Division of a sample of observations into a number of classes, together with the number of observations in each class. It serves to summarize the main characteristics of the data, such as location, shape and dispersion. In essence, it is the empirical equivalent of the probability distribution of a random variable.