The Web of Science traces its roots back more than a half-century, to when Dr. Eugene Garfield, information scientist and founder of the company that ultimately became Clarivate Analytics, oversaw the release of the Science Citation Index (SCI) in 1964. The first-ever such index, the SCI duly expanded from its original print edition to its current online incarnation as the Web of Science, now indexing the contents of some 30,000 journals, along with books, conference proceedings, and other materials.
In that time, the Web of Science has fulfilled a range of functions. In its primary role, it serves as an entry point for knowledge seekers, affording access to the premier scientific and scholarly literature, allowing researchers to identify and track all the published work most pertinent to their endeavors.
The Web of Science also offers a window into the characteristics, dynamics, and progression of research itself. Its interconnected threads of citations and clusters of specialty areas constitute a virtual mapping system through a networked landscape of knowledge.
The latter aspect has made the Web of Science indispensable to scholars whose labors fall under the heading of scientometrics, bibliometrics, or informetrics – researchers who study the dimensions and currents of science.
Countless studies have harnessed Web of Science data in examining, for example, the progression of research on a given topic, or the performance of individuals, institutions, and nations in productivity and citation impact.
Of course, when it serves as an authoritative source of data for a given paper, the Web of Science earns a formal mention in the text. And yet, until now no one has undertaken a detailed study of these actual mentions – their context, along with precise breakdowns in the journals, subject fields, and nations that have most frequently invoked the Web of Science and its related data tools.
In a recent article in Scientometrics, a trio of information scientists and data analysts representing Drexel University in Philadelphia, and Clarivate Analytics, explores this matter.
From publications indexed in the Web of Science between 1997 and 2017, the team searched English-language articles for the term “web of science,” as well as the related terms “impact factor,” “science citation index,” and “journal citation report.” The search produced nearly 20,000 papers, with a sharp year-by-year increase in occurrences. In 2017, nearly 4,000 papers mentioned the Web of Science – more than 120 times the 1997 tally.
Examining the entire cohort of papers, the analysts parsed the various entities whose authors are most prevalent in studies that mentioned Web of Science. (Spoiler alert: Among nations measured in terms of all authors listed in each paper, the United States is not the most numerous in representation in these studies.) Similarly, the team examined the knowledge domains (as reflected in the subject categories in both the Journal Citation Reports and Essential Science Indicators) accounting for the highest numbers of Web of Science-based studies, in terms of explicit mentions in the text. Individual journals were also enumerated on that basis.
Last, the study’s authors scrutinized the grammatical context of mentions, using natural language processing. This analysis identified the nouns and verbs most closely associated with mentions (e.g., “search”), and demonstrated, for instance, that the Web of Science is frequently mentioned in conjunction with other databases.
In all, the study is a testament to the myriad uses and the ever-expanding, global significance of the Web of Science over the last 20 years – proof that, five decades after its invention, Eugene Garfield’s citation index is still flourishing.
To read the complete report in Scientometrics, please click here.