A probability sample for which groups or geographic areas comprising groups were randomly selected

sample in which collections of individuals ("clusters") are sampled as units, so that whenever one individual in the cluster is included in the sample, then so are all other individuals in that cluster.

The selection of groups of elements from a population.   Typically used in multi-stage area probability designs to improve the efficiency of fieldwork by assuring that groups of neighboring households are interviewed.  Since neighbors tend to share certain characteristics, clustering almost always reduces sampling efficiency.