Test and improve a design
A test provides information for revision. Changing one feature at a time can help show which change affected the result. Adding folds to a paper beam lets it support a different load in a repeated same-gap test. A scientific explanation connects a claim with observations or a stated model. Separate what you directly notice from an inference about why it happened. Use the same labels and units when comparing records, and say when the available evidence is too limited to decide. You may record with pictures, words, objects, pointing, or another communication tool.
A worked example
Complete invented classroom source/model record: Revision record, same paper and gap: flat beam trials hold 2, 3, 2 counters; folded beam holds 6, 6, 7. Worked explanation: folds are the planned changed feature, and the higher recorded loads support trying that revision again.
Starting scenario: Adding folds to a paper beam lets it support a different load in a repeated same-gap test.
- Day 1: Read the embedded test and improve a design record. Make a labeled baseline drawing or table with all stated quantities, conditions and observations. Write a testable question; distinguish a prediction from what is already recorded.
- Day 2: Reconstruct the record as a diagram or safe adult-approved physical model. Record a first trial, change one beam feature and repeat the same load procedure with adult support. Record the method, planned changed feature and shared conditions; complete both practice questions after making your own prediction.
- Day 3: Repeat the same safe observation or model comparison at three planned times or trials. If using only the embedded record, draw three trial-planning panels and explain what each new observation could add; do not invent measured results.
- Day 4: Make a before-and-after design page explaining one evidence-based revision. Include labeled observations, a claim connected to the record and one limitation that needs further investigation. Use the embedded record and your dated observations to annotate the model. Make one justified revision and preserve both versions so a partner can trace the evidence behind your decision.
- Day 5: Present a claim, two relevant observations and reasoning about test and improve a design. Compare your prediction with the record, explain one model limitation, complete the fresh check, and propose a specific next observation.
Open the printable learning material