What Worked / What Changed

What did not work, and the fix

Plenty did not work the first time. These are the course corrections that mattered — kept here because the next round goes better when the friction is written down.

The first lines of inquiry were too abstract to research

Students could name a topic but not write a line of inquiry that led anywhere. Sentence frames and a few worked examples fixed it. Most groups rewrote their lines at least once before they were any use on the wall.

Students copied AI prompts straight into their notes

Some pasted prompts and answers from MagicSchool into their notes without reading them. The notes looked full and meant nothing. So we added a step: rewrite every finding in your own words in Cornell Notes. Slower, but the notes started to mean something.

Describing a good project map in words did nothing

Groups could not picture what “good” looked like from instructions. Once we posted two finished maps on the wall, the next round improved on its own.

Groups treated their Canva slides as the display

So we set a rule: don't print it, and don't leave an iPad at the station running the slides. Canva was a checklist for tracking tasks and keeping work moving between classes. The station was what visitors saw, not the screen.

Build week fell apart when materials were a free-for-all

Cardboard vanished, tape ran out, tables were a mess. We taught the routine directly: where things live, how to take and return them, and the rule that you plan before you build.

A poster that just hung there got ignored

The artifacts that worked were the ones students had to touch — move a sticky, fill a gap in the table, check off a fix, rewrite a line. When the wall asked for an action, the kids used it.

Feedback should lead to change.