A relationship between two variables is known as a correlation, or association. A correlation coefficient, or statistic, is a measure of the degree of linear association between paired data values. This article describes some of the causes of correlation, and discusses how correlation can be represented in a project model. Situations that cause correlation include direct cause-and-effect relationships, common drivers, and system constraints. Correlation-modeling methods include the use of randomly sampled data, probability tables, correlation coefficients, and detailed modeling. Human factors complicate the picture, but a decision analysis approach is the best one, even when some of the inputs are highly subjective.