statistical analysis that defines the variation in one variable by the variation in another, without establishing a cause-and-effect relationship. The coefficient of correlation is a measure of the strength of the relationship between the variables; that is, how well changes in one variable can be predicted by changes in another variable. For example, correlation can be shown between the frequency with which a commercial is aired and sales volumes by plotting on a graph the values of each. A line drawn through the plotted points defines the correlation algebraically. The greater the density of the points around the line, the greater the strength of the correlation. In example I, the correlation is high; in example II, the correlation is low. Although the correlation may be high between advertising exposures and sales, other factors could be the cause, such as the supply of competitive products, availability of the product in stores, and so forth.
