Stratified Sampling

Stratified sampling first splits the population into mutually exclusive subgroups that matter to the research, such as year of study, sector, or region. A random sample is then drawn within each stratum, either in proportion to its share of the population or in equal numbers when small groups need enough cases for comparison. This guarantees that key subgroups appear in the sample rather than leaving their representation to chance, and it usually produces more precise estimates than simple random sampling.

In a thesis, stratified sampling is worth reporting carefully: state the stratification variable, why it was chosen, and whether allocation was proportional. It signals to examiners that you anticipated variation in your population instead of discovering it too late. It also supports planned subgroup comparisons in your results chapter, because each group has been deliberately sampled.