Sampling Methods in Research: Probability, Purpose, and Practical Limits
A guide to probability and non-probability sampling, sample size reasoning, and how to defend your sampling section.
Sampling is an evidence boundary
Your sample defines what your findings can reasonably say. Probability samples support statistical generalization when the frame is strong. Purposeful samples support depth and relevance when the question needs specific cases or participants.
The method should be named clearly: random, stratified, cluster, convenience, purposive, snowball, theoretical, or maximum variation. Then explain why that method fits.
- Use probability sampling when representativeness is central.
- Use purposive sampling when cases are selected for information richness.
- Explain recruitment, inclusion criteria, and likely bias openly.
Sample size is justified, not guessed
For quantitative work, sample size should connect to power, precision, or expected effect size. For qualitative work, it should connect to depth, saturation, and diversity of cases.
A strong sampling section does not claim perfection. It shows readers why the sample is sufficient for the claims the study makes.
Put the guide to work
Draft with your sources visible.
Start a Thesion project, gather sources, and keep citations attached while you build the draft.