Statistical sampling is the process of finding a representative portion of a large population. In quality control, a percentage of products will be selected at random for testing, and the results extrapolated to the whole. For example, out of 100,000 cars produced, 5 will be crash-tested. The results will be assumed to hold for the cars actually sold.
In public opinion polls, the people who are contacted must be carefully selected to avoid several kinds of selection bias.