OPEN CITATIONS TO OPEN SCIENCE
发布人:高教评价院  发布时间:2019-07-16   浏览次数:424



Cassidy R. Sugimoto1, Dakota S. Murray1, and Vincent Larivière2

1. Indiana University, USA; 2. Université de Montréal, Canada



Three years ago, an advocacy group called the Initiative for Open Citations (I4OC) began the ambitious agenda of making citations open to the world. They worked arduously with a subset of receptive publishers—subscription-based and open-access, commercial and non for profit—to change the status of references contained in Crossref from closed to open. By the official launch of the initiative in April 2017, they were able to report impressive progress: whereas freely available references were available for only 1% of indexed publications before I4OC, that number stood at 40%. The initiative continued to gather steam, creating partnerships with major publishers, funders, and professional societies, and the percentage of open references continued to grow, representing more than half of all cited references in Crossref by January 2018.

 

The International Society for Scientometrics and Informetrics (ISSI) showed its support by publishing an open letter and urging members of field to demonstrate their support by signing it. Over the course of the four months, the open letter was signed by 324 individuals from 46 countries as shown in Figure 1. The most highly represented countries include the USA, Netherlands, Sweden, Spain, the UK, and China. The geographic distribution of our signatories is, in many ways, what one would expect given the demographics of our community. However, this distribution also signals two important pieces of evidence for the open citation movement.


 

Figure 1 Distribution of the signatories supporting the I4OC

 

Many of the signatories are affiliated with institutions with access to high quality citation indexes through a library subscription. This could be read as a technical limitation: that even those with access do not find that access sufficient to conduct large-scale bibliometric ****ysis. However, there are also several signatories who have access to large-scale databases for research purposes. This suggests an ideological position: adopting the proposition that citations should be open to all, regardless of resources.

 

The list of signatories also tells a story about access. The map of signatories could easily be overlaid with several other data that demonstrate the concentration of scientific resources and infrastructure across the world. The countries that are missing are those that have been historically underrepresented in elite indexes. It is hoped that opening citations may also serve as a mechani** to open the scientific world. Openness is foundational for rigorous science. We hope that by opening citations, we not only increase access to scientific knowledge, but also increase participation. As we noted in our open letter: “This is a matter of scientific integrity, scientific progress, and equity—we must ensure that all members of the scientometric community are able to participate in and validate the research in the field”. We therefore urge scientists to join our open letter and continue to advocate for open citation data.

Author Biography

Cassidy R. Sugimoto

Associate professor at the School of Informatics and Computing at Indiana University Bloomington. Cassidy researches within the domain of scholarly communication and scientometrics, examining the formal and informal ways in which knowledge producers consume and disseminate scholarship. She has edited and co-edited four books and has published numerous journal articles on this topic. Her work has been presented at numerous conferences and has received research funding from the US National Science Foundation, Institute for Museum and Library Services, and the Sloan Foundation, among other agencies. Cassidy is actively involved in teaching and service and has been rewarded in these areas with an Indiana University Trustees Teaching award (2014) and a national service award from the Association for Information Science and Technology (2009). She served as the President of the faculty at Indiana University in 2015-2016 and is currently serving as President of the International Society for Scientometrics and Informetrics. Cassidy has an undergraduate degree in music performance, an M.S. in library science, and a Ph.D. in information and library science from the University of North Carolina at Chapel Hill.

 

Dakota S. Murray

PhD Candidate at the School of Informatics and Computing at Indiana University Bloomington.

 

Vincent Larivière

Vincent Larivière holds the Canada Research Chair on the Transformations of Scholarly Communication at Université de Montréal, where he is associate professor of information science. He is also scientific director of the Érudit journal platform, associate scientific director of the Observatoire des Sciences et des Technologies (OST) and regular member of the Centre Interuniversitaire de Recherche sur la Science et la Technologie (CIRST). Over the last 15 years, Vincent’s research has investigated the characteristics of global research systems, determinants of researchers’ scientific activities, as well as the transformations, in the digital world, of the modes of production and dissemination of scientific and technological knowledge. He holds a bachelor’s degree in science, technology and society (UQAM), a master’s degree in history of science (UQAM) and a Ph.D. in information science (McGill), and has been a postdoctoral fellow at Indiana University Bloomington.

 

Appendix

Open citations: A letter from the scientometric community to scholarly publishers

December 5th, 2017

Openness is central to the research endeavor. It is essential to promote reproducibility and appraisal of research, reduce misconduct, and ensure equitable access to and participation in science. Yet, calls for increased openness in science are often met with initial resistance. The introduction of pre-print servers, open access repositories, and open data sets were, for example, initially resisted, but eventually adopted without adverse effects to the scholarly ecosystem. The launch of the Initiative for Open Citations (I4OC) is facing similar obstacles. This initiative has campaigned for scholarly publishers to make openly available the references found in articles from their journals. Many publishers, including most of the large ones, support the initiative and have opened their references. However, the initiative still lacks support from a minority of the large publishers.

 

Calls for enhanced reproducibility have been heard across all fields of science. However, scientometrics is often unable to meet these standards, largely because of the dependency of bibliometric research upon proprietary data sources. The ability to undertake large-scale and generalizable bibliometric research, both basic and applied, is limited to a few well-funded centers that can afford to pay for full access to the raw data of Web of Science or Scopus. The remaining bulk of bibliometric research is restricted to the ****ysis of **all data sets or the use of freely available data sources such as PubMed, Google Scholar, and Microsoft Academic. Although these freely available data sources are valuable, they suffer from shortcomings, such as incomplete coverage, data quality problems, lack of transparency, or limited large-scale accessibility. In order to conduct rigorous ****yses, scientometricians need a data source that is freely available and comprehensive. This is a matter of scientific integrity, scientific progress, and equity—we must ensure that all members of the scientometric community are able to participate in and validate the research in the field. I4OC is striving to create such an opportunity.

 

I4OC requests that all scholarly publishers make references openly available by providing access to the reference lists they submit to Crossref. At present, most of the large publishers—including the American Physical Society, Cambridge University Press, PLOS, SAGE, Springer Nature, and Wiley—have opened their reference lists. As a result, half of the references deposited in Crossref are now freely available. We urge all publishers who have not yet opened their reference lists to do so now. This includes the American Chemical Society, Elsevier, IEEE, and Wolters Kluwer Health. By far the largest number of closed references can be found in journals published by Elsevier: of the approximately half a billion closed references stored in Crossref, 65% are from Elsevier journals. Opening these references would place the proportion of open references at nearly 83%.

 

Open availability of citation data is important not only for the scientometric community, but also for science at large. Scientometrics is widely used to support science policy and research evaluation, with consequences for the entire scientific community. There is a need for specialized organizations, both commercial and non-commercial, that offer scientometric services. However, in order to guarantee full transparency and reproducibility of scientometric ****yses, these ****yses need to be based on open data sources. Analyses based on proprietary data sources have limited transparency and tend to be difficult to reproduce. Yet, as long as half of all references are missing in open data sources such as Crossref, ****yses based on these data sources do not offer a viable alternative. With such a large proportion of missing references, these data sources provide an incomplete portrait of the scholarly landscape, which can lead to negative effects, such as policies that ignore certain areas of research or certain countries.

 

Scientometricians have a professional obligation to promote sound practices in our field. In the current environment, advocating for open references is critical to ensure replicable and equitable research practices. We should use our relationships with journals—as authors, reviewers, and editorial board members—to advocate for openness and should expect scientometric journals to be leaders in this respect. References are a product of scholarly work and represent the backbone of science—demonstrating the origin and advancement of knowledge—and provide essential information for studying science and making decisions about the future of research. References are generated by the academic community and should be freely available to this community. We therefore issue a strong call to all publishers to make available to the academic community that which it created in the first place. To those publishers that have not already responded to the Initiative for Open Citations, our plea is: open citations now!