Sample size determination is an important step in IR as it informs how many people you need to approach to participate in your research.

Table 14: Sampling techniques used in qualitative research18
Kinds of sampling Explanation
Convenience Studies the units that are available at the time of the research. It is more convenient than a random sample because the researcher uses what is available, rather than what is selected. There is however, a risk of measurement bias. If interviewing households in the morning is most convenient, which populations might be overrepresented (housewives, elderly) and which may be underrepresented in the sample (employed, men, students)?
Purposive Used when the elements are selected based on the researcher’s judgment regarding the desired information being collected. For example, researchers may decide to identify respondents according to their involvement in a particular health programme.
Maximum variation Selects units that represent as wide a range of variation as possible (e.g. gender, socioeconomic status, population density, etc.).
Snowball Identifies a few people who will be involved with the study and then asks them to identify more people who would be relevant to include in the research. Best to start with at least two individuals so as to reach different networks of individuals. This is the most common form of sampling in qualitative research methods.
Contrasting cases Involves two or more units with distinct characteristics so that comparisons can be made when explaining problems and understanding the factors that influences them. For example, researchers may decide to study individuals living in a site where a health programme has been successful and another site where the programme has been less optimal.