suppliers/partners variable when explaining the explained
variance of the knowledge variable (56%).
In order to be able to compare the proposed hypotheses, the
precision and stability of the obtained estimations must be
assessed. For this purpose the Bootstrap technique was used
which offers the standard error and the t values of the parameters.
Following Roldán and Sánchez-Franco (2012), there was a
generation of a Bootstrap proof of 5,000 subsamples and a
one-tailed Student t distribution with (n-1) degrees of freedom,
where n is the number of subsamples to calculate the significance
of the path coefficients. From these levels, the significance of the
structural routes is obtained and, therefore, the support or not of
the hypotheses (Table 5). Specifically, the 4 hypotheses proposed
in the research have been confirmed with important levels of
significance.
The Stone–Geisser (Q2) test was used as a criterion to
measure the predictive relevance of the dependent constructs.
According to Chin (2010), if Q2 N 0, the construct has
predictive relevance. In our model all the Q2 values of the
dependent constructs display values above 0.41 (Table 5)
which is why it can be said the model has predictive relevance