Year 2010 5500 8750 727679 0814621 Year 2009 5000 8750 688393 0906464 Year 2008

Year 2010 5500 8750 727679 0814621 year 2009 5000

This preview shows page 12 - 14 out of 25 pages.

Year 2010 .5500 .8750 .727679 .0814621 Year 2009 .5000 .8750 .688393 .0906464 Year 2008 .4750 .8500 .650000 .1004619 Year 2007 .4000 .8500 .590179 .1057030 As CSR disclosure score increase over time, it may be helpful to test whether there is significant difference among CSR disclosure scores over the period under investigation. To test whether the observed changes in the CSR disclosure over the period of study are statistically significant or not, a series of statistical tests have been conducted. Descriptive statistics for the dependent and explanatory variables are presented in table 4. Table 4: Descriptive Statistics for Regression Variables Variables Mean Std. Deviation Skewness Kurtosis CSR Disclosure 0.685 0.109067 -0.28396 2.575209 Firm Size 10.8601 0.309912 -0.14234 5.821319 Firm Profitability 0.195314 0.130288 -0.43703 11.55972 Bank Age 11.21429 8.483404 0.59907 1.887174 Ownership Structure 0.257143 0.438628 1.111325 2.235043 Independent Director 0.032791 0.044015 0.969439 2.689886 Role Duality 0.928571 0.258464 -3.3282 12.07692 Board Size 13.83571 4.585888 0.417916 2.707228 From the table (4) it is observed that the overall data set is not normally distributed. As a common rule, the standard skewness of the data needs to be within the range of ±1.96 (Haniffa and Hudaib 2006; Gujarati 2003). It is observed that the minimum CSR skewness is -3.3282 which exceed the range of ±1.96 evidencing that the data is not normally distributed. The study also checked the skewness of each year's data and found that the data set is not normally distributed. Additionally, with respect to the standard kurtosis the data is not normally distributed. The data is said to be normally distributed if the standard kurtosis fall in the range of ±3 (Haniffa and Hudaib 2006; Gujarati 2003). The standard kurtosis of the total CSR disclosure and its different years exceed the range of ±3 indicating that the data is not normally distributed. Besides testing for normality, it is important to check for multicollinearity, homoscedasticity and linearity. Regarding The Variance Inflation Factor (VIF) indicated that data is normally distributed if the VIF is less than 10(Gujarati 2003; Neter et al. 1983; Mendenhall and Sincich 1989 ) . However, others suggest that the value of 5 can be used as a rule of thumb (Groebner et al. 2005). However, it can be seen from the table 5 that the maximum VIF is 1.460 and the mean VIF is 1.321. Moreover, the lowest tolerance coefficient is 0.685. Therefore, the results of VIF and tolerance coefficients; based upon the rule of thumb; indicate that there is no unacceptable level of multicollinearity in the current study. Moreover, there has been no agreement among researchers regarding the cut off correlation percentage (Alsaeed 2006). While some researchers use 0.8; e.g. Gujarati (2003); others suggest using 0.7; e.g. Tabachnick and Fidell (1996).
Image of page 12
Das, Dixon & Michael 142 Table 5: Collinearity Statistics of Variables Variables Collinearity Statistics Tolerance VIF Firm Size 0.685 1.460 Firm Profitability 0.730 1.370 Bank Age 0.827 1.210 Ownership Structure 0.762 1.312 Independent Director 0.748 1.336 Role Duality 0.879 1.138 Board Size 0.698 1.433 Mean VIF 1.321 Table 6: Correlation matrix for Dependent and Independent Variables Pearson Correlation Matrix for Dependent and Independent Variable Pearson CSR_DISC LASST ROE AGE O.SHIP I.DIR DUALIT B_SIZE CSR_DISC 1 LASST 0.479 ** (0) 1 ROE 0.109 (.199) 0.237 ** (0.005) 1 AGE -0.076 (0.373) 0.109 0.202 -0.168 * (0.047) 1 OWNERSHIP 0.280 ** (0.001) 0.162 (0.056) -.288 ** (0.000) -0.104 (0.222) 1
Image of page 13
Image of page 14

You've reached the end of your free preview.

Want to read all 25 pages?

  • Spring '20
  • Amanullah
  • Test, The Land, Corporate social responsibility

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture