Global Journal of Computer Science and Technology, G: Interdisciplinary, Volume 23 Issue 1

Table 7: Multicollinearity Test Based on Condition Index Value Source: Authors’ Representation Based on SPSS Results Assumption 03: Normality Distribution of Data Set and Test for Outliers Reponses for the survey instrument were collected from 269 respondents, which is a relatively large sample, and hence the Central Limit Theorem could be applied, posing no question on normality. Descriptive statistics relevant to normal distribution are shown in Table 8, with values of 4.47, 4.47, and 4.47 indicated as the mean, mode, and median of the dataset respectively. Table 8: Test of Normality Based on Descriptive Statistics of Dataset Remote ERP Implementation Success Mean 4.474281 Standard Error 0.02771 Median 4.472136 Mode 4.472136 Standard Deviation 0.454478 Sample Variance 0.20655 Kurtosis 1.093872 Skewness -0.96841 Range 2.289194 Minimum 2.710806 Maximum 5 Sum 1203.582 Count 269 Source: Authors’ Representation Based on SPSS Results Assumption 04: Linearity between Independent and Dependent Variables The link between dependent and independent variables can only be effectively estimated using standard multiple regression if the relationships are linear (Pallant, 2001). The findings of the regression analysis will underestimate the true relationship if the relationship between the independent factors and the dependent variable is not linear. As tested previously, both descriptively and inferentially using scatter diagrams and Pearson correlation coefficient, it was assured that a linear relationship exists between identified six factors and remote ERP implementation success, which thereby poses no concerns on linearity between variables. Assumption 05: Homoscedasticity of Independent Variables The variance of errors is the same across all levels of the independent variables, known as homoscedasticity. A visual study of a plot of the standardized residuals (errors) by the regression standardized projected value can verify this assumption (Jarque and Bera, 1980). In an ideal world, residuals are randomly spread around 0 (the horizontal line), resulting in a relatively uniform distribution that meets the homoscedasticity assumption. Dimension Eigen value Condition Index Variance Proportions (Constant) Top_Mgt_C ommitment Change_ Mgt Communicati on Implementati on_strategy User training and education Project_ Manage ment 1 6.956 1.000 .00 .00 .00 .00 .00 .00 .00 2 .015 21.406 .00 .04 .13 .01 .03 .01 .49 3 .009 27.210 .00 .18 .59 .02 .01 .03 .22 4 .008 28.751 .00 .47 .17 .08 .01 .11 .11 5 .005 38.219 .00 .02 .06 .00 .78 .38 .03 6 .004 43.304 .01 .00 .00 .77 .15 .45 .02 7 .002 56.772 .98 .29 .05 .12 .02 .01 .14 © 2023 Global Journals Global Journal of Computer Science and Technology Volume XXIII Issue I Version I 40 ( ) Year 2023 G Critical Success Factors of Remote ERP Implementation: From System Users’ Perspective