Social and structural mechanisms shaping the network architecture of instructional leadership research: Evidence from exponential random graph models

Instructional leadership has been extensively studied globally. Yet, prominent scholars consistently highlight ongoing methodological problems, observing that the field is ‘spinning intellectual wheels’ due to insufficient development of cumulative knowledge and the inability to provide a holistic structure and better insights. This study aims to advance the analysis of the Principal Instructional Management Rating Scale (PIMRS) research network by applying a novel analytical framework that integrates exponential random graph modelling (ERGM) and negative binomial regression analyses to uncover its structural characteristics. The analysis draws on a directed citation network of 66 empirical articles published between 1985 and 2025. The results revealed three key findings: (i) the ‘PIMRS research’ citation network exhibited a sparse structure with citation links concentrating around few core texts that have gained authoritative status in the field; (ii) citation ties were more likely among studies sharing the same country and thematic focus, whereas earlier publication and publication in core journals displayed higher citation counts; (iii) articles bridging different research clusters were limited, leaving cross-cluster information flow largely dependent on central studies. These findings provide empirical evidence of structural conditions associated with limited circulation and integration of knowledge within the PIMRS literature.