A technique used by auditors to draw conclusions about the accuracy of a larger population based on a sample of data.