When two sets of data are compared, how do you decide whether they are really different or whether any differences are just attributed to chance? The author describes how to address thius using the t-test.