Test Normalitas

Based on the results of testing different variability levels of profit shares between before and after the dividend announcement, the value obtained by calculating z-probability of 0.401 is greater extent of significance 5%. So that it can be concluded that there are no significant differences between mean variability level of dividend before and after the dividend increase. This result can also be interpreted that investor expectations of stock prices does not differ between before and after the dividend announcement. Or in other words the dividend announcement does not affect the return received by investors.

Normalitas test conducted to test the distribution of data, the results of the test data indicates that normalitas data variability level dividend berdistribusi normal, this is shown from the probability value (Asymp Sign) greater than 0.05. Test results data normalitas dividend rate variability shown in table 4.7 below.
Table 4.7
Test results Normalitas

Variable Description Zhitung Sign Conclusion
AR before the dividend increase
After AR 2.898
2.824 0.000
0.000 Not Normal
Not Normal
AR down before dividend
After AR 2.380
1.813 0.000
0.003 Not Normal
Not Normal

Normalitas test results show that the data of each variable berdistribusi normal, because the probability has a value greater than 0.05. Therefore, the test is done with the analysis parametrik non Wilxocon Sign Rank Test