基于Nbreg模型的SCI论文被引频次影响因素分析——中日比较研究的视角

Analysis on Factors Affecting Citation Frequency of SCI Articles Based on Negative Binomial Regression Model:From the Perspective of Comparison Between China and Japan

为更全面客观地认识论文被引频次、寻找提升中国SCI论文质量与影响力的途径,本文以中国和日本的SCI论文数据为基础,构建Nbreg负二项回归模型,分析中日SCI论文被引频次的影响因素,并从是否影响、影响方向与影响程度3个方面比较多维因素下中日SCI论文被引频次的影响异同点.研究表明,与日本相比,当参考文献数量越多,作者数量越多,合作方式为国内合作,论文语种为母语,文献类型为论文类、综述类时,中国SCI论文被引频次提高的可能性越大.据此,可通过建设更多高水平中文期刊、鼓励团队合作等方式来提升中国论文的被引频次,进而增强我国科研成果的学术影响力.

In order to understand the citation frequency of SCI articles comprehensively and seek ways to improve their quality, this paper chooses SCI articles of China and Japan respectively as the research foundation, and a negative binomial regression model is constructed to analyze the influencing factors of the citation frequency in the two countries, then compares the similarities and differences of factors between China and Japan from the three aspects of influence, influence direction and influence degree. The results show that, compared with Japan, when the number of references is more, or the number of authors is more, or the way of cooperation is domestic cooperation, or the language of the paper is mother tongue, or the type of literature is paper/review, the more likely the Chinese SCI papers are cited. Therefore, we can improve the citation frequency of Chinese papers by building more highlevel Chinese journals, encouraging team cooperation, etc., so as to enhance the academic influence of scientific research achievements in China.