Abstract
There is a rich debate in organizational theory about the contribution of supervisors to group process and performance, and about the span of control needed to make that contribution. In this paper, I summarize the debate and develop competing hypotheses. These competing hypotheses are tested using multisite survey and archival measures, and interpreted using qualitative data from the same study. I find that small supervisory spans improve performance through their positive effects on group process. In particular, supervisors with smaller spans achieved higher levels of relational coordination among their direct reports. Qualitative data suggest that supervisors with smaller spans achieved these results through working with, and providing intensive coaching and feedback to their direct reports.