Results from hierarchical analysis confirmed the formulated hypotheses for the research. The findings showed the following:
Effects of organization contexts on knowledge sharing, KMS, and Innovation
The results from hierarchical regression analysis showed that organization contexts do have an effect on knowledge sharing, KMS, and innovation. Moreover, it also indicated that knowledge sharing partially mediates the relationship between organization contexts and innovation.
According to Falk and Miller (2002), the explanatory power of R2 value greater than 10 percent is acceptable. Therefore the model that resulted from hierarchical regression results which indicated that demographic and organization contexts (i.e. information technology, top management, support, collaborative culture, and organization structure) explained 25.3% of the variance in knowledge sharing is acceptable. According to this finding it can also be concluded that since this R2 value is closer to .33, the effects organization contexts on knowledge sharing is moderate on the basis of guidelines provided in Karim (2009) which suggested that R2 value of .66 , .33, .19 should be considered as substantial, moderate and weak, respectively.
Since organization contexts and demographics explained 65.5% of the variance in KMS, it can be concluded that this model is not only acceptable but also that they have substantial effect on KMS on the basis of the guidelines provided by (Falk & Miller, 2002; Karim, 2009).
Similarly, organization contexts and demographics explained 39.5% of the variance in innovation. Although the model is acceptable, the effects of these factors on innovation can be considered as moderate according to guidelines provided in Karim (2009).
The model fit in hierarchical regression analysis for organization contexts’ effects on knowledge sharing, KMS and innovation are all significant respectively at (p<.001).
In addition, hierarchical regression also revealed that knowledge sharing and individuals’
demographics together explained 22% of the variance in innovation. This is acceptable on the basis of Falk and Miller (2002)’s guidelines. The variance in innovation explained by knowledge sharing was relatively low compared to other studies that have examined the relationships between these two variables. For example, Liao (2006) reported that 64% of the variance in firm
115
innovation was explained by knowledge sharing. According to Karim (2009), the effects of knowledge sharing on innovation is somewhat weak.
Similarly, the hierarchical regression findings indicated that individuals’ demographics together with knowledge sharing explained 28.3% of the variance in KMS. The model is also acceptable on the basis of Falk and Miller (2002)’s guidelines. In addition the effects of knowledge sharing on KMS is moderate on the basis of (Karim, 2009)’s guidelines.
The model fit (s) for knowledge sharing effects on KMS and innovation are all significant (p<.001). At last the results showed that organization contexts had a greater impact on KMS than knowledge sharing and innovation.
Effects of demographic factors on knowledge sharing, KMS and innovation
Hierarchical regression results revealed that individuals’ demographics alone explained 14.5% of the variance in knowledge sharing. Although their effect on knowledge sharing is relatively weak according to Karim (2009)’s guidelines, the model is acceptable on the basis of Falk and Miller (2002)’s guidelines and has model fit significant at (p<.01). With regards to demographic factors impact on knowledge sharing, the findings of hierarchical regression confirmed previous studies’ results and arguments on their relationships. In relations to age, the results showed no association between age and knowledge sharing. This is in accordance with Ojha (2005) and Watson and Hewett (2006) studies which also did not find any causal relationship between age and knowledge sharing. However, it somewhat contradicts Gumus (2007) study which found that age had an impact on knowledge collecting and not on knowledge donating and revealed that individuals that are in the 31-39 demographics are not willing to collect knowledge. When it comes to gender, no causal relationship was found between gender and knowledge sharing. This is also compatible with Ojha (2005) and Watson and Hewett (2006) investigations that did not report any associations between these two variables. In addition, it does not provide support for research conducted by Lin (2006) which indicated that gender may have an effect on knowledge sharing as they implied that women may be more willing to share knowledge than men. Hierarchical regression results also showed that educational levels had no effect on knowledge sharing. This confirmed results obtained in Ojha (2005) study which did not find any relationship between educational level and knowledge sharing. Overall, this study’s results do not provide support for Riege (2005) study which suggested that age differences,
116
gender differences, and differences in educational levels can have an effect on knowledge sharing. Since this study measured knowledge sharing in terms of tacit knowledge sharing and explicit knowledge sharing, the non significant effect of gender, age on knowledge sharing is partly compatible with Holste and Fields (2009) study who also found no relationship and no causal effect of these factors on tacit knowledge sharing. However, unlike these authors’ study, this study found that tenure was positively associated with knowledge sharing (i.e. tacit and explicit).
However, with respect to organization tenure, this study found a significant and positive relationship with knowledge sharing. Although this finding is in accordance with Bordia et al.
(2006), and Watson and Hewett (2006) study findings, it is contrary to Ojha (2005)’s study which indicated a significant negative relationship between organization tenure and knowledge sharing. As knowledge sharing is assessed as tacit and explicit knowledge sharing, the positive effect of organization tenure on knowledge sharing is somewhat not compatible with the study conducted by Holste and Fields (2009) which reported that tenure was not associated with tacit knowledge sharing.
Hierarchical regression showed that individuals’ demographics alone did not have any significant effect on innovation. However, surprisingly organization was found to have a positive effect on innovation. In relation to innovation, a study conducted by Carmeli, Meitar, and Weisberg (2006) showed that tenure was positively associated with innovative behavior.
Therefore it is probable that organization tenure may have an effect on innovation as the results suggested. The findings also indicated that age, gender and educational levels had no effect on innovation.
Hierarchical regression analysis also revealed that individuals’ demographics alone explained 16.1 % of the variance in KMS. Although their effects on innovation are relatively weak according to guidelines in Karim (2009), this model is acceptable on the basis of (Falk &
Miller, 2002). This model showed also an overall fit that is significant at (p<.01). Organization tenure was also found to have a positive and significant effect on KMS while education was found to be negatively at significantly related with KMS. Contrary to previous studies (Taylor et al., 2004) that suggested that gender may have an effect on KMS, this study did not found any significant relationship with KMS.
117