# Repair and rehearse, 15:10–15:45
Rotate roles: analyst repairs one consequential issue, communicator prepares the account, auditor checks evidence. Spend 15 minutes repairing or logging a remaining blocker, 12 minutes rehearsing a three-minute account and receiving feedback, and 8 minutes deciding remaining actions.

## Three-minute account
- 0:00–0:30: biological question, outcome and planned contrast.
- 0:30–1:10: allocated/sampled units, the number of animals or groups assigned to each treatment or selected for observation and analysis matched to the design.
- 1:10–2:10: main figure, estimate and uncertainty in useful units.
- 2:10–3:00: diagnostic/sensitivity finding, limitation and one follow-up design decision.

Listeners ask: What exactly is compared? Which observations provide replication? What assumption could change your conclusion? What effect is worth planning for?

## Proposed final eight-minute talk
| Time | Content |
|---|---|
| 0–1 minute | Question and context; clearly identify synthetic or observed data |
| 1–2½ minutes | Design, units, allocation/grouping and primary model |
| 2½–4½ minutes | Main figure, practical estimate and uncertainty |
| 4½–5½ minutes | Diagnostic decision and sensitivity comparison |
| 5½–7 minutes | Follow-up design and prospective simulation under two assumptions |
| 7–8 minutes | Supported conclusion, limits and next useful information |

Allow five minutes of questions. Each student should be able to explain the design and main estimate; share speaking time sensibly. Avoid reading a table of coefficients. Put extra details in backup slides.

## Remaining-actions ledger, 15:45–16:00
| Action | Named owner | Evidence of completion | Date dependent on presentation date? |
|---|---|---|---|
| | | | |

After class, 1–2 hours total: revise report/figure (30–45 minutes), prepare slides (20–35), full timed rehearsal with questions (10–20), final render/path/source check (10–20). Choose durations within the cap. Presentations are on Wednesday 2 December, 12:30–16:00. Formal assessment requirements will be shared separately. Report intended format: 3–5 narrative pages plus reproducible source/data, one outcome, contrast, diagnostic, sensitivity and future design with simulation under two assumptions.