Like all research methods, experiential simulations have limitations. To focus on the key aspects of the focal phenomena, some complexity is deliberately eliminated (Davis, Eisenhardt, and Bingham, 2007). In the case of Markstrat, there are restrictions. For example, firms cannot form alliances, make acquisitions, or enter new markets other than Vodites. Also, firms do not implement moves beyond the time and expense required to make them (Lant and Montgomery, 1992). Nonetheless, the simulation does enable a rich exploration of firm performance and key competitive moves and their implications for temporary advantage in distinct markets.
Sample and data sources
We conducted the simulation in a core masters-level class on strategy at a major U.S. west coast university. As a key component of the class, students participated in the simulation to gain hands-on, strategic experience. Groups consisting of three students formed each firm's top management team. Although Markstrat has been found to be very engaging and motivating due to its realism (Clark and Montgomery, 1996), we further motivated participants by using their firm's performance as a significant part of their course grade. Thus, participants were highly motivated to engage and perform well, just as managers of real firms are.
We gathered data during eight academic quarters, spanning 1999 to 2006. The data cover 32 industries (i.e., runs of the simulation) and 160 firms (i.e., five firms per industry). Although each of the five firms had a different starting position (i.e., relative competitive position) in an industry, these starting positions were constant across all 32 runs of the simulation. In all industries, the simulation ran for seven rounds over six weeks, with a consistent number of days between each round for teams to analyze, discuss, and make their moves. We archived all data generated by the simulations. The result is a uniquely comprehensive set of quantitative data to study competitive moves. It is our main data source.
We also gathered demographic data. The average age of participants is 24 years, and most participants have at least two to three years of work experience. Sixty-seven percent of participants are male. Forty-four percent of participants are from the U.S. We formed the teams through random assignment, but stratified them to ensure that each had members with diverse national backgrounds and work experiences. In an additional analysis (available from the authors) we used demographic variables to control for team heterogeneity (Hambrick, Cho, and Chen, 1996), and our original results held.
As in all experiential simulation research, we must use care when generalizing the results beyond the focal demographic. Prior research using Markstrat finds no empirical differences in behavior between teams of masters students and teams of executives (Lant and Montgomery, 1992; Clark and Montgomery, 1996). This may mitigate concerns about generalizing to real executive behaviors and actual industry competition. Further, because the average age of the participants in our study is relatively young, the participant teams may be especially representative of executive teams in technology-based ventures.
To supplement our main data source of simulation runs, we collected additional data (both quantitative and qualitative) on teams and their moves. First, we conducted 45-minute semi-structured interviews with eight participant teams. These interviews provide insight into the process of deciding competitive moves and reveal which moves participants thought were most crucial. Second, we reviewed all 160 team papers and final presentations completed in conjunction with the simulation. We then prepared written cases for 20 teams stratified by starting position and industry. In each industry, one case focuses on the competitive moves of a high-performing team and another focuses on the competitive moves of a low-performing team. The cases give rich understanding of the motivation for competitive moves. Third, we collected in-depth survey data for a sample of the teams—i.e., we surveyed 40 teams and asked them (1) What were the 2 to 3 most important decisions that you made this round? and (2) Why did you make them as you did? Fifty-five percent responded in all rounds, and all teams responded for at least three rounds. We coded and analyzed their responses to understand how teams viewed the competitive moves and their significance for competitive advantage. Of the 228 distinct competitive moves mentioned, 34 percent are R&D moves and 30 percent are market moves. Together, these data further confirm the importance of R&D and market moves (the focus of our hypotheses) for competitive interaction and competitive advantage.
Уважаемый посетитель!
Чтобы распечатать файл, скачайте его (в формате Word).
Ссылка на скачивание - внизу страницы.