ABSTRACT
Our first goal in this paper is to demonstrate the potential effect of the choice of research design on research findings and the interpretation of these findings. To that end, we have chosen to focus on a relatively simple issue, that of the magnitude of alliance formation and its geographic distribution. We have chosen to focus on this particular issue because studies that explore the broad demographic trends of alliances [e.g., Ghemewat, Porter and Rawlinson 1986; Beamish and Delois 1997] provide the framework necessary to test theories developed through conceptual work [Hennart 1988; Buckley & Casson 1988] and case studies [Hamel 1991; Parkhe 1991]. Consequently, it is critical to comprehend potential biases that may affect the results of the larger studies.
