Figure 7.4 presents the final measurement model of sustainable distribution construct. The standardised loading, composite reliability, Cronbach alpha and AVE results are presented in the Table 7.13-7.16 and the value of the figure is rounded up by AMOS version 21.
Environmental dimension consists of observed variables EV_10.1.6, EV_10.1.7, EV_10.1.8 and EV_10.1.10. These observed variables are shown to exhibit convergent validity, since their standardised loadings are greater than the threshold value of 0.5 (0.75 < β < 0.86) (p<0.01), and construct validity with the value of CR (.88) greater than the value of AVE (.66) (see Table 7.13 and 7.15). Moreover, they meet the discriminant validity criterion, as clustered into their respective dimensions with covariance varies between 0.52 and 0.60 (see Table 7.14). These observed variables are reliable, since their SMC exceeds the minimum threshold of 0.3 (0.56 <
SMC < 0.74) (see Table 7.13). Moreover, they are reliable, as the Cronbach’s alpha is 0.88, composite reliability is 0.88, and AVE is 0.66 (see Table 7.15).
The economic dimension consists of five observed variables EC_10.2.1, EC_10.2.5, EC_10.2.7, EC_10.2.8 and EC_10.2.14. These observed variables are shown to exhibit convergent validity since their standardised loadings are greater than the threshold value of 0.5 (0.66 < β < 0.73) (p<0.01), and construct validity with the value of CR (.82) greater than the value of AVE (.50) (see Table 7.13 and 7.15). Moreover, they demonstrate discriminant validity, as they are clearly clustered into their respective dimensions (see Table 7.14) with covariance varies between 0.60 and 0.66. These observed variables are reliable, since their SMC is greater than the minimum threshold of 0.3 (0.44 < SMC < 0.53), as well as because the Cronbach’s alpha is .83, composite reliability is .82, and AVE is .50 (see Table 7.13 and 7.15).
The social dimension consists of three observed variables SO_10.3.1, SO_10.3.6 and SO_10.3.7. These observed variables exhibit convergent validity, since the standardised loadings are greater than the threshold value of 0.5 (0.72 < β < 0.89) (p<0.01), and construct validity with the value of CR (.86) greater than the value of AVE (.67) (see Table 7.13 and 7.15). Moreover, they meet the discriminant validity criterion, as clustered into their respective dimensions with covariance varies between 0.52 and 0.66 (see Table 7.14). They are reliable since their SMC is greater than the minimum threshold of 0.3 (0.53 < SMC < 0.80) and because the Cronbach’s alpha is .85, composite reliability is .86, and AVE is .67 (see Table 7.13 and 7.15).
185 Note: EV = environmental factor, EC = economic factor, SO = social factor
Figure 7.4: Standardized estimates for sustainable distribution construct
186 Table 7.13: Standardized factor loading, squared multiple correlation and p value of sustainable
distribution construct Sustainable distribution
Environmental factor
Question items Item descriptions Standardised
Loading **
Squared Multiple Correlation
P-value
EV_10.1.6 Reduce water pollution 0.81 0.66 0.001
EV_10.1.7 Reduce visual pollution 0.82 0.67 0.001
EV_10.1.8 Reduce odour pollution 0.86 0.74 0.001
EV_10.1.10 Reduce solid waste 0.75 0.56 0.001
Economic factor
Question items Item descriptions Standardised
Loading **
Squared Multiple Correlation
P-value
EC_10.2.1 Improve company’s reputation 0.71 0.51 0.001
EC_10.2.5 lower the risk of business operation 0.66 0.44 0.001 EC_10.2.7 Help identify easier ways to attract external
sources of sponsorship
0.73 0.53 0.001
EC_10.2.8 Broaden markets and improve conditions for sales increase
0.67 0.45 0.001
EC_10.2.14 Improve market opportunities 0.69 0.48 0.001
Social factor
Question items Item descriptions Standardised
Loading **
Squared Multiple Correlation
P-value
SO_10.3.1 Increase staff motivation 0.73 0.53 0.001
SO_10.3.6 Enhance human capital value 0.83 0.69 0.001
SO_10.3.7 Improve the contribution of a firm to community development (i.e., job creation and tax breaks received)
0.89 0.80 0.001
187 Achieved Fit Indices
Chi-square = 64.07, Degrees of Freedom = 50, P = 0.09, Bollen-Stine p value = 0.29, CMIN/DF = 1.28, GFI = 0.96, AGFI = 0.93, NFI = 0.96, TLI = 0.99, CFI = 0.99, RMSEA = 0.03
Note: ** Statistically significant at p < 0.01 (two-tailed)
Table 7.14: Correlations of measurement items and sub-constructs under sustainable distribution construct
SO EC EV
SO 1.000
EC 0.66 1.000
EV 0.52 0.60 1.000 d10.3.6 0.83 0.55 0.43 d10.3.1 0.72 0.49 0.38 d10.3.7 0.89 0.59 0.47 d10.2.5 0.44 0.66 0.40 d10.2.8 0.44 0.67 0.40 d10.2.7 0.48 0.73 0.44 d10.2.14 0.46 0.69 0.42 d10.2.1 0.47 0.71 0.43 d10.1.10 0.39 0.45 0.75 d10.1.8 0.45 0.52 0.86 d10.1.7 0.43 0.50 0.82 d10.1.6 0.42 0.49 0.81
188 Table 7.15: Validity and reliability test of sustainable distribution construct
Cronbach’s alpha (α)
Composite reliability (CR)
Average variance extracted (AVE)
Sustainable distribution 0.90 0.94 0.59
EV 0.88 0.88 0.66
EC 0.83 0.82 0.50
SO 0.85 0.86 0.67
Based on the evidence presented above, environmental, economic, and social dimension are reliable and valid for sustainable distribution construct, since the composite reliability is .94, Cronbach’s alpha is 0.90 and AVE is 0.59 (see Table 7.15). The Pearson’s correlations between dimensions are less than 0.9 (0.52 < r < 0.66), which indicate discriminant validity and unidimensionality (Table 7.14). Referring to Table 7.16, all measurement dimensions meet the discriminant validity criterion, since their chi-square differences are significant. The measurement model fits the data very well, as the Chi-square = 64.07, degrees of freedom = 50, p value = 0.09 (Bollen-Stine p value = 0.29, which is not significant at the level of 0.05). Other fit measures also indicate the goodness of fit of the model to the data (CMIN/DF = 1.28, GFI = 0.96, AGFI = 0.93, NFI = 0.96, TLI = 0.99, CFI = 0.99, RMSEA = 0.03) (Table 7.13).
Table 7.16: Chi-square difference test of sustainable distribution construct Pairs of
Constructs
χ2 of model 1 (correlation is unconstrained)
df of model 1
χ2 of model 2 (correlation is constrained to 1)
df of model 2
∆ χ2 ∆df p-value Chi-Square Critical Values;
p =0.05 EV & EC 25.06 25 149.79 26 124.73 1 0.000 Significant EV & SO 17.01 13 130.71 14 113.69 1 0.000 Significant EC & SO 26.55 18 598.14 19 571.59 1 0.000 Significant