66 maximum velue is 6.00 for Gudang Garam Tbk 2010. The average value of tobins-q
variable is 3.36 with the standar deviation of 0.97 which is lower than average value, indicating that data have low deviation which means the data is distributed normally.
c. Family Ownership The result of the analysis using descriptive statistic on family ownership indicates the
minimum value is 0.0002 for PT Bakrie Sumatera Plantation 2012, while the maximum velue is 0.97 for
PT. Summarecon 2011
. The average value of tobins-q variable is 0.39 with the standar deviation of 0.28 which is lower than average value,
indicating that data have low deviation which means the data is distributed normally. d. Family DirectorManager
The result of the analysis using descriptive statistic on family ownership indicates the minimum value is 1.00 for
PT. Bakrie Brother Tbk, PT. Bakrie Sumatera Plantation Tbk
2010, while the maximum velue is 5.00 for PT Ciputra Property Tbk 2010. The average value of tobins-q variable is 1.81 with the standar deviation of 1.22 which is
lower than average value, indicating that data have low deviation which means the data is distributed normally.
67
2. Data Processing a. Classic Assumption Test
1 Normality
Normality Data test aims to test whether the dependent variable and independent variables both have a normal distribution or not in the regression model.
The results of the normality data test using Jarque-Bera test in distress category and non-distress can be shown in the following table:
Table 4.4 The Result of Normality Test
2 4
6 8
10
-1.0 -0.5
0.0 0.5
1.0 1.5
Series: Residuals Sample 1 80
Observations 80 Mean
0.003995 Median
-0.079759 Maximum
1.467058 Minimum
-1.016855 Std. Dev.
0.569449 Skewness
0.469957 Kurtosis
2.823489 Jarque-Bera
3.048649 Probability
0.217768
Source: Output Eviews 8.0
Based on the result of normality test, the value of probability is 0.21, more than significance level of 0,05. Based on Widarjono 2013:50, it can be concluded
that the data is normal distributed.
68
2. Multicollinearity
Multicollinearity test is used to test the existing of perfect relationship or near-perfect relationship between the independent variables the regression model.
The multicollinearity test results are shown in the following table:
Table 4.5 The Result Of Correlation between Independent Variable
BOD CA
FO FDM
BOD 1.000000
0.483204 0.200482
0.301886 CA
0.483204 1.000000
0.009040 -0.091258
FO 0.200482
0.009040 1.000000
0.110585 FDM
0.301886 -0.091258
0.110585 1.000000
Source: Output Eviews 8.0
Based on the table 4.5, it is known that the value of each variables BOD, CA, FO and FMD toward the other veriables is less than 0,8. According to Ghozali
2013:83, it can be concluded that the multicollinearity does not occur on board of director, size of committee, family ownership and family managerdirector variables
.
3. Autocorrelation
Autocorrelation test is used to detect the internal correlation among the groups of a series observation arrange in a series of place and time. The basic of decision
making in this test are based on Durbin-Watson Test, which can be seen in the table below: