A subset of a population. By carrying out a random survey of some school pupils for example, the pupils you survey would make up the sample, and all the pupils in that school would make up that population.
In statistics, samples are used to make inferences (estimated conclusions) about a larger population without having to survey the whole population.