Purposive Sampling

To say one will engage in purposive sampling signifies that one sees sampling as a series of strategic choices about with whom, where, and how one does one's research. This statement implies that the way that researchers sample must be tied to their objectives. A second implication follows from the first: There is no one best sampling strategy because which is best will depend on the context in which researchers are working and the nature of their research objective(s).

Purposive sampling is virtually synonymous with qualitative research. However, because there are many objectives that qualitative researchers might have, the list of purposive strategies that may be followed is virtually endless, and any given list will reflect only the range of situations the author of that list has considered.

Nevertheless, some specific objectives and interests characterize qualitative research. For one thing, qualitative researchers are less often interested in asking about central tendency in a larger group (e.g., “What do most people in this population think about an issue?”), and much more interested in case study analysis—why particular people (or groups) feel particular ways, the processes by which these attitudes are constructed, and the role they play in dynamic processes within the organization or group. Embedded in this is the idea that who a person is and where that person is located within a group is important, unlike other forms of research where people are viewed as essentially interchangeable. Research participants are not always created equal—one well-placed articulate informant will often advance the research far better than any randomly chosen sample of 50—and researchers need to take this into account in choosing a sample.

The general theme here is that the biggest questions all researchers need to ask themselves are what they want to accomplish and what they want to know. The appropriate sampling strategy will follow from that. Some examples of the kinds of purposive alternatives available include the following:

Stakeholder Sampling: Particularly useful in the context of evaluation research and policy analysis, this strategy involves identifying who the major stakeholders are who are involved in designing, giving, receiving, or administering the program or service being evaluated, and who might otherwise be affected by it.

Extreme or Deviant Case Sampling: Sometimes extreme cases are of interest because they represent the purest or most clea-cut instance of a phenomenon researchers are interested in. For example, if researchers were interested in studying management styles, it might be most interesting to study an organization that did exceptionally well and/or exceptionally poorly.

Typical Case Sampling: Sometimes researchers are interested in cases simply because they are not unusual in any way. For example, years ago Howard Becker and some of his colleagues were interested in studying how medical students were socialized into the profession. They did their research at the University of Kansas Medical School precisely because there was nothing unusual about it, and for that reason it was probably typical of the medical school experience.

Paradigmatic Case Sampling: A case is paradigmatic when it is considered the exemplar for a certain class. For example, if one wanted to study the management of professional sports teams, the paradigmatic case in hockey of a successful franchise would be the Montreal Canadians; for baseball it would be the New York Yankees.

...