Regression analysis is a technique used for the modeling and analysis of numerical data consisting of values of a dependent variable (response variable) and of one or more independent variables (explanatory variables). Multiple regression deals with more than one independent variable, and is often used in marketing research to determine key drivers of overall liking, satisfaction, or purchase intent.
So what is multiple regression analysis used for?
Multiple regression can be used for prediction (including forecasting of time-series data), inference, hypothesis testing, and modeling of causal relationships. In marketing research, multiple regression analysis is used to look at causal relationships between the dependent variable and independent variables, thus determining key drivers. The key drivers (the variables you suspect have a direct influence on overall satisfaction, purchase intent, or overall liking) become the independent variables, whereas overall liking, satisfaction, or purchase intent (depending on what it is you’re examining) becomes the dependent variable.
How is it calculated?
With the help of a statistical program (e.g. SPSS, SAS), regression coefficients are calculated, which show the order of association of each independent variable on the dependent variable. Standardized beta coefficients are used for comparison, as they indicate the relative importance of alternative predictor variables. The higher the value of the standardized beta coefficient, the higher the effect on the dependent variable is. Using a stepwise regression analysis will prevent you from classifying two independent variables that are highly correlated as two separate key drivers. Only those effects that are statistically significant are included in the model and can be qualified as key drivers.
For more information about Multiple Regression, go to
http://www.tns-global.com/corporate/Doc/0/3PEH43LO30B4B2TOTS8N9H4QCF/23322%20Modeling_2-4.pdf
