Which sampling technique is most appropriate when the population is heterogeneous and you need proportional representation across subgroups?

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

Answer: (D)

When the population is diverse, you want to make sure every important subgroup shows up in the sample in the same proportion as in the population. That’s what stratified sampling does: you split the population into homogeneous subgroups, or strata (like age groups, gender, or income levels), and then draw a sample from each stratum in proportion to its size in the population. By doing this, the final sample mirrors the population’s mix, so estimates reflect the true structure and you reduce sampling error from any one subgroup being over- or underrepresented.

This approach is better for heterogeneity than simple random sampling, which treats all units the same and can stumble into uneven subgroup representation by chance. Systematic sampling selects units at regular intervals, which might miss or misrepresent subgroups if there’s any pattern in the list. Cluster sampling focuses on whole groups and can be efficient, but it may not capture the required proportional representation across subgroups unless you use many clusters and still maintain the balance. Stratified sampling explicitly guarantees representation across subgroups by design, and often yields more precise, generalizable results.