Sample

In survey research, a sample is a subset of elements drawn from a larger population. If all elements from the larger population are "sampled" for measurement, then a census is being conducted, not a sample survey.

In a broad context, survey researchers are interested in obtaining some type of information for some population, or universe, of interest. A sampling frame, that is, a frame that represents the population of interest, must be defined. The sampling frame may be identical to the population, or it may be only part of the population and is therefore subject to some under-coverage, or it may have an indirect relationship to the population (e. g. the population is males and the frame is telephone numbers). It is sometimes possible to obtain the desired information from the entire population through a census. Usually, however, for reasons of cost and time, survey researchers will only obtain information for part of it, referred to as a sample of the population. There may be several different samples selected, one for each stage of a multi-stage sample. For example, there may be a sample of counties, a sample of blocks within sampled counties, a sample of addresses within sampled blocks, and a sample of persons from sampled addresses.

A sample can be obtained in many different ways, as defined by the sample design. Survey researchers usually will want to have a probability sample, which ensures that all units in the frame have a known nonzero probability of selection, rather than a convenience sample or nonprobability sample, which do not sample respondents with known nonzero probabilities of selection.