Answers are grounded in this paper and cite their sources.
Meta-regression analysis (MRA) has a long-standing history in literature and publication bias. However classical frameworks systematically overlook the network dependencies and spatial-type interactions. This paper addresses this methodological gap by introducing a spatial approach into MRA. Utilising comprehensive datasets from previously published empirical studies, we construct spatial weight matrices, extending from first-order and second-order contiguity networks, to put forth unobserved co-authorship. Methodologically, we apply spatial diagnostics and extensive Monte Carlo simulations. This approach not only enables the analysis of publication bias, as traditionally done, but also investigates whether studies/estimations exhibit a networking effect. The results appear promising regarding the applicability of MRA to spatial models.
No takes yet. Share an insight, caveat, or question.
Yeşilyurt et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: