What is the primary characteristic of cluster sampling?

Cluster sampling is a method that involves dividing the population into separate groups, known as clusters, and then sampling entire clusters rather than individuals. The primary characteristic of cluster sampling is that it often requires sampling in multiple stages. This means that initially, a selection of clusters is randomly chosen, and then all individuals within the selected clusters may be included in the sample or a further sampling may occur within those clusters.

This approach is particularly useful when the population is widespread geographically or difficult to list comprehensively. It allows researchers to effectively manage sampling logistics and resource allocation while still obtaining a representative sample of the entire population. The multiple stages can involve selecting which clusters to include and then determining the sampling method within those clusters.

In contrast, the other options don't capture the essence of cluster sampling as accurately. For example, single-stage random selection applies more to simple random sampling methods rather than the structured approach in cluster sampling. Sorting the population into categories before sampling aligns more with stratified sampling rather than cluster sampling. Focusing exclusively on homogeneous populations contradicts the idea that clusters are often more diverse and collectively represent the overall population well.