About the science

Fair tests beat one impressive model

Change one design feature at a time — insulation, glazing, or orientation. Keep the model size and test conditions the same. Compare indoor temperature against outdoor temperature, and log your design version so the pattern across versions becomes the evidence, not any single reading.

Good to know

• One change at a time

  Changing insulation and orientation together means you can't tell which one mattered.

• Design versions matter

  Your third model teaches more than your first — the record of what changed is the evidence.