A type of random sampling, based first on the random selection of certain subgroups, from which the sample can be taken.
A cluster sample is one in which the whole population being studied is divided into clusters or groups, and a selection of these clusters is then made, and the entire population or a random sample of subjects within these sub-clusters are sampled. (e.g. if the sampling frame was an entire city, then the city would be broken down into city blocks and certain randomly selected city blocks would be sampled instead of the entire city).
A sampling design commonly used in retrospective mortality surveys when comprehensive lists of individual households cannot be obtained. Clusters are groups of households of which the first is chosen at random, and the remainder by a rule of proximity (e.g. second closest). In a cluster mortality survey, 30 or more clusters are usually sampled from the target study population, and each cluster usually contains at least 30 households.