Multiple Regression

This chapter describes the concepts underlying and assumptions of multiple regression, a method of predicting a normally distributed dependent variable from a set of normally distributed or dichotomous independent variables. There are several different ways of computing multiple regression that are used under somewhat different circumstances. This chapter guides the reader to use two of these approaches: hierarchical multiple regression , in which one enters predictor variables in a series of predetermined steps so that one can see how the entry of later predictors impacts the relations of earlier variables to the dependent variable, and simultaneous multiple regression , in which one enters all predictors at the same time. By comparing these two analyses, readers will be able to see that the method one uses to compute multiple regression influences the information one obtains from the analysis. Neither stepwise nor forward entry nor backward elimination methods of multiple regression are demonstrated, as these are not considered best practices, since they may capitalize on chance.