Stratified Sampling
Methodology & design
Probability sampling that divides the population into strata (e.g. grade, sector, region) and samples randomly within each — guaranteeing subgroup representation and usually improving precision.
Stratified sampling divides the population into mutually exclusive subgroups, or strata, that share a relevant characteristic — gender, department, region, firm size — and then samples independently within each. It is a probability technique, so it supports statistical inference to the population.
Its advantage over simple random sampling is precision and guaranteed representation. Because every stratum is sampled, small but important subgroups cannot be missed by chance, and sampling error is reduced when the stratifying variable genuinely relates to the outcome being measured.
Proportionate allocation samples each stratum in proportion to its share of the population, preserving the overall profile. Disproportionate allocation over-samples small strata so they can be analysed separately, which then requires weighting before population-level estimates are computed. Report the strata, the allocation method and the achieved sample within each.
Where it's used
- Employee surveys sampled proportionally across levels
- Multi-region studies requiring representative state-wise coverage
Related concepts
Guides that use this concept
All guidesNot sure where to start?
Book a free 15-minute consult. We'll map your next three steps — no obligation.