# Correlation Analysis

Go Back

Definition

Correlation Analysis compares two variables by plotting paired values. One variable is plotted on the “Y” axis (vertical or ordinate scale). The other variable is plotted on the “X” axis (horizontal or abscissa scale). This plot, called a scatter-plot, helps identify if a relationship exists between the two variables.The strength of the relationship is assessed by looking at how much scatter there is in the paired data points. A random scatter indicates no relationship.

Examples

A value called the correlation coefficient is used to quantify the degree of correspondence between the two variables. A coefficient of 0 indicates no relation, +1 indicates a perfect positive relation and -1 is a perfect negative relation. Thus the correlation value (r) varies between -1 and +1. Values closer to the extremes imply stronger associations.

Note: Although the scatter plot might suggest a non linear relationship, the correlation coefficient only measures linear association.