AtlasSession 4L10: Sample size calculationsp.7, 11–12, 19–20
No previous study
- 9 earlier topics
- Beta error and power
- Why sample size matters
- Continuous outcomes
- Binary outcomes
- No previous study
Fig. 1
First MSC-hydrogel trial for disc degeneration, MRI disc height index (%): where does each sample size input come from?
Fig. 1: decision flowchart “Can you borrow numbers from an earlier study?”. Case: A brand-new disc treatment. Nobody has tested this disc gel before, so no earlier study says how much disc height varies between patients. Question: Is there an earlier study that tells you how much the outcome varies? Answers: 1 Yes, 2 No.
- Case
- A brand-new disc treatment
- Question
- earlier study?
- Your answers
- none yet
- How it’s going
- — (answer the first question to see how it’s going)
- Score
- 0/2 (no cases finished yet)
Decision Flowchart Walker · First look
Level 1 · First look · Lecture 10 · p.7, 11–12, 19–20
Explain that sizing a study needs facts from earlier work, and what to do when none exist.
Trace where each sample size input comes from and decide when a feasibility study or expert advice is needed instead.
Cooking for 100 (the step you are on)
A new recipe needs a trial run before you know how much to buy.
Cooking for 100 from a new recipe? Do a trial run first to learn how much to buy. Sizing a study works the same way.