How to Interpret Multiple Regressions in SPSS
- 1). Examine the "variables entered/removed" table. This shows you which independent variables were included in your regression. If you ran more than one model, each will be shown. This is good for checking that SPSS did what you wanted it to do.
- 2). Examine the model summary table. This includes the R, R-squared and adjusted R-squared for the model, and the standard error of the estimate. R-squared is a measure of how much of the variation in the dependent variable is accounted for by the model. Adjusted R-squared attempts to adjust this for the complexity of the model. More complex models will explain more variance than simpler models.
- 3). Examine the ANOVA (Analysis of Variance) table. In particular, the F and its df (degrees of freedom) are indicators of how good the model is. Sig. (statistical significance) is a measure of how likely it is that an F this high or higher could have arisen if there was no relationship in the whole population from which the sample analyzed was drawn.
- 4). Examine the coefficients table. This tells you how statistically significant each independent variable is (Sig.). The unstandardized coefficient is an estimate which you can use to build a predictive model. The output also includes a confidence interval for each variable's parameter estimate, which tells you the range that you can be 95 percent confident that the parameter falls within.
