NONPROBABILITY SAMPLING
    Reliance on Available Subjects
    Purposive or Judgmental Sampling
    Snowball Sampling
    Quota Sampling
    Selecting Informants

    Probability sampling, sometimes called random sampling, is the technique used in the 1948 election polls and subsequently.  Often, however, this technique is not possible or appropriate.  This section deals with a number of nonprobability sampling techniques.

    Some studies rely on available subjects, such as religious cult members attending a group event, tourists at the airport to leave or arrive at a vacation area, etc.

    In qualitative research, it is often appropriate to make a purposive or judgmental selection of subjects or informants, choosing people that represent different subgroups, for example.  In the religous cult example, you might choose some of the leaders, some of the active followers, and some not so active to insure that you would be sure to tap into the variations among members.

    In studying a relatively rare kind of person, you may use snowball sampling in which you ask those subjects you know to identify others for you to study--and to ask them for futher referrals.

    Quota sampling, the technique used successfully by Gallup in 1936 and unsuccessfully in 1948, attempts to select subjects from all the different kinds of people in the population and then weight those studied on the basis or how common or rare their type is.

    Qualitative researchers, working within the culture or subculture they wish to study, often use informants: people who provide information about the culture rather than just about themselves.  We'll look at some guidelines for choosing informants.