Results
We used confirmatory factor analysis to verify the measurement models. As we show
in Table III, the Cronbach’s a statistics for the constructs range from 0.718 for
proactiveness to 0.895 for product advantages, which suggests the scales are sufficiently
reliable. Each item loads significantly on the expected constructs, in support of
convergent validity. All constructs also exhibit excellent convergent validity; their
average variance extracted exceeds 0.50.
We use the factor scores for the measured variables in each construct to test the
hypotheses with multiple linear regression analysis (Table IV). The overall regression
model is statistically significant at a ¼ 5 percent and the data explain 36 percent
(adjusted R2
) of the model variance. The regression coefficient for the control variable
is statistically significant at a ¼ 5 percent; that is, the environmental uncertainty
variable (b ¼ 0.151) affects the antecedents of SCM practices. Next, in terms of
corporate entrepreneurship, we find that innovativeness has a significant positive
influence on SCM practices (b ¼ 0.175), in support of H1a. However, H1b venturing/
risk-taking (b ¼ 0.113) does not have a significant effect on the dependent variable.
As predicted, proactiveness (b ¼ 0.196) has a positive significant influence on SCM
practices, in support of H1c.
For social capital, the results support H2a and H2b, suggesting that supplier
relationships (b ¼ 0.275) and customer relationships (b ¼ 0.187) are positively related
to the dependent variable, SCM practices. Finally, H3 that argues for a positive
relationship between firm resource in terms of product advantages and SCM practices
is not supported (b ¼ 0.006). Overall, in addition to the control variable, four of the six
hypotheses were supported by the research results.