Sampling Methods

Key points

• • A sample is a subset of a population chosen for study; a good sampling method makes the sample represent the whole population so findings can be generalised.
• • In probability (random) sampling, every person has a known, non-zero chance of being chosen (not always an equal one), which reduces selection bias and lets results be generalised with more confidence.
• • Simple random sampling picks people entirely by chance (as in a lottery), while systematic sampling takes every nth person from a list.
• • Stratified sampling first splits the population into groups (strata), such as age bands, then samples from each, to make sure every group is represented.
• • Cluster sampling picks whole groups (such as villages or schools) at random, which is practical when the population is spread out.
• • Non-random sampling (such as picking whoever is convenient) is easier but risks a biased, unrepresentative sample.