We integrate social network theories and information richness theory to understand which social structures are associated with effective knowledge transfer and higher productivity in face-to-face communication networks. Using novel data collection tools and methodologies, we record precise data on face-to-face interaction networks, tonal conversational variation and physical proximity among a group of IT configuration specialists over one month. By linking these data to detailed performance and productivity metrics we find that 1) network cohesion is associated with higher worker productivity, in contrast to previous findings for email data; 2) cohesion in face-to-face networks is associated with even higher performance during complex tasks, suggesting that social cohesion complements information-rich communication media when tasks are complex; 3) dominant structures for “latent” networks differ from “in-task” networks; and 4) face-to-face networks have more explanatory power than physical-proximity networks. Our research opens new avenues for the measurement of face-to-face interaction and information worker productivity.
Wu, Lynn; Waber, Ben; Aral, Sinan; Brynjolfsson, Erik; and Pentland, Alex (Sandy), "Mining Face-to-Face Interaction Networks Using Sociometric Badges: Predicting Productivity in an IT Configuration Task" (2008). ICIS 2008 Proceedings. Paper 127.