Nested Sampling

Nested sampling (or double sampling) involves selecting a sample (first phase) where one set of variables is measured followed by selecting a sub-sample (second phase) of the first sample where an additional second set of variables is measured. In most cases the second set of variables are "expensive" to measure compared to the first set of variables. The purpose of double sampling is to obtain estimates for the second set of variables with lower sampling error by using relationships between variables in the first set and variables in the second set. Note that it is possible to use more than two phases of sampling. Nested sampling is also used to denote two-stage (or multi-stage) sampling where the higher stage units are "nested" within the lower stage units. In two-stage sampling the first stage units (primary sampling units) consist of secondary sampling units at the second stage. A term used in two somewhat different senses: (1) as equivalent to multi-stage sampling because the higher stage units are "nested" in the lower stage units; (2) where the sampling is such that certain units are imbedded in larger units which form part of the whole sample, e.g. the entry-plots of clusters are "nested" in this sense. OECD Source: A Dictionary of Statistical Terms, 5th edition, prepared for the International Statistical Institute by F.H.C. Marriott. Published for the International Statistical Institute by Longman Scientific and Technical.