Cluster sampling
Cluster sampling randomly selects entire groups from a population, then surveys units within those groups. It cuts costs for dispersed populations but sacrifices precision because units within a cluster tend to be similar to one another.
FrameworkSampling design
See it move
Cluster sampling selects entire groups — school classes, branches, census blocks — at random, then surveys the units inside them, which is far cheaper than reaching scattered individuals across a dispersed population. Simple random sampling instead draws individual units directly. Because units within a cluster tend to resemble each other, cluster sampling loses precision compared with a simple random sample of the same size, trading statistical efficiency for lower cost.
Check yourself
A researcher studying employee wellbeing at a nationwide supermarket chain randomly selects 20 branches and surveys every employee within those branches. What sampling method is this, and what is its primary statistical limitation?