What is stratified sampling and when would you use it in surveillance?

• A. Divide population into strata and sample from each to ensure representation across subgroups (e.g., age groups, geographic regions).
• B. Randomly sample from the entire population without considering subgroups.
• C. Select cases only from the largest subgroup.
• D. Use convenience sampling.

Answer: (B)

Stratified sampling divides the population into subgroups that share a characteristic (strata) and then samples from each subgroup to ensure representation across those subgroups. This approach is especially useful in surveillance when you need accurate estimates for specific subgroups or when outcomes differ across groups, such as different age ranges or geographic regions. By sampling from every subgroup, you prevent the results from being dominated by the larger groups and improve the precision of comparisons between subgroups. The other described methods don’t guarantee that representation: sampling from the whole population without considering subgroups can miss important differences, selecting only from the largest subgroup ignores diversity, and convenience sampling often introduces bias because it relies on what’s easiest to obtain rather than what’s truly representative.