Populations and Samples

The $\textbf{population}$ is the whole set of items or people you are interested in. A $\textbf{sample}$ is a smaller selection taken from it. The individual members are $\textbf{sampling units}$, and a list of them is the $\textbf{sampling frame}$.

A $\textbf{census}$ measures every member of the population. It is completely accurate but is often expensive, slow, and sometimes impossible (e.g. testing every item destroys them). A $\textbf{sample}$ is cheaper and faster but may not perfectly represent the population.

Samples save time and money and are essential when testing is destructive. The downside is potential bias and the fact that conclusions are estimates. A good sample is large enough and chosen to fairly represent the population.