Figure 7.2 presents the final measurement model of the freight consolidation construct. The standardised loading, composite reliability, Cronbach alpha and AVE results are presented in the Table 7.5-7.8 and the value of the figure is rounded up by AMOS version 21. Location of freight consolidation centre dimension consists of the observed variables LO_4.5, LO_4.6, LO_4.7, and LO_4.10. These observed variables meet convergent validity criterion with the standardised loadings greater than the threshold value of 0.5 (0.57 < β < 0.82) (p<0.01), and construct validity with the value of CR (.81) greater than the value of AVE (.51) (see Table 7.5 and 7.7). Moreover, they demonstrate discriminant validity, as they are highly correlated to the location of freight consolidation centre (LO) dimension while having lower correlation with other dimensions (covariance varies between 0.78 and 0.81) (Table 7.6). These observed variables are reliable, since their SMC exceeds the minimum threshold of 0.3 (0.32 < SMC <
0.67) (see Table 7.5). Moreover, these observed variables are reliable because the Cronbach’s alpha is .80, composite reliability is .81, and AVE is .51 (see Table 7.7).
The geographical coverage dimension consists of three observed variables GC_5.1, GC_5.4 and GC_5.5. These observed variables are shown to exhibit convergent validity as the standardised loadings are greater than the threshold value of 0.5 (0.68 < β < 0.81) (p<0.01) and construct validity with the value of CR (.80) greater than the value of AVE (.57) (see Table 7.5 and 7.7). Moreover, they meet the discriminant validity criterion, as clearly clustered into their respective dimensions with covariance varies between 0.78 and 0.89 (see Table 7.6). These observed variables are reliable, as their SMC is greater than the minimum threshold of 0.3 (0.46 < SMC < 0.65), as well as because the Cronbach’s alpha is .79, composite reliability is .80, and AVE is .57 (see Table 7.5 and 7.7).
The utilization of transport modes dimension consists of three observed variables UT_6.1, UT_6.6 and UT_6.7. The error term of variable UT_6.6 and 6.7 are correlated since they are sharing something in common to the latent variable (Byrne 2009). This is justified because the reduction of delivery vehicle would automatically lead to the reduction of driver. These observed variables are shown to exhibit convergent validity criterion since the standardised loadings exceed the threshold value of 0.5 (0.59 < β < 0.83) (p<0.01) and construct validity with the value of CR (.73) greater than the value of AVE (.50) (see Table 7.5 and 7.7).
Moreover, they demonstrate discriminant validity, as clustered into their respective dimensions with covariance varies between 0.81 and 0.89 (see Table 7.6). These observed variables are
173 reliable, as their SMC is greater than the minimum threshold of 0.3 (0.35 < SMC < 0.70), as well as because the Cronbach’s alpha is .79, composite reliability is .73, and AVE is .50 (see Table 7.5 and 7.7).
Note: LO = location of freight consolidation centre,
GC = geographical coverage, UT = utilization of transport modes.
Figure 7.2: Standardized estimates for freight consolidation construct
174 Table 7.5: Standardized factor loading, squared multiple correlation and p value of freight
consolidation construct Freight consolidation
Location of freight consolidation centre Question
items
Item descriptions Standardised
Loading **
Squared Multiple Correlation
P-value
LO_4.5 You are going to implement freight consolidation, if the proper location of freight consolidation centre can improve inbound and outbound flow of products.
0.72 0.51 0.001
LO_4.6 You are going to implement freight consolidation, if the proper location of freight consolidation centre can reduce distribution costs.
0.82 0.67 0.001
LO_4.7 You are going to implement freight consolidation, if the proper location of freight consolidation centre can improve delivery flexibility.
0.74 0.55 0.001
LO_4.10 You are going to implement freight consolidation, if the proper location of freight consolidation centre can improve the flow of product returns.
0.57 0.33 0.001
Geographical coverage Question
items
Item descriptions Standardised
Loading **
Squared Multiple Correlation
P-value
GC_5.1 You are going to implement freight consolidation, if it can improve on time delivery of each drop-off point.
0.68 0.46 0.001
GC_5.4 You are going to implement freight 0.78 0.61 0.001
175 consolidation, if it can reduce travel
distance.
GC_5.5 You are going to implement freight consolidation, if it can reduce fuel consumption.
0.81 0.65 0.001
Utilization of transport mode Question
items
Item descriptions Standardised
Loading **
Squared Multiple Correlation
P-value
UT_6.1 You are going to implement freight consolidation centre, if it can reduce transportation costs.
0.84 0.70 0.001
UT_6.6 You are going to implement freight consolidation centre, if it can reduce the number of delivery vehicles.
0.64 0.41 0.001
UT_6.7 You are going to implement freight consolidation centre, if it can reduce the number of drivers.
0.59 0.35 0.001
Achieved Fit Indices
Chi-square =49.35, Degrees of Freedom = 31, P = 0.02, Bollen-Stine p value = 0.21, CMIN/DF = 1.59, GFI = 0.96, AGFI = 0.93, NFI = 0.96, TLI = 0.97, CFI = 0.98, RMSEA = 0.05
Note: ** Statistically significant at p < 0.01 (two-tailed)
176 Table 7.6: Correlations of measurement items and sub-constructs under freight consolidation
construct
UT GC LO
UT 1.000
GC 0.89 1.000
LO 0.81 0.78 1.000 b6.1 0.83 0.64 0.57 b6.6 0.64 0.57 0.52 b6.7 0.59 0.52 0.48 b5.1 0.50 0.68 0.53 b5.4 0.51 0.78 0.51 b5.5 0.48 0.81 0.53 b4.5 0.58 0.56 0.72 b4.6 0.57 0.64 0.82 b4.7 0.40 0.58 0.74 b4.10 0.46 0.45 0.57
Table 7.7: Validity and reliability test of freight consolidation construct Cronbach’s alpha
(α)
Composite reliability (CR)
Average variance extracted (AVE) Freight
consolidation
0.90 0.92 0.52
LO 0.80 0.81 0.51
GC 0.79 0.80 0.57
UT 0.79 0.73 0.50
177 Location of freight consolidation centre dimension, geographical coverage dimension and utilization of transport modes dimension are, therefore, reliable and valid for freight consolidation construct because the composite reliability of .92, Cronbach alpha of 0.90, and AVE of 0.52 is greater than the threshold value (Table 7.7). Moreover, the Pearson’s correlations between dimensions are below 0.9 (.79 < r < .89), which indicates discriminant validity and unidimensionality (Table 7.6). Nonetheless, all measurement dimensions demonstrate discriminate validity, as their chi-square differences are significant (Table 7.8).
The measurement model fits the data very well with parameter like Chi-square = 49.347, Degrees of freedom = 31, p value = 0.019 (Bollen-Stine p value = 0.209, which is not significant at the 0.05 level). Other fit measures also indicate the goodness of fit of the model (CMIN/DF = 1.592, GFI = 0.963, AGFI = 0.934, NFI = 0.956, TLI = 0.975, CFI = 0.983, RMSEA = 0.050) (Table 7.5).
Table 7.8: Chi-square difference test of freight consolidation 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 LO & GC 10.48 13 686.41 14 675.93 1 0.000 Significant
LO & UT 21.32 12 760.21 13 738.89 1 0.000 Significant
GC & UT 11.57 7 614.10 8 602.53 1 0.000 Significant