Guide + free calculator
Justify your sample size with confidence.
Sample-size questions are one of the most common reasons proposals come back from ethics committees. This guide explains the choices and lets you run the numbers.
Worked example
Estimating ANC attendance prevalence (~70%) with 5% margin and 95% confidence using Kish Leslie:
n = (1.96² × 0.70 × 0.30) / 0.05² ≈ 323
Inflate by 10% for non-response → final sample ≈ 356.
Run your own numbers Frequently asked
Questions researchers ask
- What formula should I use?
- It depends on your design and outcome. Cross-sectional prevalence studies often use Kish Leslie; comparative studies use formulas based on expected effect size.
- What is a reasonable margin of error?
- 5% is standard. Tighter margins increase sample size sharply, looser margins reduce precision.
- Do I need to add for non-response?
- Yes. A 10–15% inflation is common; adjust based on your fieldwork context.
- What if I am doing qualitative research?
- Sample by saturation: when new interviews stop producing new themes: typically 12–25 in-depth interviews or 4–8 focus groups.
- Does my ethics committee need a justification?
- Yes. Always cite your formula, your assumptions and your source.
Related tools and guides