This project was funded by the Nuffield Foundation, but the views exp ressed are those of the authors and not nec essarily those of the Foundation.
A Sample is a selection of data chosen from all of that possibly available. Sampling is needed in
almost all forms of data collection as in most research processes it is simply not possible to gain
data from every available source. For example, if one wished to conduct some interviews within a
town it would not be possible to interview absolutely every resident of that town. Instead, a
selection of the population would be used to try to get a representation of the town’s answers. The
method you use to select this sample is known as your Sampling Technique.
Sampling not only makes conducting your data collection possible, it can also make more efficient
use of the time you have to collect your data. It may not even be necessary to collect all points of
data for your research, and in many cases your overall population size (or your Sampling Frame –
the pool of data from which you are drawing a sample) may be an unknown quantity; as a
researcher you would therefore not know when to stop collecting data.
Researchers frequently put great effort into deciding on the size of their data sample. The larger
the sample size, the more representative it is likely to be of your overall sampling frame, and as a
result, the more justifiable your conclusions will be. However, the size of any sample is also
dependent on the time and resources you have available and how manageable the data collection,
and data analysis will be as a result.
While there are many different sampling techniques, there are three common methods used
frequently by researchers: Random, Systematic and Stratified Sampling. In the following
explanations, two examples will be used. In the first, Researcher A is trying to find a sample of ten
data collection locations on a map that contains three different geographical zones. In the second,
Researcher B is trying to select a sample of ten interviewees from a population of twenty people of
different ages.
______________________________________________________________________________
Random Sampling
Random sampling is where sources of data are chosen in a completely haphazard way. Once the
size of the sample has been decided (maybe as a percentage of the overall sampling frame),
researchers use random number generators, which can be found online, to give completely
random sets of numbers. These can then be used to create grid references for data collection sites
on a map or tell researchers which house numbers to survey within a street.
2g – A Guide to
Sampling
Techniques
