The E-book Power User in Academic and Research Libraries: Deep Log Analysis and User Customisation

Document Type

Journal Article


Taylor & Francis (Routledge)


Faculty of Health, Engineering and Science


School of Computer and Security Science




This article was originally published as: Ahmad P., Brogan M., Johnstone M.N. (2014). The E-book Power User in Academic and Research Libraries: Deep Log Analysis and User Customisation. Australian Academic and Research Libraries, 45(1), 35-47. Original article available here


In the literature on e-book adoption in academic and research libraries, user acceptance is seldom rigorously examined. As take-up of e-books has grown to achieve mass adoption, what industry analysts Gartner describe as the "plateau of productivity", the question of user acceptance may seem less relevant. However, if attention is switched from downloads to information behaviour, expectations and gratification, the picture of acceptance is more nuanced, with some studies pointing towards user disaffection.This paper reports on the information behaviour of another category of e-book users, i.e. apparently satisfied or intensive users. The paper is based on data analysis and interpretation of data from transaction logs generated by the Ebook Library (EBL) platform in an academic library. The paper forms part of a broader investigation of technology acceptance and options for improving the user experience of e-books within an academic library context. Three years of e-books transaction logs were mined for evidence of "power user" behaviour. The paper demonstrates how power user behaviour is different from other user behaviour, shows which variables determine such behaviour, and creates a probabilistic model that can determine a power user based on these variables. The paper also describes how this model was validated against the log data.