Business Networks and Performance: A Spatial Approach

... table 1 continued

5. Results

Table 2 shows the coefficients of estimating the tobit models for PMIR and PSE. It is
important to note that the two variables indicating networking activities for suppliers
and customers are highly significant. It is also evident that networking for finance does
not affect the firm’s use of locally produced material and/or its exporting activities.
Furthermore, the type of formal or informal agreements again does not have an impact
on the firm’s use of locally produced inputs and of its exporting activity. Surprisingly
enough the sector of economic activity does not exert a statistically significant impact.
Comparing the means of PMIR and PSE for the two values of the sectoral dummy
variable (SECTOR) does not reveal statistically significant differences in either the
mean (a non-parametric Mann-Whitney test) or the median (a non-parametric Kruskal-
Wallis test). Furthermore, the location of the business in the remote or the less remote
area is not statistically significant. Finally, the size of the business again is not
statistically significant. Other variables concerning entrepreneurial and enterprise
characteristics were also entered in the tobit model but did not significantly improve the
fit of the models.

Table 3 shows the estimated marginal effects of the independent variables on the PMIR
and PSE. Between two firms with all their characteristics equal at sample’s means, the

17

<<   <   14   15   16   17   18   19   20   >   >>