图书馆杂志

图书馆杂志 ›› 2025, Vol. 44 ›› Issue (408): 98-109.

• 信息管理 • 上一篇    下一篇

国际合作中国家间科研主导力及影响因素研究——基于DBLP数据的空间计量分析

章成志 胡子乐 (南京理工大学经济管理学院信息管理系)   

  • 出版日期:2025-04-15 发布日期:2025-04-24
  • 作者简介:

    章成志 南京理工大学经济管理学院,教授,博士生导师。研究方向:信息组织、信息检索、文本挖掘与自然语言处理。作者贡献:完善研究思路与方法、论文撰写与修改。E-mail:zhangcz @njust.edu.cn 江苏南京 210094

    胡子乐 南京理工大学经济管理学院,硕士研究生。研究方向:科学计量学与科学学。作者贡献:完成实验、论文撰写与修改。江苏南京 210094

    赵 毅 南京理工大学经济管理学院,博士研究生。研究方向:科学计量学与科学学。作者贡献:设计研究方案、提出多轮修改意见。 江苏南京 210094

Scientific Leadership and Influencing Factors in International Collaboration: A Spatial Econometric Analysis Based on DBLP Data

Zhang Chengzhi, Hu Zile, Zhao Yi (Department of Information Management, School of Economics and Management, Nanjing University of Science & Technology)   

  • Online:2025-04-15 Published:2025-04-24
  • About author:Zhang Chengzhi, Hu Zile, Zhao Yi (Department of Information Management, School of Economics and Management, Nanjing University of Science & Technology)

摘要:

在全球化背景下,国际科研合作备受关注。现有研究通常将团队中的合作关系视为同等, 忽视层级关系。当前的科研合作组合多样,仅从论文作者位次反映主导—参与关系存在局限。通过合作网络分析研究团队,更能反映出节点的重要性。此外,已有研究表明,距离因素会影响国家间合作强度,但忽略空间因素对科研主导力的影响,会影响模型估计结果的准确性。为此,本研究从网络全局视角,分析计算机科学学科内国际科研合作中国家科研主导力,利用网络节点间相互依赖关系确定国家间主导地位,并借助空间计量模型剖析影响因素。研究发现,科研主导力在空间上呈现出明显的聚集效应,高科研主导力国家集中在欧美地区。然而,一些欧洲国家虽有较多合作伙伴且参与的合作论文较多,但未被列为高科研主导力国家,表明它们可能处于参与地位。空间回归结果显示:一个国家的科研主导力受其周边国家、地区科研主导力水平及其他因素的影响。国民生产总值和教育支出对科研主导力有积极影响,人口规模则有负面影响。引入国家科研主导力,为深入了解国际合作中的主导地位提供新视角,对推动国家间合作共赢、提高创新协同效率具有重要理论和实践意义。

Abstract: In the context of globalization, international scientific collaboration has garnered significant attention. Existing research often assumes equal status among collaborators, neglecting the hierarchical dynamics within research teams. The diversity of current scientific collaborations means that authorship alone is insufficient to fully capture leadership roles; analyzing research teams through collaborative networks can better identify key nodes. Additionally, while existing studies have shown that geographic distance can affect the intensity of international collaboration, they have largely overlooked the impact of spatial factors on scientific leadership, which can affect the accuracy of model estimations. Therefore, this study analyzes the scientific leadership of countries in international scientific collaborations within the field of Computer Science from a global network perspective, identifying leading countries through network centrality and applying spatial econometric models to analyze influencing factors. Findings reveal a notable spatial clustering effect, with countries demonstrating strong scientific leadership primarily concentrated in Europe and North America. However, some European countries, despite having numerous partners and participating in many collaborative papers, are not classified as scientific leaders, indicating a more participatory role. Spatial regression analysis shows that a country’s scientific leadership is influenced by the scientific leadership levels of its neighboring countries. Gross domestic product (GDP) and education expenditure positively impact scientific leadership, while population size has a negative impact. Introducing the concept of national scientific leadership provides a new analytical perspective for understanding the leading positions in international collaboration, which is of significant theoretical and practical importance for promoting win-win cooperation and enhancing collaborative innovation efficiency among countries.