Academia.eduAcademia.edu

Figure 2 – uploaded by Dr.Patthareeya Lakpetch

See full PDFdownloadDownload figure
Then, structural model and hypothesis testing were performed. Seven of the ten hypotheses were confirmed with positive relationships. The proposed model had a significant mediating effect on knowledge transfer effectiveness. It can be concluded that seven out of the ten hypothesized relationships, H2-H6, H8, and H10, were found to be statistically significant at the 0.05 level in direct causal relationship, while three

Table 2 Then, structural model and hypothesis testing were performed. Seven of the ten hypotheses were confirmed with positive relationships. The proposed model had a significant mediating effect on knowledge transfer effectiveness. It can be concluded that seven out of the ten hypothesized relationships, H2-H6, H8, and H10, were found to be statistically significant at the 0.05 level in direct causal relationship, while three

Related Figures (36)

The role of partner complementarities in knowledge transfer effectiveness The role of partner complementarities in knowledge transfer effectiveness capabilities, firms can initiate and perform compe have done alone (Harrigan, 1985). Dyer and Singh (] contribute certain critical resources, such as tec both the assets and know-how at its disposal and t inputs. Accordingly, it is suggested that when t partner may be helpful in providing financing, co access to the market channel. Thus, the contributions of each partner are de he comparative advantage of different 1998) argue that while one he partners can synergis pooled resources and knowledge, this practice will reduce the risk of fa tical collaboration. Likewise, Emden et al. (2005) further state that to sustain the Based on RBV literature, by pooling complementary and supplementary resources anc titively on projects that they could no partner may hnological skills and assets, another mplementary technical know- how, 01 termined by y aligr ilure of the alliance partnership, the potential for creating synergistic value through co-deve opment alliance hinges on some similar aspects:selecting a partner with maximum potential tc collaborate that is strategic alignment with the partner; andselecting the partner with maximum potential for creating resources and competences in terms of valuable resources and knowledge complementarities. Thus, the following hypothesis is proposed:
The role of partner complementarities in knowledge transfer effectiveness The role of partner complementarities in knowledge transfer effectiveness capabilities, firms can initiate and perform compe have done alone (Harrigan, 1985). Dyer and Singh (] contribute certain critical resources, such as tec both the assets and know-how at its disposal and t inputs. Accordingly, it is suggested that when t partner may be helpful in providing financing, co access to the market channel. Thus, the contributions of each partner are de he comparative advantage of different 1998) argue that while one he partners can synergis pooled resources and knowledge, this practice will reduce the risk of fa tical collaboration. Likewise, Emden et al. (2005) further state that to sustain the Based on RBV literature, by pooling complementary and supplementary resources anc titively on projects that they could no partner may hnological skills and assets, another mplementary technical know- how, 01 termined by y aligr ilure of the alliance partnership, the potential for creating synergistic value through co-deve opment alliance hinges on some similar aspects:selecting a partner with maximum potential tc collaborate that is strategic alignment with the partner; andselecting the partner with maximum potential for creating resources and competences in terms of valuable resources and knowledge complementarities. Thus, the following hypothesis is proposed:
Notes: *p < 0.05; x? = 369.905, df = 305, p = 0.253, CMIN/DF = 1.210. GFI = 0.977, RMSEA = 0.040 The first measure was the y” of 369.905(df (x?) = 305). If the model is to provide a satisfactory representation of the data, it is important for the y7-value to be
Notes: *p < 0.05; x? = 369.905, df = 305, p = 0.253, CMIN/DF = 1.210. GFI = 0.977, RMSEA = 0.040 The first measure was the y” of 369.905(df (x?) = 305). If the model is to provide a satisfactory representation of the data, it is important for the y7-value to be
NN EIEN IES I IDE EE EEE I EE LI EE This principle is fundamental to Shrout and Bolger’s (2002) method for testing the statistical significance of an indirect effect. Empirical distributions resulting from bootstrap analyses are especially useful for estimating the confidence interval of a given population parameter to ensure the precision in estimating the confidence interval of each indirect effect. To perform the bootstrap analysis, we used the graphical interface of Amos 6.0 (Arbuckle, 2003), which is the one currently available software package we examined that directly produces bootstrapped percentile at 95 percent confidence intervals for indirect effects. We requested 1,000 bootstrap samples, drawn by default with replacement from the full data set of 240 cases. Note that a request for 95 percent confidence intervals must be used to override the Amos default that provides 90 percent confidence intervals. The confidence intervals show results of the bootstrapped percentile method reported by Amos. These values correspond to the 2.0th and 95th percentile
NN EIEN IES I IDE EE EEE I EE LI EE This principle is fundamental to Shrout and Bolger’s (2002) method for testing the statistical significance of an indirect effect. Empirical distributions resulting from bootstrap analyses are especially useful for estimating the confidence interval of a given population parameter to ensure the precision in estimating the confidence interval of each indirect effect. To perform the bootstrap analysis, we used the graphical interface of Amos 6.0 (Arbuckle, 2003), which is the one currently available software package we examined that directly produces bootstrapped percentile at 95 percent confidence intervals for indirect effects. We requested 1,000 bootstrap samples, drawn by default with replacement from the full data set of 240 cases. Note that a request for 95 percent confidence intervals must be used to override the Amos default that provides 90 percent confidence intervals. The confidence intervals show results of the bootstrapped percentile method reported by Amos. These values correspond to the 2.0th and 95th percentile
Parameter summary (Group number 1) Computation of degrees of freedom (Default model)
Parameter summary (Group number 1) Computation of degrees of freedom (Default model)
Regression Weights: (Group number 1 - Default model)
Regression Weights: (Group number 1 - Default model)
Correlations: (Group number 1- Default model)
Correlations: (Group number 1- Default model)
Variances: (Group number 1- Default model)
Variances: (Group number 1- Default model)
Squared Multiple Correlations: (Group number 1- Default model)
Squared Multiple Correlations: (Group number 1- Default model)
ML discrepancy (implied vs sample) (Default model)
ML discrepancy (implied vs sample) (Default model)
ML discrepancy (implied vs pop) (Default model)
ML discrepancy (implied vs pop) (Default model)
K-L overoptimism (stabilized) (Default model)
K-L overoptimism (stabilized) (Default model)
GLS discrepancy (implied vs pop) (Default model)
GLS discrepancy (implied vs pop) (Default model)
Baseline Comparisons
Baseline Comparisons
Related topics:
BusinessComputer ScienceBusiness and ManagementTourism and HospitalityOrganizational Analysis
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved