Modular and nested structures in workplace mobbing within the construction industry: A bipartite network diagnostic approach

Abstract

This study investigates workplace mobbing dynamics in the construction industry using a bipartite network approach, inspired by ecological network analysis (ENA). Mobbing, as a form of persistent psychological harassment, poses significant threats to employee well-being and organizational efficiency, particularly in construction. Traditional approaches often overlook the relational patterns between perpetrators and victims. This research introduces a network-based perspective by constructing bipartite matrices that link perpetrators, mobbing reasons, and types of harassment experienced. Metrics such as nestedness (NODF) and modularity (Leading Eigenvector method) are used to analyze behavioral structures across 180 construction professionals. The findings reveal significant modularity in certain mobbing patterns, indicating that such behaviors are concentrated within specific subgroups. Nestedness analysis uncovers hierarchical structures, where specific mobbing types emerge as subsets of broader abuse patterns, suggesting systemic organizational issues. This study offers two key methodological contributions: (1) a bipartite network framework to identify and assess mobbing clusters, enabling proactive risk identification in construction settings; and (2) the integration of ENA metrics for analyzing clustered and hierarchical harassment behaviors. These insights provide a foundation for targeted interventions, supporting workforce well-being. Furthermore, the approach is compatible with management information systems (MIS), offering construction managers a data-driven tool to mitigate mobbing, enhance employee morale, and improve overall project performance.