
• Metacognitive development: articulating what "good" means, not just producing good work

Instructions¶

1. Present 3–5 anonymized work samples spanning the quality range; have learners rank them and justify the rankings in small groups (Case Studies function as exemplars here)
2. Facilitate whole-group discussion to extract the dimensions that distinguish the ranks; name each as a criterion
3. Draft performance-level descriptors collaboratively, using language from the learners' own justifications; the instructor sharpens wording for measurability
4. Have learners apply the co-constructed rubric to a sample (Assessment as practice), then compare judgments with peers and the instructor to calibrate
5. Use the same rubric for the actual task, and revisit it afterward — revision of the rubric after real use deepens understanding of quality

Related Strategies¶

• Self-Assessment — co-constructed rubrics give self-assessment a shared, calibrated instrument
• Peer Feedback — peer review is only as good as the criteria; co-construction supplies them
• Exemplar Analysis — the exemplar-sorting activity that typically launches co-construction

Examples¶

• Writing workshops (e.g., National Writing Project classrooms): teachers and students examine mentor texts, name the moves that make them effective, and build the rubric used for peer response and revision
• Project-based learning (e.g., PBLWorks Gold Standard PBL): project rubrics are co-developed with students during the "build the culture" phase so students help define what a high-quality product means
• Design studio courses: instructors and students jointly critique past project portfolios to derive presentation and critique criteria before the first studio review

Key Sources¶

• Black, P., & Wiliam, D. (1998). Assessment and classroom learning. Assessment in Education: Principles, Policy & Practice, 5(1), 7–74. doi:10.1080/0969595980050102
• Sadler, D. R. (1989). Formative assessment and the design of instructional systems. Instructional Science, 18(2), 119–144. doi:10.1007/BF00117714
• Andrade, H. L. (2005). Teaching with rubrics: The good, the bad, and the ugly. College Teaching, 53(1), 27–30. doi:10.3200/ctch.53.1.27-31
• Panadero, E., & Jonsson, A. (2013). The use of scoring rubrics for formative assessment purposes revisited: A review. Educational Research Review, 9, 129–144. doi:10.1016/j.edurev.2013.01.002