Jenis-jenis Pengambilan Sampel pada Riset Kualitatif
Creswell, J. W. (2007). Qualitative inquiry & research design: Choosing among five approaches (2nd ed.). Thousand Oaks, CA: SAGE. Publications.
Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook. 1994. Beverly Hills: Sage Publications.
- Homogeneous samples (Holloway and Wheeler, 1996; Patton, 2002; Robson, 2002) chosen to give a detailed picture of a particular phenomenon – for example, individuals who belong to the same subculture or have the same characteristics. This allows for detailed investigation of social processes in a specified context.
- Heterogeneous samples (Holloway and Wheeler, 1996; Robson, 2002) or maximum variation sampling (Patton, 2002) where there is a deliberate strategy to include phenomena which vary widely from each other. The aim is to identify central themes which cut across the variety of cases or people.
- Extreme case or deviant sampling (Patton, 2002; Robson, 2002). Cases are chosen because they are unusual or special and therefore potentially enlightening. The logic is that learning about phenomena is heightened by looking at exceptions or extremes (for example, ethnomethodologists sometimes use deviant sampling to expose implicit assumptions and norms).
- Intensity sampling (Patton, 2002) which employs similar logic to extreme or deviant case sampling but focuses on cases which strongly represent the phenomena of interest rather than unusual cases.
- Typical case sampling (Patton, 2002). Cases which characterize positions that are ‘normal’ or ‘average’ are selected to provide detailed profiling. This requires prior knowledge about overall patterns of response so that what is ‘typical’ is known (for example, participants might be selected from their responses to a survey).
- Stratified purposive sampling (Patton, 2002), a hybrid approach in which the aim is to select groups that display variation on a particular phenomena but each of which is fairly homogeneous, so that subgroups can be compared.
- Critical case sampling (Patton, 2002) in which cases are chosen on the basis that they demonstrate a phenomenon or position ‘dramatically’ or are pivotal in the delivery of a process or operation. The logic is that these cases will be ‘critical’ to any understanding offered by the research. Patton sees this approach as particularly valuable in evaluative research because it helps to draw attention to particular features of a process and can thus heighten the impact of the research.