Sampling is the process of collecting units (e.g., people) that represent a bigger population of interest. The accessible population is the portion of the population that the researcher can access and sample from, otherwise referred to as the study population. Finally, the group of people or objects that are selected from the accessible population is the sample. The number of cases in the accessible population is represented by an uppercase N, while the number of cases in the sample is represented by a lowercase n.
Cluster sampling involves two layers of sampling; the first layer selects clusters, or naturally occurring groups (e.g., yoga classes), and the second layer selects individuals from within those clusters (e.g., students of those yoga classes). Researchers can assess differences between classes as a whole (i.e., the cluster) or differences between individual students. Whatever major entity is being assessed in the results is referred to as the unit of analysis. For example, if a study compares student performance across different students, the unit of analysis is the student. If a study compares average student performance across different classes, the unit of analysis is the class.