Discussion of Model Findings
Having presented the findings of the model testing, the next step is to interpret the findings. First, the smallest common denominator from the survey results for the six sub-samples will be extracted. This will crystallize out those findings that appear to be most generalizable for a larger population. Second, the group results will be examined more closely. In particular, differences in the IS sourcing behavior between countries (Germany versus USA), industries (Finance versus Machinery), and IS functions (Application Development versus Maintenance) will be pointed out and interpreted. Third, the major theoretical contributions will be summarized. Fourth, practical implications from the study results will be presented. Fifth, the theoretical and methodological limitations of this study will be identified and, finally, avenues for future research will be suggested.
KeywordsTransaction Cost Intellectual Capital Finance Industry Maintenance Work Transaction Cost Theory
Unable to display preview. Download preview PDF.
- 18.Moreover, standard computer aided design (CAD) tools are well established. It should be kept in mind, however, that these are average tendencies. Moreover, there are still ares in which standard software is less established, e.g., in the area of product data management (PDM).Google Scholar
- 19.For an overview of TCE research on vertical integration and make-or-buy in general, see Shelanski and Klein (1995). For an overview of IS outsourcing research that tested “market failure arguments”, see Dibbern et al. (2003).Google Scholar
- 20.It has to be noted that IS research appears to be predominantly seen as empirical research by Lee (1999). This neglects the more constructive type of IS research, which is concerned with actually building IS. This includes important issues like developing prototypes of particular IS components, designing IS architectures, or establishing principles of IS development (Krcmar, 1999; Lyytinen, 1999).Google Scholar