Sampling a population of interest
A topic in research methodology
Sampling is usually discussed in terms of data collection. It refers to which members of a population to collect data from, or how to select data from some canon (e.g. if analysing examination questions from an archive of examination papers going back decades, offering too many for close consideration of each question).
Read about 'populations' in research
Sampling may also be considered at the data analysis stage if data collection has produced too much data to analyse within the resources of a project, or some data is clearly redundant.
Ethical considerations suggest we should not deliberately collect data that will not be analysed, but when we are unsure how much data may be generated we may have to err on the side of caution. We may need to plan for possible drop-out of participants that does not in the event occur, for example. We may be selecting a case for case study depending upon the quality of data available from different participants or institutions.
Two approaches to sampling
"There are two general strategies or logics for selecting units (organizations, events, people, documents, locations, etc.) to study in qualitative work: an empirical or statistical strategy and a theoretical or purposive strategy….
In the logic of sampling based on an empirical or statistical strategy, sample units are chosen based on their representativeness of some wider population of units. …
In the logic of sampling based on a theoretical or purposive strategy, units are chosen … for their relevance to the research questions, analytical framework, and explanation or account being developed in the research."
Schwandt, 2001: 232
The sampling approach must be coherent with the rest of the research design.
In confirmatory research, using statistical approaches it may be important fro results to be valid that sampling has a random element so that all members of some population have the same chance of being included in the sample (Taber, 2019).
In educational studies, it is not unusual to see studies that are nominally about some specified population, where there is no assurance that the sample is representative of the population.
Read about some examples of populations sampled in research studies
In discovery modes of research, theoretical grounds may be used to select the (purposive) sample which offers the most information in relation to our research questions. For example, in grounded theory methodology, a general iterative strategy of 'theoretical sampling' is used during the research.
Sources cited:
- Schwandt, T. A. (2001). Dictionary of Qualitative Inquiry (2nd ed.). Thousand Oaks, California: Sage Publications.
- Taber, K. S. (2019). Experimental research into teaching innovations: responding to methodological and ethical challenges. Studies in Science Education. doi:10.1080/03057267.2019.1658058)
My introduction to educational research:
Taber, K. S. (2013). Classroom-based Research and Evidence-based Practice: An introduction (2nd ed.). London: Sage.
