Decoding Conflict: Linguistic Markers of Disagreement in Hackathon Design Teams

Abstract

Abstract. In design teams, conflict can serve as a catalyst for innovation if managed effectively. While prior research has explored the causes and outcomes of conflict, less is known about how language reflects conflict dynamics in real-time. This study examines linguistic markers of conflict in teams at hackathons—fast-paced, high-stakes environments where disagreement is frequent. Using deductive coding, we first identify instances of conflict in transcript datasets from two hackathon teams. Then, using text analysis with the Linguistic Inquiry and Word Count (LIWC) tool, we analyze differences in language patterns between conflict and nonconflict interactions. One hackathon team studied was composed of strangers, and the other of pre-existing teammates, allowing us to examine the role of familiarity in conflict communication. Our findings reveal distinct linguistic characteristics during conflict, including reduced negative emotion and prosocial language, as well as differences in sentence complexity and cohesion based on team familiarity. These insights not only contribute to the growing body of research on team dynamics but also demonstrate the potential for deriving team insights from transcripts captured in situ at hackathons.