Abstract:Carte, Chidambaram, and Becker (2006) conducted a longitudinal study of 22 self-managed virtual teams in order to better understand the differences in leadership behaviors engaged in by members of high-versus low-performing teams. We conducted a conceptual replication of this study in an attempt to examine the robustness of its findings. Our data were collected in a different country (Sri Lanka versus US), using different subjects (MBA versus undergraduate students) grouped into 24 teams that were collocated (rather than geographically distributed) but still using a CT and engaged in different tasks (judgement versus intellective) than the original study. Two of the five hypothesized results were replicated, indicating that certain characteristics of high-performing virtual teams seem universal. However, the remaining differences between the studies point to the influence of task, culture, and geographic dispersion of members in determining effective leadership behaviors in self-managed virtual teams.