Assumptions, Limitations, Practical Considerations - Normality Assumption
It is assumed in multiple regression that the residuals (observed minus predicted values) are distributed normally (i.e., follow the normal distribution). Again, even though most tests (specifically the F-test) are quite robust with regard to violations of this assumption, it is always a good idea, before drawing final conclusions, to review the distributions of the major variables of interest. In Multiple Regression you can produce a histogram (with the normal curve superimposed) from within the descriptive statistics spreadsheet that is available at all times via the Descriptive statistics option of the Basic Statistics and Tables module. Also, you can produce histograms for the residuals as well as normal probability plots in order to inspect the distribution of the residual values.