What is Purposive sampling?

Finding my sample

Finding the right people at the right time is crucial in collecting data that is usable, viable and valuable. In this post I want to discuss the importance of developing a clear sampling strategy, and why understanding and articulating the decisions you make at this early stage are essential for any research papers, book chapters, articles and dissertations you may wish to write in the future. This paper is a brief continuation of my previous article on what is sampling?

Just to recap in a previous blog post I discussed the crucial question on  precisely “what is sampling?”  and why its important to any research. As I mention in the post sampling is critical in helping you to identify your sampling population, in other words who it is you are trying to interview and why that individual is crucial to your research. Today’s blog post is to look at sampling in more detail, primarily the different methods we can utilise to undertake sampling. One of the more popular methods is called “purposive sampling”.

Purposive sampling in qualitative circles is also referred to as “non-probability” sampling. It is called this due to the researcher applying their own criteria when defining their sample, so in layman terms the researcher selects their own individuals as part of their study. For a more academic definition Ritchie et al., (2003) defines this sampling approach as a strategy where “Members of a sample are chosen with a purpose to represent a location or type in relation to the criterion” (p. 77). So in summary the researchers choose  specific people within a sample population. This is in contrast to random studies where elements of your sample are known and you may deliberately include multiple criteria such as gender, ethnicity, culture etc.

The main advantage of such an approach is the ability to critically think and define the parameters of the population that is intended to be studied at an early stage (Miles & Huberman, 1994). So for example when defining your parameters you may want to only talk to female students within the ages of 16-24, as you feel they are more appropriate to your study. This approach can make sampling easier, but you will need to spend some time  justifying some of your criterion later in your research.

However as with all things there are disadvantages also, the main issue I have found in the past, especially with my undergraduate students is the researcher bias. In some scenarios (not all) students will interview their friends/colleagues/acquittances for an easy life, but thankfully this is always in a minority. This happens for numerous reasons but the ability to create a sample based entirely on your own judgement where you are trying to answer your own research question, can be a pressurised environment, so in these cases your sample may become a little bit skewed. This then leads to the second issue, defending your sample! If your research ever gets to a stage where there is potential to publish then be ready to defend your sample and by extension your findings to your critics. In scenarios like this non-probability sampling can be difficult to defend.

Anyway I hope this blog post has helped you to understand the relative merits of sampling and why it can be useful. Whatever you decide to do good luck!

References

Ritchie, J., Lewis, J., & Elam, G. (2003). Designing and selecting samples. In J. Ritchie & J. Lewis (Eds.), Qualitative research practice: A guide for social science students and researchers (pp. 77–108). Book Section, London: Sage.

Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook. Book, Sage.