Purposive Sample

A purposive sample, also referred to as a judgmental or expert sample, is a type of nonprobability sample. The main objective of a purposive sample is to produce a sample that can be logically assumed to be representative of the population. This is often accomplished by applying expert knowledge of the population to select in a nonrandom manner a sample of elements that represents a cross-section of the population.

In probability sampling, each element in the population has a known nonzero chance of being selected through the use of a random selection procedure. In contrast, nonprobability sampling does not involve known nonzero probabilities of selection. Rather, subjective methods are used to decide which elements should be included in the sample. In nonprobability sampling, the population may not be well defined. Nonprobability sampling is often divided into three categories: purposive sampling, convenience sampling, and quota sampling.

An example of purposive sampling would be the selection of a sample of universities in the United States that represent a cross-section of U.S. universities, using expert knowledge of the population first to decide with characteristics are important to be represented in the sample and then to identify a sample of universities that meet the various characteristics that are viewed as being most important. For example, this might involve selecting large, medium, and small public and private universities. This process is referred to as two-stage sampling, but the first-stage units are not selected using probability sampling techniques. This kind of two-stage sampling should not be confused with stratified sampling, which is a probability method of sampling. Instead, in this example of purposive sampling, such “strata” (size of university and type of university) are used without the scientific rigor of probability sampling. Nonetheless, they are used to assure a more representative mix of elements than may otherwise occur if the expert were not explicitly choosing universities of different sizes and types.

Another example of purposive sampling could be the selection of a sample of food stamp offices from which participants will be sampled. Here, the first-stage units are selected to represent key food stamp recipient dimensions, with expert subject matter judgment used to select the specifie food stamp offices that are included in the study.

One limitation of purposive sampling is that another expert would likely come up with different sampled elements from the target population in terms of important characteristics and typical elements to be in the sample. Given the subjectivity of the selection mechanism, purposive sampling is generally considered most appropriate for the selection of small samples often from a limited geographic area or from a restricted population definition, when inference to the population is not the highest priority.