Sampling refers to the practice of selecting a subgroup of people from a defined population to include them in a study. In qualitative research, the aim is to select people who can provide different perspectives and experiences relevant for the study topic. Including different, even contradictory, viewpoints is important to generate a complex understanding and ‘information-richness’ about the research topic.
Take, for example, a study in which a researcher aims to understand the impact of war and trauma on the mental health and wellbeing of refugee women. To gain in-depth insight into the women’s situation, the researcher might want to include people with different perspectives on survivors’ mental health in the study. These could include:
- women survivors of war
- family members
- humanitarian aid providers
- mental health and psychosocial support workers
- community leaders and organisers
Creating a sample like this is referred to as ‘purposeful sampling’. When creating a purposeful sample, researchers pre-define which types of participants can provide valuable views for the research objective. In other words, the ‘researcher exercises his or her judgment about who will provide the best perspective on the phenomenon of interest, and then intentionally invites those specific perspectives to the study’ (Abrams, 2010, p. 538).
Key assumptions about qualitative samples:
- the sample is usually relatively small, ranging from five to 20 participants (it can be smaller or larger than this depending on the study)
- people included in the sample are not assumed to share the same experiences or viewpoints. In other words, they cannot ‘stand-in’ for one another
- the particular experiences and perceptions of participants are of interest. So-called ‘outliers’ or ‘special cases’ are insightful as they can add complexity and depth.
Assessment of an Optimal Qualitative Sample
To assess whether your sampling procedure is optimal, you can ask yourself the following questions:
- Is the sampling strategy relevant to the conceptual framework and questions addressed by my research?
- Is my sample likely to generate rich information on the type of phenomena which need to be studied?
- Is it likely that my sample produces believable descriptions and explanations?
- Is my sample ethical?
- Is my sample feasible?
[Questions adapted from Miles and Huberman (1994)]
In summary, the approach to sampling must match your research objective and questions as well as your overall study design. It needs to be carefully developed for valuable data to emerge as part of your research.
(Author: Hanna Kienzler)