Step 2: Selecting suitable methods of multiple linear regression
There are three methods in MLR which are forward selection, backward elimination and stepwise regression. All
three methods can be categorized into stepwise-type procedures. In this research, only stepwise regression method
was applied. Stepwise regression method is a combination of forward selection and backward elimination. By
Intan Martina Md Ghani and Sabri Ahmad / Procedia Social and Behavioral Sciences 8 (2010) 549–554 551
referring Minitab Methods and Formulas, standard stepwise regression both adds and removes controlled variables
as needed for each step. Minitab ended its procedure when all variables not in the model have p-value that are less
than the specified Alpha-to-Enter value and when all variables in the model have p-value that are greater than or
equal to the specified Alpha-to-Remove value. Based on Minitab StatGuide, Alpha-to-Enter is a value that
determines if any of the predictors that not currently in the model should be added to the model. While Alpha-toRemove
is a value that determines if any of the predictors in the model should be removed from the model.