Universities in Google: Towards a multilevel analysis model of academic web positioning

Abstract

We propose a model of analysis of the web positioning of universities based on a vocabulary of keywords categorized according to the different university mission; it is applied to a university (Universitat Politecnica de Valencia) to check its suitability. From a vocabulary of 164 keywords, 290 web queries were built and executed on Google, collecting the first 20 results obtained for each query. The results confirm that the universities obtain a variable web positioning according to the dimension linked to the web query and that the web pages related to teaching (especially Degrees) obtain the best ranks, even when executing web queries oriented to research. However, a low position is observed not only for the UPV, but also for all the Spanish on-site public universities (only a 27% of the Top 20 ranks correspond to these universities). It is concluded that a multilevel analysis is necessary to study the web positioning of the universities and that the proposed model is both viable and scalable. However certain limitations must be taken into account, as the dependence of the vocabulary used and the high variability of data, that must be considered in the design of this type of analysis models.