Source-linked AI summary
China's Shrinking Home Bias and Rising Disruptive Impact: Evidence from a Global Citation Network Analysis
Nan Deng, Zheng Ma, Xiaohui Liu
TL;DR
Concerns remain that China’s rapid publication growth and domestic citation practices may inflate its apparent scientific influence, while the respective roles of citation preference and publication volume remain underexplored. The paper analyzes a Web of Science citation network with a network reshuffling null model and persistent disruption measures, finding that China’s home citation bias is declining and its disruptive research is converging toward Western levels.
Problem
The paper examines whether China’s elevated home citation rate reflects preferential citation behavior or the mechanical effect of rapidly expanding publication volume.
Method
The study analyzes a Web of Science citation network from 1980 to 2025 using a network reshuffling null model and the persistent disruption framework.
Results
China’s home citation bias is lower than commonly assumed and declining, while its papers are increasingly capable of producing paradigm-shifting work and approaching Western persistent-disruption levels.
Takeaways & Limitations
Chinese science appears increasingly integrated into the global knowledge network rather than characterized by intensifying excessive self-citation.
Takeaways & Limitations
Country attribution uses the first author’s primary affiliation, and the null model controls publication volume but not field-level citation norms.
Abstract
from arXiv · showhide
China has become the world's largest producer of scientific publications, yet concerns persist that this growth is inflated by excessive domestic citation practices. In this study, we analyze a citation network of over 45 million publications from Web of Science (1980-2025) to investigate China's home citation bias and research impact. Using a network reshuffling null model to control for the structural effect of publication volume, we find that China's home citation bias is less pronounced than commonly assumed and has been steadily declining over the past two decades. Chinese researchers do not exhibit a significantly stronger home citation preference than other major countries, indicating increasing internationalization rather than insularity. Furthermore, using the persistent disruption framework, we show that Chinese papers are converging toward American papers in their capacity to produce paradigm-shifting work. These findings challenge prevailing narratives about Chinese scientific home bias and suggest that China's advances in research impact.
1. Introduction
China’s rapid growth in scientific output and influence has prompted concerns that domestic citation practices inflate its apparent strength. Existing research documents widespread home citation patterns but leaves unresolved how much China’s elevated rates reflect citation preference versus publication-volume effects.
- China surpassed the United States in scientific research output by 2016, becoming the world’s largest producer.
- China’s growing presence in high-impact research and rising submission and acceptance rates have accompanied its expansion in publication volume.
- 62% of citations to China’s top 10% most-cited articles came from within China, compared with 24% for the United States.
- China’s top-journal influence ranking fell from second to fourth when national self-citations were excluded, fueling skepticism about its scientific strength.
- Home bias is widespread across countries, while national citation patterns vary by field, citation impact, and research discipline.
- Publication volume mechanically raises a country’s self-citation rate, but China’s preference-based contribution remains underexplored.
2. Results
China’s observed domestic citation rates rise with its expanding publication output, but null-model comparisons show that relative home bias is modest and declining. Persistent-disruption measures likewise indicate convergence toward Western research impact.
- Publication and citation impact: China’s publication output has grown steadily to become the world’s largest, while its citation impact has also improved.CNCI controls for differences in research field and publication year.
- Observed home citation patterns: China’s observed home reference and home citation rates are high, with home citations consistently exceeding 50%.The home reference rate rose notably over the past decade, while China maintained the highest home citation rate among major countries.
- Null-model adjustment: The network-reshuffling null model preserves publication-year and degree-distribution constraints to estimate home-citation rates expected under random citation behavior.This separates publication-volume effects from citation preferences.
- Relative home bias: China’s relative home reference tendency remained above 1 but declined steadily, converging toward the random baseline as publication output expanded.Relative home reference tendencies for the USA, India, the UK, and Germany remained stable.
- Relative home bias: China’s relative home citation ratio was comparatively low and declining, while remaining above 1 relative to null-model expectations.The pattern indicates shrinking rather than intensifying home bias.
- Interpretation: Overall, the findings challenge claims of excessive Chinese self-citation and suggest increasing integration into the global knowledge network.The authors attribute the apparent home-bias pattern partly to publication scale and describe the declining bias as consistent with internationalization.
- Research disruption: Persistent disruption shows China’s reference disruption increasing toward Western levels and its citation-disruption gap narrowing.The framework captures both departure from cited references and influence on citing papers.
- Limitations: The conclusions are qualified by first-author country attribution and by a null model that controls publication volume but not field-level citation norms.Some domestic concentration may reflect legitimate expertise rather than genuine preference.
4. Method
The study measures domestic citation orientation with two complementary rates and tests observed patterns against a structurally constrained reshuffled network. It also evaluates paradigm disruption using a two-dimensional framework and notes that the underlying Web of Science data cannot be publicly shared.
- Data: The dataset contains over 44 million publication nodes and over 1.8 billion citation links from Web of Science records spanning 1980–2025.Articles and reviews with incomplete metadata were removed before constructing the directed citation network.
- Home Reference & Citation Rate: Home Reference Rate measures the share of a paper’s cited references originating from its assigned country, while Home Citation Rate measures the share of incoming citations from that country.Both indicators are proportions bounded in [0,1] and differ from citations received per paper.
- Null Model: The null model randomly rewires citation connections while preserving focal and reference publication years, network degree constraints, and other structural features.It generates expected home reference and home citation rates under random citation behavior for comparison with observed rates.
- Persistent Disruption: Persistent disruption combines reference disruption, which captures departure from cited references, with citation disruption, which captures whether citing papers consolidate or disrupt the focal work.Persistently disruptive papers have high reference disruption and low citation disruption; lower citation disruption indicates stronger consolidation of foundational status.
- Open Science Practices: The study’s raw Web of Science bibliometric data cannot be publicly shared because they were obtained under an institutional license.This licensing condition limits direct public access to the underlying records.