Size of web domains and interlinking behavior of higher education institutions in Europe

• Published in: Scientometrics Vol. 100; no. 2; pp. 497 - 518
• Main Authors: Lepori, Benedetto, Aguillo, Isidro F., Seeber, Marco
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01-08-2014; Springer
• Subjects: Bibliometrics. Scientometrics; Bibliometrics. Scientometrics. Evaluation; Colleges and universities; Computer Science; Educational policy; Europe; Exact sciences and technology; Higher education; Information and communication sciences; Information science. Documentation; Information Storage and Retrieval; Library and information science. General aspects; Library Science; Sciences and techniques of general use; Social networks; Web pages; Webometrics; Social relationships; D85; 62P20; Higher education; Document network; Weblinks; 91B74; 62J12; Correlation analysis; Scientific research; Lognormal distribution
• ISSN: 0138-9130; 1588-2861
• DOI: 10.1007/s11192-014-1242-6

The aim of this paper is to empirically test whether interlinking patterns between higher education institutions (HEIs) conform to a document model, where links are motivated by webpage content, or a social relationship model, where they are markers of underlying social relationships between HEIs. To this aim, we analyzed a sample of approximately 400 European HEIs, using the number of pages on their web domains and the total number of links sent and received; in addition we test whether these two characteristics are associated with organizational size, reputation, and the volume of teaching and research activities. Our main findings are as follows: first, the number of webpages of HEI websites is strongly associated with their size, and to a lesser extent, with the volume of their educational activities, research orientation, and reputation; differences between European countries are rather limited, supporting the insight that the academic Web has reached a mature stage. Second, the distribution of connectivity (as measured by the total degree of HEI’s) follows a lognormal distribution typical of social networks between organizations, while counts of weblinks can be predicted with good precision from organizational characteristics. HEIs with larger websites tend to send and receive more links, but the effect is rather limited and does not fundamentally modify the resulting network structure. We conclude that aggregated counts of weblinks between pairs of HEIs are not significantly affected by the web policies of HEIs and thus can be considered as reasonably robust measures. Furthermore, interlinking should be considered as proxies of social relationships between HEIs rather than as reputational measures of the content published on their websites.