Can wearable sensors be used to capture engineering design team interactions? An investigation into the reliability of sociometric badges

Engineering organizations around the word are increasingly becoming team-based due to a teams' ability to generate solutions to complex problems. This is thought to be attributed to the "wisdom of the collectives" where teams outperform the sum of their individual members. Despite heavy emphasis of teamwork in engineering design, our understanding of how to cultivate teamwork skills is poorly understood. This is due in part to the fact that research on engineering design teams is often based on 'snap-shot' ethnographic methods that do not account for the dynamic changes that happen over the course of a project. Research that does account for these interactions are hindered by the human processing required to code and analyze the immense amounts of video data acquired through team studies. Recent technological advances in the way of sociometric badges provide a potential avenue to explore intricate communication patterns and help researchers identify when and how team interventions should be developed. However, these technologies have not been validated for use in the dynamic context of engineering design teams. Thus, the current study was developed to examine sociometric badges for their accuracy, precision, and ability to determine speech dominance in engineering design teams. Our results show that sociometric badges can accurately capture total speaking time. However, the results also show that environmental conditions can impact their accuracy. In addition, we found that sociometric badges did not capture speech precisely and often overestimated when a team member was speaking. These results support the use of sociometric badges for capturing high level team interactions (e.g. participation and total speaking time) in engineering design research.