Networked “bubbles”: study workgroups’ spatial adjacency preference using social network analysis methods

Purpose: In large organizations, space planning relies on workgroup leaders to indicate spatial adjacency preferences. However, many factors affect workgroups’ adjacency preferences, and it is not clear how the choices are made. This paper aims to explore whether the adjacency preferences are influenced by the collaboration relationship or constrained by the organizational structure. Design/methodology/approach: The authors studied a large company’s spatial adjacency planning with an in-depth analysis of its formal organizational structure and collaboration network. A sample of 183 managers was surveyed regarding groups with whom they want to be spatially adjacent and groups with whom they mostly interact. The data enabled us to test three structural factors related to adjacency preference: department affiliation, workgroup’s prestige and collaboration relation. The authors used the quadratic assignment procedure analysis to examine the correlations between network matrices. Findings: The results suggest that department affiliation and collaboration relations are significantly correlated to adjacency preferences. The authors did not find evidence supporting the notion that a workgroup’s prestige affects the preference. Among the three factors, collaboration relation best predicts the preference, which echoes Pena et al.’s (1977) argument that space planners should look into how groups function, rather than merely following the organizational chart. Originality/value: To the best of the authors’ knowledge, this research was the first to explore the choice of spatially adjacent workgroup through a detailed network analysis of the formal structure, work collaboration relations and other group-level characteristics. The findings have noteworthy cross-disciplinary implications, given that spatial proximity can be taken as a human resource management strategy to facilitate the overall interactions between workgroups.