The Constitution Project That Fell Apart in Two Days

Diagnosis: this is group work, not cooperative learning. Placing four students at a table with one product creates a setting, not a structure. Four specific design gaps explain all four symptoms:

1. No positive interdependence. Nothing in the task requires students to need each other, so arguing over who does what is a rational response to an ambiguous assignment.
2. No individual accountability. The grade attaches to one poster, so the strongest reader writing everything is the efficient strategy, not a character flaw. The two silent students can free-ride with no consequence.
3. No assigned or rotating roles. Silence from the two English learners is predictable when speaking is optional and there is no rehearsal time.
4. Cognitive demand too low. A poster that can be completed in 20 minutes by copying the glossary tells you the task asked for transcription, not reasoning. Students who finish early are not off task; they are correctly completing a low-level task quickly.

Redesign.

• Jigsaw the content. Each team member becomes the expert on one branch plus one check (e.g., veto override, judicial review, treaty ratification). Experts meet in like-role groups first, then teach their home team. Now every student is a required source of information for the others - that is interdependence.
• Individual accountability. A short individual quiz on all four branches after the jigsaw, averaged into a team score only if every member meets a threshold. This makes teaching your teammates the path to your own grade.
• Named roles that rotate: facilitator, evidence finder, visual recorder, reporter. The reporter must be able to answer questions the recorder did not prepare.
• Raise the task. Replace the poster with a written advisory memo on a current conflict between Congress and the president that must cite two specific constitutional clauses and predict the outcome. It cannot be copied from a glossary.
• Structured talk. Three-minute think time, then a round-robin where each member must speak once before anyone speaks twice.
• Group processing. Two minutes at the end: what worked, what one thing will the team change tomorrow.

Why this addresses the symptoms. The arguing disappears when roles are assigned; the writing monopoly disappears when the product requires each member's specialized knowledge; the silence disappears when speaking is required and rehearsed; the glossary poster disappears when the product demands an argument.

The rule to apply. Scaffolds change the route to the reasoning, never the reasoning itself. The target stays: explain separation of powers and checks and balances with evidence. What can flex is text complexity, vocabulary load, and the format of the response.

Specific scaffolds.

• Text sets, not a single textbook. Three versions of the same content at different reading levels, all containing the same three key ideas. The below-grade-level readers get a version at their instructional level; the eleventh-grade-level text is available for anyone who wants it. All students answer the same question.
• Vocabulary pre-teaching for the two English learners two days ahead: six to eight terms with visuals and a cognate check (separation, branch, veto, override, treaty, judicial). Front-loading terms removes the decoding barrier without removing the concept.
• A structured note-catcher with headings tied to the jigsaw roles, so students know what they are listening for when a peer teaches.
• Discussion frames posted and practiced aloud: 'The check on the president is ___ because ___.' Frames carry academic grammar so students can spend their effort on argument.
• Visual anchor - a three-column chart of branch, power, check - that students complete rather than copy.
• Scaffolded writing for the memo: a template that still requires the student to choose the clause and justify the prediction.

Defending it. The distinction to state plainly is between differentiation of input and differentiation of expectations. Every student in this class writes the same memo, answers the same quiz items, and is scored with the same rubric on the same reasoning. The leveled text substitutes for a textbook chapter that is written three grade levels above half the class - a student who cannot decode the chapter has zero access to the reasoning, so the leveling is the precondition for rigor, not a departure from it.

The evidence to bring. Offer to show the chair one student's memo written from the plain-language text and ask whether the argument meets the standard. If it does, the scaffold worked. Also note the alternative: without the text set, the six below-grade-level readers produce nothing, which is not higher expectations - it is a hidden reduction in what they are asked to learn.

The planning problem. Three weeks at 42 minutes is about fifteen periods, minus the two laptop days per week (six total). A project that assumes daily technology will collapse; a project that assumes none wastes what he has. The fix is to assign technology a specific, narrow job and build the rest of the work around print and people.

Structure.

• Week 1 - build knowledge, no laptops. Jigsaw expert groups work from the leveled text set and printed primary-source excerpts (two short ones: the veto message and the Federalist excerpt on checks). Daily product: a completed note-catcher section, collected.
• Week 2 - apply, laptops on the two available days. Laptop day 1: research the current conflict and gather two sources with citation. Laptop day 2: draft the memo in a shared document. Off days: peer critique using the rubric, revision by hand, and evidence conferences with the teacher.
• Week 3 - present and assess. Two days of team presentations of the memo (four minutes each, all members answer one question), one day of individual written assessment, one day of reteach based on what the assessment shows.

Micro-structure inside a 42-minute period. A predictable five-part frame so no time is lost to transitions: five-minute warm-up with the day's question posted, five-minute role assignment and materials, twenty-five minutes of work, five-minute team processing, two-minute exit ticket. Post the frame permanently.

Daily monitoring, not end-of-project monitoring.

• The exit ticket asks one content question and one process question ('what did your team accomplish, what is left, who needs help tomorrow').
• A circulation checklist with one row per team: on task, roles active, evidence cited, equity of talk. Ten seconds per team, three passes per period.
• Artifact check at the end of each class: the day's product must be turned in or photographed. A team cannot be 'finished' with nothing to show.

Early warning signal. The team that produced the glossary poster in 20 minutes in the original design would have been flagged on day one by the artifact check, because the product did not answer the day's question. Monitoring the artifact, not the noise level, is what catches disengagement.