Inferential statistics use laws of probability to make inferences about a population based on information gleaned from a sample. Such statistics are particularly useful given the unlikely occurrence that a researcher has access to an entire population. The population is the larger group of interest from which a sample is drawn. For example, if we wanted to study education levels among women incarcerated in state prisons, all female state prisoners comprise the population. Studying all the women would be time consuming and expensive. Instead, the researcher could randomly choose a sample of one hundred women from this group and use inferential statistics to suggest results to the entire population. There are always some differences between the characteristics of a sample and the characteristics of a population simply because of natural variation, referred to as sampling error which is always estimated in inferential statistics.
(from Oxford Reference Online)