Results of Validity Test

81 Table 4.9 Result of Reliability Test Variables Cronbachs Alpha Number of Items Explanation Organizational Structure 0,755 4 Reliable Organizational Procedure 0,793 9 Reliable Organizational Strategy 0,786 7 Reliable Organizational Culture 0,779 7 Reliable Taxpayer Compliance 0,749 6 Reliable Source: Primary Data Output from SPSS 20 Table 4.9 shows the value of cronbachs alpha. The value of alpha in organizational structure is 0.755, in organizational procedure is 0.793, in organizational strategy is 0.786, in organizational culture is 0.779, and in taxpayer compliance is 0.749. Thus, it can be concluded that all variables are reliable because the value of cronbachs alpha is greater than 0.60. Every question used will be able to obtain consistent data. It means that the question will provide relatively the same answer even though it was tasted many times. 82

3. Result of Classic Assumption

a. Result of Normality Test

The aim of normality test is to know the distribution of data used in the research. A good data should be normally distributed. One of the ways to detect whether the data are normally distributed data or not can be done by using graph analysis namely Normal Probability Plot P-P Plot. The data is normally distributed when the dots spread around the diagonal line and follow the direction of the diagonal line. The result of normality test can be seen in figure 4.1 Figure 4.1 Result of Normality Data Test Source: Primary Data Output from SPSS 20 From the P-P Plots diagram above, it can be seen that the dots spread around the diagonal line and follow the direction of a diagonal 83 spread. Thus, it can be concluded that the distribution of the data close to normal or have met the assumptions of normality. This result also tells that the regression model is qualified to apply due to the assumption of normality test Ghozali 2009: 149.

b. Result of Multicollinearity Test

The aim from multicolinearity test is to test whether the regression model found a correlation among the independent variables. In this research, detection of the presence or absence in multicolinearity can be done by calculating value of variance inflation factor VIF of each independent variable. Table 4.10 Result of Multicollinearity Test Coefficients a Model Collinearity Statistics Tolerance VIF 1 Constant ST .526 1.902 PO .389 2.568 SO .568 1.760 BO .532 1.879 a. Dependent Variable: KWP Source: Primary Data Output from SPSS 20 Based on table 4.10 above, the result shows that there is no value of variance inflation factor VIF of each independent variable which is less than 0.1 or more than 10. So, it can be concluded that there is no multicolinearity. 84

c. Result of Heteroscedasticity Test

The aim from heteroscedastisity test is to test whether the regression model occur the variance inequality of the residual from one observation to another observation. A good regression model is homocedastisity or there is no heteroscedastisity. The results of heteroscedastisity by using scatter plot can be seen in the following figure 4.2 Figure 4.2 Result of Heteroscedastic Test Source: Primary Data Output from SPSS 20 Based on figure 4.2 above, it can be seen that the dots spread above and below zero 0 on the Y axis. So, it indicates that there is no heteroscedasticity or homocedasticity.