摘要
[目的/意义]不同合作者在科研合作中扮演不同角色,承担相应职责。区分科研合作对象中的不同角色,对于科研人才评价和人力资源配置具有重要意义。本文基于H指数的测算方式,使用合作次数来识别科学家合作关系中的核心合作者。[方法/过程]以中国与G7国家计算机领域高产科学家为研究对象。首先,识别出核心合作者并分析其规模与占比。其次,基于H指数拟合公式提出核心合作者人数的估算公式。最后,利用该式比较理论值与实际值的差异。[结果/结论]研究发现,在核心合作者的规模与占比上,各国核心合作者人数普遍为3~7人,且比例均不超过10%。核心合作者人数可使用发文量与篇均合作者人数的幂次方乘积来估算,美国、德国和英国的核心合作者占比更低,人才的流动和交流较为频繁,而意大利、日本和中国的核心合作者人数占比偏高,表明人才流动性不足,科研合作相对固化。
[Purpose/Significance]Different collaborators play different roles and assume corresponding responsibilities in scientific research collaboration.Distinguishing the different roles in research collaborators is important for the evaluation of research talent and human resources allocation.Previous studies have defined the roles of collaborators from multiple perspectives,both qualitative and quantitative,but lack a simple and efficient way to identify core collaborators.In this paper,we use the number of collaborations to identify core collaborators in scientists'collaborative relationships based on the H-index measure,which is a very easy to calculate and intuitively understandable method.[Method/Process]Using the OpenAlex database as a data source,an empirical analysis of approximately 5.05 million journal papers in the field of computing in China and G7 countries over 20 years(2000-2021)was conducted.First,the core collaborators of highly productive scientists were studied and their collaboration characteristics were analyzed from the perspective of size and share.Second,based on the H-index fitting formula proposed by previous authors,a formula for estimating the number of core collaborators based on the number of publications and the average number of collaborators per article was proposed.Finally,the formula was used to compare the differences between the theoretical and actual values of the number of core collaborators across countries.[Results/Conclusions]The study found that in terms of size and proportion of core collaborators,China had the highest average total number of collaborators among highly productive scientists,followed by the USA,Germany and the UK,while Italy had the lowest.The number of core collaborators was generally 3-7 across countries,with China and Italy having a higher rate of cooperation and the UK,France and Canada having a lower rate of cooperation.In terms of the number of core collaborators as a percentage,no country has more than 10%,with Italy having the highest percentage of core collaborators at 7.42%,followed by Japan,France and Canada,while the US has the lowest percentage of core collaborators.In terms of the total number of collaborators,there is no significant difference between China,the US and Germany,while there is a significant difference among all five other countries.In terms of the number of core collaborators,China is not significantly different from Italy and is significantly different from all other six countries.The number of core collaborators can be estimated by using the formula of the product of the number of publications and the power of the average number of collaborators per article,which has a good fit of 0.8 or more.Among China and the G7,the US,Germany and the UK have a lower proportion of core collaborators,with more frequent mobility and exchange of talent,while Italy,Japan and China have a higher proportion of core collaborators,indicating a lack of talent mobility and a relative consolidation of research collaboration.
作者
罗雯
胡志刚
LUO Wen;HU Zhigang(Institution of Science of Science and S.&T.Management,Dalian University of Technology,Dalian 116024)
出处
《农业图书情报学报》
2023年第4期48-58,共11页
Journal of Library and Information Science in Agriculture
基金
辽宁省“兴辽英才”青年拔尖人才“科技自立自强背景下的开放合作新模式研究”(XLYC2007149)。
关键词
科研合作
核心合作者
科研合作模式
H指数
scientific collaboration
core collaborators
scientific collaboration model
H-index
