RECOUP/Qualitative research/Handout on Sampling in Qualitative Research

Sampling in qualitative research

Or What can you say if you don't have a large random sample?

"More discoveries have arisen from intense observation than from statistics applied to large groups." (W. I. B. Beveridge, 1951).

1. Choosing between large quantitative surveys and small-scale qualitative studies is not Either/Or: good social science depends on Both/And. Good social science employs those methods that best help answer a given research question and purpose. Often, a combination of qualitative and quantitative methods will do the task best.
2. The advantage of large samples is breadth, whereas their problem is one of limited depth. For the case study, the situation is the reverse.
3. For research using qualitative methods, we are interested in depth, especially for
   1. Issues that are too complex to be covered in a survey or a census
   2. Issues that are too sensitive for a simple question/answer approach
4. A random sample may not the most appropriate strategy, because the typical of average case may not always be the richest in information. Atypical or extreme cases often reveal more about a social issue
5. To clarify the deeper causes behind a given problem and its consequences than to describe the symptoms of the problem and how frequently they occur. Rather, we need some deeper insights by spending more time on an issue, and gathering information on the full context in which it occurs.

Sampling

Random and stratified sample are not the only kinds that can be used in social science: quota sampling is often used in market surveys, and strategic sampling, or the choice of cases to study, is also common, especially in qualitative projects.

Layers of 'Case' in qualitative projects:

• • Selection of sites for intensive study (e.g. a village or an urban neighbourhood)
  • Selection of households/individuals for collecting contextual information
  • Selection of sample individuals etc for more sustained data collection

Most useful for qualitative studies:

Type of Strategic Selection	Purpose
	To maximize the utility of information from small samples and single cases. Cases are selected on the basis of expectations about what they can tell us.
1. Extreme/deviant cases	To obtain information on unusual cases, which can be especially 'problematic' or especially 'good'.
2. Maximum variation cases	To obtain information about the significance of various circumstances for processes and outcome (e.g. three to four cases that are very different on one dimension: e.g. largest, median and smallest size; government, aided, not-for-profit and commercial funding patterns; city, town and rural area).
3. Critical cases	To test a hypothesis by choosing the case that permits logical deductions of the type, "If this is valid for this case, then it should apply to all cases." Or "If it is not valid for this case, it is unlikely to be valid for any other cases".

---

Singal, N., and Jeffery, R. (2008). Qualitative Research Skills Workshop: A Facilitator's Reference Manual, http://oer.educ.cam.ac.uk/wiki/RECOUP, Cambridge: RECOUP (Research Consortium on Educational Outcomes and Poverty, http://recoup.educ.cam.ac.uk/). CC BY-NC-SA 4.0. (original page)