Linear regression is used for:

- Determining if measurement variables are associated. There is a dependent variable (
*y)*and one or more explanatory/independent variables (*x*) (e.g.: blood pressure and drug intake) - Measuring the strength of this association. This strength is indicated by the r
^{2}value, which is a value between 0 and 1. The higher r^{2}is, the stronger the relationship between the measurement variables is. - Finding an equation describing the relationship between the measurement variables, which can then be used for prediction purposes. This equation finds the line that fits the cloud of data points the best. However, there’s various definitions of ‘best fit’, which means there’s also different corresponding methods of linear regression.

You can read more about correlation and linear regression here.

Source: ‘Handbook of Biological Statistics’ by John H. McDonald