use test data and kind feedback to compare designs and find what to improve.
Phase 4: Build
Do Now. Look at your data. Circle the result you are proudest of. Put a star by the result that needs the most work.
Which comment gives the most useful feedback?
The reading opens over the lesson; close it to come back to exactly where you were.
Data helps us compare. Do not just ask, “Which latch do we like?” Ask, “Which latch meets more criteria? Which stays within our constraints?”
Look across trials. Are the numbers close together? Then the latch acts in a steady way. Do they jump around? Something may be loose or out of line.
Find the failure point. Name the exact part and the exact criterion. “The latch is bad” does not help. “The tab tears on pull 3” helps a lot.
Critique means kind, useful feedback. Use these starters: I like…, I wonder…, What if…?
Worked example 1: Ruby compares two latches (sample data). The Washer Catch scored 11, 12 and 11. The Two-Magnet Catch scored 29, 30 and 31. Ruby says the Washer Catch is inside the sweet spot. The Two-Magnet Catch is too hard to open.
Worked example 2: Theo gives feedback. Theo tells a team, “I like that your scores are steady. I wonder why the door missed 4 clicks. What if the washer were bigger, so it is easier for the magnet to catch?”
Trade data pages with another team. Practice with these questions first.
Write your team’s findings.
Which design met our criteria best? Use at least two pieces of data.
Name one failure point in your latch. Which criterion does it miss? How do you know?
Feedback is a gift. Tomorrow you will use it to make Version 2.
Exit ticket: What is one idea you got from another team?
How confident are you that you can I can give and use helpful feedback.
Pacing: Do Now 5, instruction 10, gallery critique and guided practice 15, independent 5, closure 5.
Materials: data tables, prototypes, sticky notes.
Norms: critique the design, not the person. Require one piece of evidence in every comment.
Misconceptions: the highest Pull Score wins; a rare problem can be ignored; percent is hard, so read 5% as 5 out of 100 only.
Supports: color the sweet spot on a number line and mark each score. Stretch: compare steadiness and strength together.
Answer guide: comparisons must cite data against the Design Card. Failure points must name a part and a criterion.
NASA, NASA's Perseverance Rover Deposits First Sample on Mars Surface · U.S. government work, public domain
On December 21, 2022, the Perseverance rover set a sample tube on the ground on Mars. The tube held a piece of rock about the size of a piece of chalk. It was dropped from about 3 feet high.
Each tube has a flat end piece called a glove. Future missions can grab the tube by it. But that flat end caused a question. Could a dropped tube land standing straight up?
Engineers tested this on Earth with OPTIMISM, the rover’s full-size copy. They looked at the results of their drops. The tube stood on its end less than 5% of the time. That means fewer than 5 drops out of every 100.
The team did not ignore this small chance. They wrote commands so the rover could gently knock a standing tube over.
Engineers look at many results before they decide. They plan for problems, even ones that do not happen often. You can do that with your latch data too.
Source: NASA, NASA's Perseverance Rover Deposits First Sample on Mars Surface (nasa.gov)