Collaborative Creation of Exemplars¶

Strategy · All strategies
Evidence · 6 claims (6 for) · 13 studies (5 causal, 5 quant-synthesis, 1 review, 1 associational, 1 theoretical), q2–q4 · 4 of 13 report an effect size · 1 claim rests on one study

Description¶

After completing collaborative problem-solving tasks, learners examine leveled exemplars of responses and identify the characteristics that distinguish each level. They then create posters or anchor charts — often using their own work — that illustrate the differences between levels. The activity converts assessment criteria from teacher-held secrets into shared, learner-articulated standards, promoting metacognition, self-assessment, and a common understanding of quality work.

Design Implications¶

The strategy works because comparing multiple worked responses at different quality levels supports abstraction of the underlying criteria — learners induce what "good" looks like rather than being told Comparing multiple contrasting cases supports abstraction of deep structure. [+M]. Explaining why an exemplar merits its level is a form of self-explanation that deepens conceptual understanding of both the content and the evaluative criteria Self-explanation improves conceptual understanding. [+S]. Because the discussion centers on task and process qualities rather than the self, the resulting peer feedback aligns with the feedback levels most associated with learning gains Feedback is most effective at the task and process levels. [+S].

Context¶

Requirements¶

• Exemplars of varying quality (typically 3–4 levels) with clear differentiation between adjacent levels
• Completed learner work from the collaborative task, so criteria are grounded in learners' own products
• Structured discussion prompts ("What makes this a 3 and not a 2?") and materials for co-constructed anchor charts
• Time for groups to articulate, debate, and record criteria in their own words

Constraints¶

• Poorly differentiated exemplars collapse the levels into a single vague impression of "good," undermining the abstraction the activity depends on [~M]
• Learners with weak self-assessment skills may rate their own work generously or defer entirely to peers, producing feedback that is inaccurate or unactionable [~M]
• If the teacher supplies the criteria list up front, learners stop analyzing and simply transcribe, eliminating the generative processing that drives the effect [-M]
• Time-intensive; the analysis and chart creation can crowd out the practice the exemplars were meant to improve

Implementation Variability¶

• Analyze-then-create: groups study teacher-provided leveled exemplars, then build anchor charts from their own work
• Anonymized peer work: swap products between groups so evaluation is less ego-threatening
• Erroneous exemplars: include a flawed sample and ask groups to diagnose and repair it Erroneous examples build conceptual knowledge. [+M]
• Co-construction with rubrics: learners draft rubric language first, then test it against exemplars and revise — a formative-assessment cycle in the spirit of Assessment for Learning
• Digital variants: shared annotation tools (e.g., Perusall, Google Docs comments) replace physical posters for remote or older learners

Target Learners¶

• Upper elementary through secondary learners (grades 3–12); the leveled-exemplar structure is adaptable to adult and higher education with discipline-specific work samples
• Novices benefit most from seeing concrete level differences rather than abstract criteria lists Worked examples reduce unnecessary search for novices. [+M]
• Learners developing self-regulation, since comparing their work to leveled standards builds monitoring accuracy Self-monitoring improves self-regulation. [+M]
• Less beneficial for learners who already possess accurate internal standards for the task, for whom the leveled comparison is redundant [~W]

Target Learning Goals¶

• Metacognition: building accurate internal standards for judging one's own work
• Self- and peer-assessment: generating and applying evaluative criteria
• Conceptual understanding of quality: what distinguishes strong from weak performance in a domain
• Collaborative discourse: negotiating and justifying evaluative judgments with peers

Instructions¶

1. Complete a collaborative problem-solving task and collect the resulting work products.
2. Present leveled exemplars (e.g., levels 1–4) without labels; groups rank them and justify their ordering — a Demonstration of quality enacted through contrasting cases rather than a single model.
3. Facilitate Peer Discussion in which groups name the specific characteristics that separate each level, pushing beyond surface features ("longer," "neater") toward process and reasoning criteria.
4. Have groups create an anchor chart or poster using their own work samples to illustrate the level differences, with Individual Reflection on where their own product falls and what would move it up a level.
5. Post the co-constructed criteria and use them for subsequent self- and peer-assessment, revising them as understanding deepens.

Related Strategies¶

• Gallery Walk — a natural follow-on where learners apply co-constructed criteria to peers' posted work
• Two Stars and a Wish — structured peer feedback that can use the co-constructed criteria as its vocabulary
• Rubric Co-Construction — the same criteria-building move applied to formal scoring tools
• Structure reflection after collaboration for both students and teachers using prompts, exit tickets, journals, rubrics, plus-and-delta, and gallery walks
• Use multiple exemplars with scaffolded discussion rather than mere provision, balancing teacher guidance against student co-construction

Related Elements¶

• Demonstration — exemplars function as demonstrations of quality at multiple levels
• Peer Discussion — the negotiation mechanism through which criteria are articulated
• Individual Reflection — the step that converts shared criteria into personal self-assessment
• Collaboration — the task structure that supplies authentic work samples for analysis

Examples¶

• A 5th-grade class completes a collaborative math problem, then examines anonymized responses rated levels 1–4. Groups debate what distinguishes each level, then build a poster using their own work to show the differences; the poster becomes the class reference for future problem write-ups.
• Writing instruction (e.g., The Write Tools and similar trait-based programs): students rank writing samples by trait levels (ideas, organization, voice) and co-construct anchor charts that hang in the classroom as revision references.
• Secondary science: lab-report exemplars at three quality levels are analyzed by lab groups, who generate criteria for "evidence-based reasoning" before writing their next report.

Key Sources¶

• Sadler, D. R. (1989). Formative assessment and the design of instructional systems. Instructional Science, 18(2), 119–144. doi:10.1007/BF00117714
• Black, P., & Wiliam, D. (1998). Assessment and classroom learning. Assessment in Education: Principles, Policy & Practice, 5(1), 7–74. doi:10.1080/0969595980050102
• Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1), 81–112. doi:10.3102/003465430298487
• Andrade, H. L. (2010). Students as the definitive source of formative assessment: Academic self-assessment and the self-regulation of learning. In H. L. Andrade & G. J. Cizek (Eds.), Handbook of formative assessment (pp. 90–105). Routledge.
• Chi, M. T. H., & Wylie, R. (2014). The ICAP framework: Linking cognitive engagement to active learning outcomes. Educational Psychologist, 49(4), 219–243. doi:10.1080/00461520.2014.965823