Multiple regression generally explains the relationship between multiple independent or multiple predictor variables and a dependent or criterion variable. In multiple regression, a dependent variable is modeled as a function of several independent variables with corresponding multiple regression coefficients, along with the constant term. Multiple regression requires a large number of observations, and this is why it is called multiple regression.
The multiple regression equation explained above takes the following form:
y = b1x1 + b2x2 + … + bnxn + c.
Here, bi’s (i=1,2…n) are the regression coefficients, which in multiple regression represents the value at which the criterion variable changes when the predictor variable changes simultaneously.
As an example, let’s say that the test score of a student in an exam will be dependent on various factors like his focus while attending the class, his intake of food before the exam and the amount of sleep he gets before the exam. With the help of multiple regression, one can estimate the appropriate relationship among these factors.
Multiple regression in SPSS is done by selecting “analyze” from the menu. Then, from analyze, select “regression,” and from regression select “linear.” You can then perform multiple regression.
There are certain terminologies in multiple regression that help in understanding multiple regression. These terminologies are as follows:
The beta value in multiple regression is used in measuring how effectively the predictor variable influences the criterion variable. In multiple regression, it is measured in terms of standard deviation.
R, in multiple regression is the measure of association between the observed value and the predicted value of the criterion variable. R Square, or R2 in multiple regression is the square of the measure of association which indicates the variability in the criterion variable. Adjusted R2 in multiple regression takes the number of variables in the model and the number of data points in the model under consideration.
Assumptions of Multiple Regression
There should be proper specification of the model in multiple regression. This means that only relevant variables must be included in the multiple regression model. This means that in multiple regression the model should be reliable.
Linearity must be assumed in multiple regression. This means that in multiple regression, the model should be linear in nature.
Normality must be assumed in multiple regression. This means that in multiple regression, variables must have normal distribution.
Homoscedasticity must be assumed in multiple regression. This means that in multiple regression, the variability in errors must be the same across all levels.
Multiple Regression Resources
Achen, C. H. (1982). Interpreting and using regression. Newbury Park, CA: Sage Publications.
Afifi, A. A., Kotlerman, J. B., Ettner, S. L., & Cowan, M. (2007). Methods for improving regression analysis for skewed continuous or counted responses. Annual Review of Public Health, 28, 95-111.
Aguinis, H. (2004). Regression analysis for categorical moderators. New York: Guilford Press.
Algina, J., & Olejnik, S. (2003). Sample size tables for correlation analysis with applications in partial correlation and multiple regression analysis. Multivariate Behavioral Research, 38(3), 309-323.
Allison, P. D. (1999). Multiple regression. Thousand Oaks, CA: Pine Forge Press.
Anderson, E. B. (2004). Latent regression analysis based on the rating scale model. Psychological Science, 46(2), 209-226.
Belsley, D. A., Kuh, E., & Welsch, R. E. (1980). Regression diagnostics: Identification influential data and sources of collinearity.New York: John Wiley & Sons.
Berk, R. A. (2003). Regression analysis: A constructive critique. Thousand Oaks, CA: Sage Publications.
Berry, W. D. (1993). Understanding regression assumptions. Newbury Park, CA: Sage Publications.
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis for the behavioral sciences. (3rd ed.). Mahwah, NJ: Lawrence Erlbaum Associates.
Cook, R. D., & Weisberg, S. (1982). Residuals and influence in regression. New York: Chapman and Hall.
Fox, J. (1991). Regression diagnostics. Newbury Park, CA: Sage Publications.
Fox, J. (2000a). Nonparametric simple regression: Smoothing scatterplots. Thousand Oaks, CA: Sage Publications.
Fox, J. (2000b). Multiple and generalized nonparametric regression. Thousand Oaks, CA: Sage Publications.
Hardy, M. A. (1993). Regression with dummy variables. Newbury Park, CA: Sage Publications.
Jaccard, J. (2001). Interaction effects in logistic regression. Thousand Oaks, CA: Sage Publications.
Kahane, L. H. (2001). Regression basics. Thousand Oaks, CA: Sage Publications.
Long, J. S. (1997). Regression models for categorical and limited dependent variables. Thousand Oaks, CA: Sage Publications.
Miles, J., & Shevlin, M. (2001). Applying regression and correlation: A guide for students and researchers. Thousand Oaks, CA: Sage Publications.
Pedhazur, E. J. (1997). Multiple regression in behavioral research (3rd ed.). Fort Worth, TX: Harcourt Brace.
Schroeder, L. D., Sjoquist, D. L., & Stephan, P. E. (1986). Understanding regression analysis: An introductory guide. Newbury Park, CA: Sage Publications.
Serlin, R. C., & Harwell, M. R. (2004). More powerful tests of predictor subsets in regression analysis under nonnormality. Psychological Methods, 9(4), 492-509.
