Unit 01 · Day 6
Conduct & Analyze
Run the preregistered paper-helicopter investigation with a complete audit trail: preserve raw observations, document deviations, summarize variation, visualize the relationship, and separate what the data show from what you expected to see.
Learning objectives
- Carry out an approved investigation consistently and stop when the approved conditions cannot be maintained.
- Record every valid raw trial with units, condition labels, and relevant observations before calculating summaries.
- Distinguish an unusual observation from a justified exclusion and preserve an audit trail for any exclusion.
- Calculate and verify mean and range for each rotor condition.
- Create a correctly labeled graph and describe the observed pattern without overstating causation or certainty.
- Evaluate variation, confounders, measurement limitations, and the most useful next improvement.
Materials
- Approved Day 05 investigation plan and raw-data table
- Completed paper helicopters built to the approved dimensions
- Stopwatch/timer
- Metric ruler or tape
- Science notebook
- Khaemenes Scientific Calculator and Dataset Lab through Science Studio when useful
Use the approved clear drop area and stable floor-level footing. Never climb to increase drop height. Keep people, animals, airflow sources, liquids, heat, electrical equipment, fragile objects, and traffic paths clear. Stop immediately if the environment changes in a way that makes the approved method unsafe or scientifically incomparable.
1. Pre-run integrity check
Do not begin merely because the helicopters are ready. Compare today's conditions with yesterday's approved plan.
| Check | Confirm before Trial 1 |
|---|---|
| Rotor levels and construction | Match the approved dimensions and controlled features |
| Drop height | Measured using the approved reference points |
| Release rule | Same operational definition for every condition |
| Timing rule | Same start/stop definition and unit |
| Environment | No material airflow or safety change |
| Raw-data table | Ready before observations begin |
| Stop conditions | Known before testing |
2. Collect raw data without result hunting
- Run conditions in the planned order.
- Record each fall time immediately with its unit.
- Record relevant observations such as collision, damaged rotor, release problem, unexpected airflow, or timing interruption.
- Do not round differently because a value looks inconvenient.
- Do not repeat only trials that disagree with the prediction.
- Do not erase a surprising value. Preserve it and evaluate it later.
If a trial is invalid because a documented protocol failure occurred, keep the original value visible, label the reason, and record any replacement trial separately. “It did not fit the trend” is never a valid exclusion reason.
3. Anomaly is a question, not a delete button
An anomalous observation is unusually different from the rest of the data or expected pattern. It may reflect natural variation, measurement error, a procedural failure, or a real phenomenon. Its unusual appearance alone does not tell you which.
For every flagged trial, write: (1) what is unusual, (2) whether a protocol/event record explains it, (3) whether it remains in the primary analysis, and (4) how the conclusion changes if it is included versus excluded when that comparison is useful.
4. Process each condition transparently
For each rotor length, calculate the number of analyzed trials, mean fall time, minimum, maximum, and range. Keep raw data visible beside the summaries.
Use the Khaemenes Scientific Calculator to verify arithmetic if desired. A tool can catch arithmetic mistakes; it cannot decide whether the experiment was fair or whether an exclusion is justified.
| Rotor length (cm) | n | Mean fall time (s) | Min (s) | Max (s) | Range (s) |
|---|---|---|---|---|---|
| _____ | _____ | _____ | _____ | _____ | _____ |
| _____ | _____ | _____ | _____ | _____ | _____ |
| _____ | _____ | _____ | _____ | _____ | _____ |
5. Graph the relationship
Create a graph with rotor length (cm) on the x-axis and mean fall time (s) on the y-axis. Include a descriptive title, labeled axes, units, sensible numerical scales, and plotted points that can be traced back to the summary table.
Use the Dataset Lab if available to compare your hand-built interpretation with a digital visualization. Do not let software choose the scientific claim for you.
Does your scale distort the apparent size of differences? Are units present? Can a reader identify every plotted condition? Does connecting points imply continuity that your design did not actually test? Explain one graphing choice.
6. Describe before you explain
First write an observation-only description: direction, shape, size of differences, variation, and any unusual points. Then write a separate interpretation.
Interpretation language: “This pattern is consistent with the idea that changing rotor length affected descent behavior under these test conditions, although ___ limits confidence.”
7. Evidence-quality evaluation
Answer with reference to your actual data and run notes:
- Was the predicted direction observed across the tested range?
- How large was within-condition variation relative to differences between condition means?
- Which control was hardest to maintain?
- Did any documented deviation threaten comparability?
- Which measurement limitation mattered most?
- What alternative explanation remains plausible?
- What single improvement would most strengthen a replication?
Write one paragraph that distinguishes result, interpretation, and limitation. Do not call a limitation “human error” without naming the actual mechanism.
8. Mentor + Science Studio evidence conference
Bring your raw table, calculations, graph, exclusions/deviations, and interpretation to the Science Mentor. Be prepared to answer:
- Show me one conclusion that comes directly from a measurement rather than your prediction.
- Why did you keep or exclude your most unusual trial?
- Which graphing choice could change how a reader perceives the pattern?
- What would you expect an independent replication to reproduce?
- What evidence would make you revise your present explanation?
Science Studio tools and Mentor feedback support checking and revision. Neither substitutes for the student's raw evidence or ≥80% mastery.
9. Choose your pathway
| Pathway | Required analysis |
|---|---|
| Foundation | Complete all planned trials, preserve raw data, calculate mean and range, make a labeled graph, and identify one limitation. |
| Core | Complete full audit trail, mean/range for every condition, graph audit, anomaly decision, alternative explanation, and evidence-quality evaluation. |
| Extended | In addition to Core, compare whether the pattern appears linear across tested levels; examine how justified inclusion/exclusion choices affect summaries; propose a preregistered independent replication or improved measurement method. |
10. Exit check
- Why is an unusual result not automatically an error?
- What information must accompany a justified exclusion?
- Why should raw data remain available after means are calculated?
- What does range tell you that the mean does not?
- What is the difference between describing a graph and explaining its pattern?
- Why should conclusions be limited to the conditions actually tested?
Save raw data, protocol notes, calculations, graph, anomaly decisions, evaluation, and revised interpretation in the portfolio. Completing Day 06 does not bypass the ≥80% Unit 01 mastery gate.