Primary Sampling Unit (PSU)

In sample surveys, primary sampling unit (commonly abbreviated as PSU) arises in samples in which population elements are grouped into aggregates and the aggregates become units in sample selection. The aggregates are, due to their intended usage, called “sampling units.” Primary sampling unit refers to sampling units that are selected in the first (primary) stage of a multi-stage sample ultimately aimed at selecting individual elements.

In selecting a sample, one may choose elements directly; in such a design, the elements are the only sampling units. One may also choose to group the elements into aggregates and choose the aggregates in a first stage of selection and then elements at a later stage of selection. The aggregates and the elements are both sampling units in such a design. For example, if a survey is selecting households as elements, then counties may serve as the primary sampling unit, with blocks and households on those blocks serving as the sampling units in subsequent sampling stages of the survey. Thus, sampling unit is a term that combines sample selection with the units used as potential selections at any point in the sampling process. For example, in a systematic selection of elements from a list, the elements are the sampling units. The sampling unit contains only one element. But in order to reduce cost of sample selection and data collection, samples may be drawn in multiple stages. Elements are grouped into aggregates (e.g. households on blocks in counties), and a sample of aggregates (e.g. counties) is selected at the first stage. Later elements (e.g. households on blocks) are selected from elements in selected aggregates (e.g. blocks). The aggregates selected at the first stage (e.g. counties) are called “primary sampling units.” In multi-stage samples, another level of aggregation may occur within selected primary sampling units, and the aggregates at the second stage (e.g. blocks) are selected, that is, the second stage units. The procedure of creating and [Page 611]selecting aggregates within previously selected aggregates may proceed through several levels of a hierarchy until, at the last stage, the individual element (e.g. households) is selected. The selected elements are chosen only within already chosen aggregates within the hierarchy.

Primary sampling units are not necessarily formed from a list of elements or individuals. They may be conceptual representations of groups for which lists could be obtained, one from each primary sampling unit. For example, suppose a sample of adults living in a large country with widespread population is to be selected, and no list of adults of sufficient quality (e.g. good coverage, covering nearly all adults) is available. Aggregation of adults on an existing list cannot take place. However, units can be formed that are for all intent aggregates of adults within them. A common procedure, area sampling, uses geographic areas at successive levels of a geographic hierarchy as sampling units, with implicit aggregation of adults (or other elements) who usually reside within them.

For example, in the United States, a common area primary sampling unit is the county. Within counties, geographic areas such as towns or townships or other administrative units defined by geographic borders can be identified. Within towns or other administrative units, the geography may be further divided into smaller units, such as city blocks or enumeration areas with boundaries formed by streets or highways, rivers, and other relatively permanent, readily identified features. These blocks or enumeration areas are created by a government statistical agency for the purpose of providing census counts of various types of units within them.

...