Explicit capstone instruction
Use teacher-approved safe materials or a simulation. Preserve raw trial data before averaging or scoring.
Calculate the model residual or percent error, then compare performance with criteria.
If results disagree, distinguish between model failure, measurement uncertainty, build variation, and an incorrect system assumption.
Vocabulary: trial data; residual; percent error; validation; revision.
Worked example
A model predicts 50 units; trials are 46, 47, 45. Mean is 46, so residual is −4 and percent error relative to prediction is 8%.
Guided question
Would one 50-unit trial erase the evidence from several 46-unit trials? Why not?
Independent work
- Preserve all raw trials.
- Calculate mean and spread.
- Compare with prediction.
- Revise one model or design assumption.
Capstone Investigation
Use safe teacher-approved materials or simulation. Preserve raw trial data, model prediction, uncertainty, residuals, and revision history.
Misconception check
Do not discard inconvenient trials without a documented methodological reason.
Formative check
What makes a capstone test auditable?
Reveal response
Raw data, calibration, calculations, criteria, assumptions, and revision history.
Mentor evidence conference
Ask Archaemenes to challenge system boundaries, predictions, uncertainty, and whether revisions follow the evidence.
Teacher guidance
Foundation: Use supplied data if physical testing is unavailable.
Core: Complete prediction-versus-test analysis.
Extended: Quantify sensitivity and uncertainty.