Unveiling Latent Dispute Causes in Construction: A Social Network Perspective

Abstract

Construction disputes often arise not from isolated incidents but from interconnected latent causes, underlying systemic issues that remain hidden until activated by project dynamics. This study explores these latent dispute causes within the Kenyan construction industry using Social Network Analysis (SNA). A dataset of 30 documented dispute cases was analyzed, supported by expert validation, to identify 14 latent causes such as inequitable contractual practices, ambiguous contract terms, conflicts of interest, and incomplete designs. The analysis was conducted using the University of California Network (UCINET) software, which enabled the generation of adjacency matrices and calculation of centrality measures to map the interdependencies and influence among dispute causes. Results indicate that disputes often stem from power asymmetries, unclear scopes, and overlapping roles, particularly prevalent in Design-Bid-Build procurement systems where a single agent handles design, supervision, and certification. Centrality scores revealed inequitable practices and contract ambiguities as the most influential latent factors. The study highlights the inadequacy of traditional linear models in capturing these non-linear and weighted relationships. Key recommendations include clear role segregation, adaptive risk management, transparent contract structuring, and integration of predictive analytics like SNA. This research provides a robust, system-level framework for proactively identifying and mitigating latent dispute risks in construction.