Cluster Sample vs Stratified Random Sample
Both terms below come up in the same part of the course, and students mix them up. Here is each one defined on its own, side by side, so you can see where they part company.
Cluster sample
Collecting data and study design
A cluster sample divides the population into groups called clusters, randomly selects whole clusters, and includes every individual in the chosen clusters.
Cluster sampling is used when the population naturally falls into groups that are each meant to resemble the whole population. For example, to survey a school district you might randomly choose 5 schools and then survey every student in those 5 schools. Unlike strata, clusters are ideally diverse within and similar to one another, and you sample entire clusters rather than individuals. It is often cheaper because you only travel to a few locations.
Stratified random sample
Collecting data and study design
A stratified random sample splits the population into similar groups called strata, then takes a separate simple random sample from each stratum.
Stratifying groups individuals who resemble each other on something relevant, then samples within every group, which can give more precise estimates than a plain simple random sample. For example, you might split students into grade levels and draw a random sample from each grade so every grade is represented. The strata should be internally similar but different from one another. You then combine the results across strata to estimate the population value.