Which sampling method divides the population into subgroups and samples within each subgroup?

• A. Simple random sampling
• B. Stratified sampling
• C. Systematic sampling
• D. Cluster sampling

Answer: (B)

Dividing the population into subgroups that are similar on a characteristic and then sampling from each subgroup ensures representation of every subgroup and often improves precision when subgroups differ from one another. This approach, known as stratified sampling, is particularly helpful when you want comparisons across subgroups or accurate overall estimates that reflect the whole population. For example, surveying customers across a country by region or by age group, and then drawing samples from each region or age group, guarantees that each subgroup contributes data. Sampling within each subgroup reduces the risk that a random sample would over- or under-represent any one group and can increase accuracy when within-subgroup variation is smaller than between-subgroup variation. You can allocate samples proportionally to the subgroup's size or allocate equally to give each subgroup the same weight. By contrast, simple random sampling pulls from the entire population without enforcing subgroup structure, systematic sampling selects every nth unit, and cluster sampling divides units into clusters and then samples within clusters, which does not require sampling from every subgroup.