Measuring Variability in Decision Prediction with Standard Error of Estimate TEST • • • • • • • • BI 1 variable has a positive and significant effect on loyalty. BI 2 variable has a positive and significant effect on loyalty. BI 3 variables have a positive and significant effect on loyalty. BI 4 variables have a positive and significant effect on loyalty. CS1 variable has a positive and significant effect on loyalty. CS2 variable has a positive and significant effect on loyalty. CS3 variables have a positive and significant effect on loyalty. CS4 variable has a positive and significant effect on loyalty. Model Summary Adjusted R ModelR R Square Square Std. Error of the Estimate 1 .225a .051 -.043 .89916 a. Predictors: (Constant), CS4, BI2, CS3, BI1, CS2, BI3, BI4, CS1 ANOVAa Sum of ModelSquares df 1 Mean Square Regression 3.501 8 .438 Residuals 65.488 81 .808 Total 68.989 89 F .541 Sig. .822b a. Dependent Variable: SB3 b. Predictors: (Constant), CS4, BI2, CS3, BI1, CS2, BI3, BI4, CS1 Coefficientsa ModeltSig 1 (Constant) 4.563 1.176 BI1 .124 .140 BI2 -.037 BI3 -.035 . 3.881 .000 .113 .880 .381 .140 -.035 -.266 .791 .144 -.033 -.240 .811 BI4 -.141 .142 -.134 -.990 .325 CS1 -.182 .167 -.149 -1.091 .279 CS2 -.064 .127 -.063 -.499 .619 CS3 .051 .154 .043 .330 .742 CS4 .133 .111 .145 1.201 .233 a. Dependent Variable: SB3 • The model shows R = 0.225, meaning that the relationship between quality and decision is 22.5%. This means that the relationship is not close. The greater the R means the tighter the relationship. To ensure the type of relationship between variables can be seen in the following table: Value 0.0 - 0.19 0.2 - 0.39 0.4 - 0.59 0.6 - 0.79 0.8 - 0.99 Interpretation Very heavy Not heavy Quite heavy Erat Sangaterat • R square of 0.051 means that 0.51% of the decision variable can be explained by quality, while the remaining 99.49% can be explained by other factors examined by this study. • Standard Error of Estimate means measuring the variation of the predicted value. Standard Error of Estimate can also be referred to as standard deviation. The Standard Error of Estimate is 0.89916. then the smaller the standard deviation means the better the model. b). In the table, it can be seen that the significant value is 0.000 (more than 0.05), meaning that quality has a positive and significant effect on decisions. Means reject Ho and accept Hi. The regression equation is Y = 4.563 + 7.67 X, meaning that if we increase the quality by an amount of One-unit will increase the decision by 7.67.