Overview of the Analysis
To better understand the nature of the relationship among the five variables presented in Figure 14.7, the data are analyzed using two SAS procedures. First, PROC CORR will compute Pearson correlations between variables. This is useful for understanding the big picture: the simple bivariate relations among the variables.
Next, PROC REG is used to perform a multiple regression in which commitment is regressed on the four predictor variables simultaneously. This provides a number of important pieces of information. First, you will learn whether there is a significant relationship between commitment and the linear combination of predictors (i.e., whether there is a significant relationship between commitment and the four variables taken as a group). In addition, you will review the multiple regression coefficients for each of the four predictors to determine which are statistically significant and which standardized coefficients are relatively large.
Finally, the SELECTION=RSQUARE option is used with PROC REG in a separate analysis to determine the amount of variance in commitment that is accounted for by every possible combination of predictor variables. Results from this analysis are used to determine the uniqueness index for each of the four predictors. The formula for testing the significance of the difference between two R2 values are then used to determine which of the uniqueness indices significantly differ from zero.