Sampling methods explained
Sampling methods split into two families. In probability sampling, every member of the population has a known, non-zero chance of being chosen — which is what lets you generalize your results to that population with statistics. In non-probability sampling, selection isn’t random; you can’t generalize statistically, but it’s often exactly right for qualitative, exploratory, or hard-to-reach studies.
Probability sampling (for generalizing)
| Method | How it works | Note |
|---|---|---|
| Simple random | Every member has an equal chance | Gold standard; needs a full sampling frame |
| Stratified | Random sampling within subgroups (strata) | Ensures key subgroups are represented |
| Cluster | Randomly select whole groups, sample within | Cheaper for dispersed populations; less precise |
| Systematic | Every k-th member from an ordered list | Simple; beware hidden periodic patterns |
Non-probability sampling (for depth / access)
| Method | How it works | Common in |
|---|---|---|
| Convenience | Whoever is easiest to reach | Pilots; quick studies (highest bias risk) |
| Purposive | Deliberately pick information-rich cases | Qualitative research |
| Quota | Fill preset quotas per subgroup (non-randomly) | Market/opinion research |
| Snowball | Participants refer others | Hard-to-reach / hidden populations |
How to choose
- Need to generalize to a population with statistics? Use a probability method and a calculated sample size.
- Exploring meaning, or studying a specific/hidden group? Purposive or snowball sampling usually fits.
- Be honest about bias. Convenience sampling is common in student projects — acknowledge its limits rather than over-claiming representativeness.
Get the free Research Design Decision Tree
A step-by-step decision tree from your research question to the design that fits it — qualitative, quantitative, or mixed methods — with the sampling approach each path implies, five matched question-to-design examples, and a ten-point pre-commitment checklist, from Research Design Simplified. We’ll email you the download link.
Frequently asked questions
Probability vs non-probability sampling?
Probability = known chance of selection, supports generalizing; non-probability = non-random, limits statistical generalization but fits qualitative/exploratory work.
What are the main probability methods?
Simple random, stratified, cluster, and systematic.
What is purposive sampling?
Deliberately selecting information-rich cases that fit the study’s purpose — common in qualitative research.
Which should I use?
Probability + a calculated sample size to generalize; purposive/snowball to explore or reach specific groups.
The Sample Size & Power Analysis Workbook
“How many participants do I need?” — the answer, and the paragraph that proves it.
64 fill-in templates. Paperback $32.99 on Amazon
See inside the workbookJust want the templates? The Sample Size & Power Analysis Template Kit: every template as a fillable PDF, $19.