Page 144 - WHERE WE WORK - Schlegelmilch

Ideas in place
= .76). There were no significant correlations of the control variables with idea sharing. For idea implementation, there were low correlations with gender, work hours, and role.
4.4.3 Hypothesis testing
In H1, we argued for the positive direct relationship between workplace transparency and idea sharing (Table 4.3). We found support for this hypothesis (B = 0.399, p < .001). In H3, we hypothesized that high workplace flexibility would weaken the positive relationship between workplace transparency and idea sharing. In support of H3, we found a significant, negative interaction between transparency and workplace flexibility (B = -0.261, p < .01) in relation to idea sharing. All results can be found in Table 3. The incidence rate ratio for workplace transparency (1.538 [1.285, 1.841]) suggests that idea sharing increased by approximately 54% with every unit increase in workplace transparency. However, the IRR for the higher order interaction term indicates a 23% decrease of idea sharing with every increase of the interaction.
Figure 4.2 shows the relation between task workplace transparency and idea sharing at low and high levels of workplace flexibility (note: negative binomial distribution). Further analysis of the difference between the slopes indicated that only under conditions of low workplace flexibility, the slope was significantly different from other slopes (Table 4.4). This suggests that the relationship between transparency and idea sharing was more positive under low workplace flexibility than under high workplace flexibility. These results thus support our hypothesis that transparency is positively associated with idea sharing when workplace flexibility is low.
Regarding idea implementation, we did not find support for H2, where we hypothesized that workplace transparency relates positively to idea implementation (B = 0.035, p = .13). The interaction between the moderator and workplace transparency was not significant (B = 0.042, p = .07). Figure 4.3 provides a summary of our findings.
142