3.2.3—Sampling

A sample is a subset of the target population. Random, systematic, quota, stratified and convenience methods trade representativeness, cost, access and control in different ways.

Define the population, sampling frame, sample size and selection rule before collecting data. Bias can arise when accessible people differ from the population.

A school survey taken only from volunteers may over-represent highly engaged pupils; stratifying by year group and using a consistent selection rule can improve coverage.

A large sample is not automatically representative, and a small purposive sample may be appropriate for exploratory interviews.