One figure, one paragraph
Week 12 · In-class workshop
Gavin Simpson
Aarhus University
Wednesday, 25 November 2026
Resume: communicate the lesson result
| 13:55–14:30 |
Cress teaching figure and paragraph |
| 14:30–14:55 |
Peer review and revise |
| 14:55–15:10 |
Break |
| 15:10–15:45 |
Your project analysis workshop |
| 15:45–16:00 |
Share decisions and individual exit |
What should a reader recover?
Question · comparison · direction · units · time · conditions · uncertainty
The caption should say that researchers assigned watering to trays, describe the model and comparison, and identify the data as invented.
A figure should support one defensible conclusion.
Write 100–150 words
Include:
- The watering question, how researchers assigned trays, and how they counted sprouts.
- Model, estimate and interval in useful units.
- Practical interpretation and one caveat.
Say the data are invented and the tray deviation is set to zero.
Write, label and check
Use the cress result and worksheet.
- Label comparison, day 10, units and interval.
- State what is held fixed and what is averaged.
- Distinguish estimation uncertainty from seed variation.
By 14:30: one figure, caption and paragraph.
Peer review: recover before discussing
Swap teaching-example drafts.
- Recover the conclusion unaided.
- Check the watering comparison, units, interval and number of trays receiving each method.
- Give one change that improves the scientific explanation and one feature to keep.
- Authors revise immediately.
Break 14:55–15:10.
Your project: justify your own analysis
15:10–15:45:
- Say what researchers do to the animals or groups and what they measure.
- Choose the quantity to estimate and candidate model.
- Explain what you hold fixed or average over, and which model checks you use.
- Record an alternative and a consulting question.
The cress model is a lesson example, not a prescribed project analysis.
Share and individual exit
Each group: one justified project decision + one unresolved issue.
Individually explain:
- Why crossing zero differs from showing no effect.
- Why precision does not establish causality.
- One project decision you must justify next.