Pathway modeling and quantitative evaluation of rural sports development’s impact on rural revitalization via multi-source data integration

Abstract

Against the backdrop of the deepening implementation of China’s rural revitalization strategy, rural sports have been recognized as a critical vehicle for enhancing rural social and cultural development. Their advancement has been closely associated with economic growth, improved quality of life, and the promotion of social cohesion in rural areas. With the rapid evolution of information technologies, diverse data streams—ranging from rural sports infrastructure and resident participation to industrial and economic indicators—have emerged, offering a rich foundation for analyzing the influence of rural sports development on rural revitalization. However, the complex and latent interconnections between the two remain insufficiently understood. The challenges of effectively integrating multi-source data and accurately identifying causal pathways have yet to be fully addressed. Existing studies often rely on single-source data, failing to capture the multidimensional nature of the relationship. Moreover, conventional regression-based approaches have demonstrated limited capacity in discerning causality and handling endogeneity, while current methods of data fusion tend to involve simplistic concatenation, lacking mechanisms for deep collaborative analysis. To address these limitations, a causal structure inference model suited for large-scale multi-source data integration was developed in this study to investigate the pathways through which rural sports development impacts rural revitalization. Specifically, structural equation modeling (SEM) was employed to explore latent influence relationships, followed by the introduction of a directed acyclic constraint within a low-dimensional factor node space to determine causal directions and hierarchical structures. Finally, specificity modules were constructed to enhance the integrative regulation of influencing factor nodes, thereby enabling effective causal structure inference within large-scale fused data. The principal innovation lies in the establishment of a causal inference framework tailored to large-scale multi-source data, integrating SEM, a directed acyclic constraint, and specificity modules. This framework addresses the limitations of single-source data and traditional methods, significantly improving the accuracy of causal relationship identification and the capacity for integrated regulation. The proposed approach provides a novel methodological foundation for revealing the intrinsic links between rural sports development and rural revitalization. This study is situated within the differentiated development contexts of rural areas across eastern, central, and western China and is positioned at the intersection of rural development and sports management. The constructed causal structural inference model applicable to large-scale, multi-source integrated data can precisely investigate the pathways by which rural sports development contributes to rural revitalization.