A random sampling method where the population is divided into natural groupings (clusters) of people, for example, your school or college, or the neighbourhood within which you live. Clusters to include in the study are randomly selected and either a census is conducted on all the people in the selected clusters, or a number of individuals within each cluster selected to be surveyed at random. (The latter is known as multi-stage sampling). Cluster sampling does not require a complete list of the population – only a list of the population within each cluster. This helps keep down preparation time and costs. Travel and administration costs are also kept down as samples are not spread out geographically.