Purpose This study aims to explore the heterogeneous effects of network embeddedness on project-based innovation performance (PIP) within cross-regional construction collaboration networks, focusing on differences in network effects between project owners (POs) and cross-regional project teams (CPTs). Design/methodology/approach Using 153 cross-regional construction projects completed in the Greater Bay Area between 2013 and 2023, this study applies social network analysis to construct lagged cumulative collaboration networks and measure five dimensions of network embeddedness. Hierarchical negative binomial regression models examine the linear and nonlinear effects of PO and CPT network positions on PIP. Findings The network analysis shows that CPTs have broader connections, stronger bridging positions, and greater local cohesion, whereas POs are more concentrated within particular network communities. For POs, degree centrality shows a monotonically decreasing relationship with PIP, betweenness centrality has a negative effect, structural holes have a positive effect, clustering coefficient has an inverted U-shaped effect, and closeness centrality shows no stable effect. For CPTs, degree centrality and structural holes show inverted U-shaped relationships; closeness centrality and clustering coefficient have positive effects, whereas betweenness centrality has a negative effect. Originality/value This study introduces a role embeddedness perspective by conceptualizing actor role as a boundary condition in translating network positions into PIP. Managerial embeddedness for POs and technical embeddedness for CPTs explain why the same type of network position can generate different innovation effects, clarifying the role dependence of these effects and informing differentiated network governance.
No takes yet. Share an insight, caveat, or question.
Miao et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: