Taxonomy of inlinked Web entities: What does it imply for webometric research?

• Published in: Library & information science research Vol. 27; no. 1; pp. 8 - 27
• Main Author: Chu, Heting
• Format: Journal Article
• Language: English
• Published: San Diego, CA Elsevier Inc 01.01.2005; Elsevier
• Subjects: Bibliometrics. Scientometrics. Evaluation; Citation analysis; Exact sciences and technology; Information and communication sciences; Information science. Documentation; Library and information science. General aspects; Links; Sciences and techniques of general use; Webometrics; World Wide Web; United States; North America; America; Information science; Higher education; Citation analysis; Informetrics; World wide web; Webometrics; Librarianship; Typology; Web site; Hyperlink
• ISSN: 0740-8188; 1873-1848
• DOI: 10.1016/j.lisr.2004.09.002

Hyperlinks from other Web sites are, in some respect, similar to bibliographical citations. Link analysis, like citation analysis in bibliometrics, has emerged as a research area of webometrics in recent years. But why are links made and where do they point? A sample of inlinked Web entities (i.e., Web pages or Web sites) was randomly selected from a group of academic institutions' Web sites. The inlinked sites, along with the hyperlink data and outlinking sites, were analyzed and categorized to form a taxonomy of inlinked sites. Based on this taxonomy, a list of reasons for hyperlinking, grouped in four top-level categories (teaching/learning, research, service, and home page), was identified. Compared with bibliographical citations, hyperlinks were made for a different set of reasons. Hyperlinking also has fewer dimensions, less complexity, and little negative implication. On the whole, almost 50% of all the inlinks examined were created for pointing to resource or directory information provided at the target Web sites. In addition, nearly three fourths (73%) of all the inlinked sites analyzed in this study were linked to for reasons relating to service or home page while less than one third (27%) of the links were made out of research or teaching/learning motivations. Yet, teaching and research are the two major criteria traditionally used for evaluating academic institutions. These findings, although by no means conclusive, suggest that evaluative link-based studies should not only consider link counts but also reasons for hyperlinking in order to ensure the validity of such research.