Research data of the study "Service Innovation Performance: The Role of Customer Involvement, Dynamic Capability, and their Interactions"
Description
This is the data of the study "Service Innovation Performance: The Role of Customer Involvement, Dynamic Capability, and their Interactions". In this study , six hypotheses were proposed. H1:CI has a positive impact on the operational SIP; H2:CI has a positive impact on customer-related SIP. H3:Dynamic capability, including (a)adaptive capability, (b)absorptive capability, and (c)innovative capability, has a positive impact on the operational SIP. H4:Dynamic capability, including (a)adaptive capability, (b)absorptive capability, and (c)innovative capability, has a positive impact on the customer-related SIP. H5:The interactions between CI and DC (including (a)adaptive capability, (b)absorptive capability, and (c)innovative capability) positively affect the operational SIP. H6:The interactions between CI and DC (including (a)adaptive capability, (b)absorptive capability, and (c)innovative capability.) positively affect customer-related SIP. This study adopted a questionnaire approach to collect this data. Employing data collected from 298 Chinese companies, and using a hierarchical regression analysis, the study finds that customer involvement positively affects both operational and customer-related SIP, dynamic capabilities promotes the operational SIP, but only two dimensions of dynamic capability (i.e. absorptive and innovation capability) improve customer-related SIP. In addition, the interaction between customer involvement and absorptive capability is positively co-related with operational SIP, and the interaction between customer involvement and adaptive capability positively affects customer-related SIP.
Files
Steps to reproduce
Firstly, we used Amos.20 to conducted a confirmatory factory analysis to test the reliability and validity of the questionaire. Secondly, discriminant validity was tested by examining the square root of the average variance extracted for each construct. Thirdly, to test for the possibility of common method bias, we conducted post hoc Harman’s single-factor test. And the result shows that common method bias is not a major concern. Then, descriptive and correlational analyses were conducted. The results were shown in Table 3. Finally, we did hierarchical regression analysis by SPSS.20 to test hypotheses, and the results of regression are shown in Table 4. In order to limit the potential multicolinearity of interaction terms, we mean centered the independent variables before constructing the interaction terms. Model 1 and model7 contains the control variables and the subsequent models add the main and moderating effects. We discuss the results of the full model, Model 6 and Model 12.