This study aims to investigate how geographic proximity shapes the flow of scientific knowledge, focusing on the persistence and variation of home bias in citations at both regional and national scales. We examine how this pattern differs across countries, disciplines and research environments.
The analysis draws on global scientific publication and citation data from 1990 to 2023, sourced from OpenAlex and filtered to Web of Science–indexed journals. Using large-scale bibliometric indicators, we quantify regional and national home bias and examine how it varies with cognitive distance between fields, regional research productivity and prior collaboration networks. The unit of analysis is the region, focusing on region-to-region citation flows estimated with ordinary least squares including cited-region and citing-region fixed effects, with standard errors clustered at the region-pair level.
Results show robust and persistent home bias in scientific citations worldwide, though its magnitude is heterogeneous. High-performing regions exhibit stronger local reliance, while scientifically advanced countries display weaker national-level bias. Cognitive distance and collaboration ties significantly shape the degree of localization, with closer intellectual proximity and stronger prior networks amplifying local spillovers.
Understanding the geography of knowledge flows can inform research funding strategies, regional innovation policies and the design of collaboration programs. Policymakers and research institutions can use these insights to strengthen cross-regional linkages, reduce unnecessary fragmentation and support more balanced scientific development.
Regions with limited research capacity may face structural disadvantages if knowledge circulates primarily within established hubs. Recognizing these dynamics can support efforts to broaden participation in global science and reduce disparities in access to influential knowledge networks.
This study provides a global, longitudinal assessment of citation-based home bias over three decades, integrating regional and national perspectives. By identifying the moderators that shape localized knowledge flows, it offers a nuanced understanding of how geography continues to influence scientific diffusion in an increasingly interconnected world.
