Journal of Data and Information Science ›› 2021, Vol. 6 ›› Issue (4): 62-75.doi: 10.2478/jdis-2021-0033

• Research Papers • Previous Articles     Next Articles

Scientific Value Weights more than Being Open or Toll Access: An analysis of the OA advantage in Nature and Science

Howell Y. Wang1,*, Shelia X. Wei1,*, Cong Cao2, Xianwen Wang3,(), Fred Y. Ye1,()   

  1. 1International Joint Informatics Laboratory & Jiangsu Key Laboratory of Data Engineering and Knowledge Service, School of Information Management, Nanjing University, Nanjing 210023, China
    2Nottingham University Business School China, University of Nottingham Ningbo China, Ningbo, 315100, China
    3WISE Lab., Dalian University of Technology, Dalian 116024, China
  • Received:2021-01-30 Revised:2021-02-10 Accepted:2021-09-10 Online:2021-11-20 Published:2021-11-01
  • Contact: Xianwen Wang,Fred Y. Ye E-mail:xwang@dlut.edu.cn;yye@nju.edu.cn
  • About author:* The authors contribute equally and are listed alphabetically.

Abstract:

Purpose: We attempt to find out whether OA or TA really affects the dissemination of scientific discoveries.
Design/methodology/approach: We design the indicators, hot-degree, and R-index to indicate a topic OA or TA advantages. First, according to the OA classification of the Web of Science (WoS), we collect data from the WoS by downloading OA and TA articles, letters, and reviews published in Nature and Science during 2010-2019. These papers are divided into three broad disciplines, namely biomedicine, physics, and others. Then, taking a discipline in a journal and using the classical Latent Dirichlet Allocation (LDA) to cluster 100 topics of OA and TA papers respectively, we apply the Pearson correlation coefficient to match the topics of OA and TA, and calculate the hot-degree and R-index of every OA-TA topic pair. Finally, characteristics of the discipline can be presented. In qualitative comparison, we choose some high-quality papers which belong to Nature remarkable papers or Science breakthroughs, and analyze the relations between OA/TA and citation numbers.
Findings: The result shows that OA hot-degree in biomedicine is significantly greater than that of TA, but significantly less than that of TA in physics. Based on the R-index, it is found that OA advantages exist in biomedicine and TA advantages do in physics. Therefore, the dissemination of average scientific discoveries in all fields is not necessarily affected by OA or TA. However, OA promotes the spread of important scientific discoveries in high-quality papers.
Research limitations: We lost some citations by ignoring other open sources such as arXiv and bioArxiv. Another limitation came from that Nature employs some strong measures for access-promoting subscription-based articles, on which the boundary between OA and TA became fuzzy.
Practical implications: It is useful to select hot topics in a set of publications by the hot-degree index. The finding comprehensively reflects the differences of OA and TA in different disciplines, which is a useful reference when researchers choose the publishing way as OA or TA.
Originality/value: We propose a new method, including two indicators, to explore and measure OA or TA advantages.

Key words: Open access, Toll access, Scientific discovery, Academic dissemination, Data analysis