因果推断视角下OA对论文影响力提升效应研究——以AI领域Altmetrics为例

• 作者： 王旭 ¹ 戴春燕 ¹
• 作者单位：
  1.  燕山大学经济管理学院
• 发布时间：2024-12-10

摘要: 【目的/意义】深入理解OA学术论文（简称：论文）影响力指标间关系并分析OA与论文影响力指标间因果关系，对于揭示OA论文影响力传播途径的内在优势、提升论文质量、促进学科间的知识交流与共享具有重要意义。【方法/过程】本文以人工智能领域论文为研究样本，从OA与非OA两层面出发，采用描述性统计分析、负二项回归分析以及双重差分模型，综合考虑社会影响力和学术影响力的关键指标，揭示人工智能领域相关论文指标数据分布规律与特征、分析OA论文影响力指标间关系并挖掘OA优势。【结果/结论】（1）人工智能领域OA与非OA论文各指标值差异较小，指标数据分布较为集中，推特提及量呈现右偏分布，而其余变量呈正态分布。OA论文各指标均值均高于非OA论文指标。（2）无论是OA论文还是非OA论文，各个Altmetrics指标对Dimensions被引次数均产生显著的正向影响，但OA论文的影响程度显著更高。其中，OA模式下，专利提及量对Dimensions被引次数的影响程度最为显著，而政策提及量、新闻提及量、推特提及量对Dimensions被引次数的影响程度依次减弱。（3）OA与论文新闻提及量和推特提及量存在显著因果关系，即OA显著提升了论文在新闻和推特平台上的提及量。但OA Altmetrics优势并非持续稳定，而是随时间呈现逐渐下降的趋势。【创新/价值】从相关性分析与因果推断的视角出发，逐步挖掘OA优势。在此基础上，本文进一步提出了针对性的策略，以期提升人工智能领域论文的可见性和影响力。

Abstract: 【Purpose/Significance】 An in-depth understanding of the relationship among the impact indicators of OA academic papers (referred to as: papers) and analyzing the causal relationship between OA and the impact indicators of papers are of great significance in revealing the intrinsic advantages of the dissemination pathways of OA papers' impact, improving the quality of papers, and facilitating the exchange and sharing of knowledge among disciplines. 【Method/Process】 This article takes the papers in the field of artificial intelligence as the research sample, and adopts descriptive statistical analysis, negative binomial regression analysis and Difference-in-Difference (DID) model from both OA and non-OA levels. At the same time, this article comprehensively considers the key indicators of social impact and academic impact, reveals the indicators data distribution pattern and characteristics of AI papers, analyzes the relationships among the impact indicators of OA papers and explores the advantages of OA. 【Results/Conclusion】 (1) The difference in the values of each indicator between OA and non-OA papers in the field of AI is small, and the distribution of indicator data is more concentrated. Twitter mentions show a right-skewed distribution, while the rest of the variables show a normal distribution.The mean values of each indicator for OA papers are higher than the indicators for non-OA papers. (2) Whether OA papers or non-OA papers, each altmetrics indicator has a significant positive influence on the number of citations to Dimensions, but the influence of OA papers is significantly higher. Among them, patent mentions have the most significant influence on Dimensions citations in OA mode, while policy mentions, news mentions, and twitter mentions have weaker influence on Dimensions citations in that order. (3) There is a significant causal relationship between OA and news mentions and twitter mentions of papers, i.e., OA significantly enhances the mentions of papers on news and twitter platforms. However, the OA altmetrics advantage is not consistently stable, but shows a gradual decline over time.【Originality/Value】 This article gradually explores the advantages of OA from the perspectives of correlation analysis and causal inference. On this basis, this article further proposes targeted strategies in order to enhance the visibility and impact of papers in the field of artificial intelligence.

• 来自： 王旭
• 分类： 信息资源管理
• 稿件状态： 已投稿期刊
• 引用： PSSXiv:202412.00402 (或此版本 PSSXiv:202412.00402V1)
  DOI:10.12451/202412.00402
  CSTR:32012.36.PSSXiv.202412.00402
• 推荐引用方式： 王旭,戴春燕.因果推断视角下OA对论文影响力提升效应研究——以AI领域Altmetrics为例.哲学社会科学预印本平台:https://zsyyb.cn/abs/202412.00402.[PSSXiv:202412.00402V1] (点此复制)