Single units from a category (e.g. human, animal, farm) selected to be representative of the category as a whole.

A list of all sampling units in the source population.

The basic element(s) of the population we are sampling. This is often the case when sampling is conducted in multiple steps, or stages. For example, we may have: 

• hospitals (primary sampling unit) 
• wards (secondary sampling unit) 
• individual patients (tertiary sampling unit).

The actual population from which eligible study subjects are drawn, such as hospitalised patients at a single hospital in the capital city. Ideally, the source population should be the same as the target (reference) population, but this is often not practical.

The units selected from the source population; for example, 40 patients randomly selected from inpatients on a surgical ward over a defined time period. The study sample should be similar to the source population.