Date of this Version
Power and non-power use, involving user discrimination on the basis of expertise, is an idea from end-user computing with potential applicability as an interpretive tool for analyzing e-book user behavior. Can academic e-book power users be reliably identified in system-generated log data? A case study set of three-year e-book user transaction log data generated by the Ebook Library (EBL) platform was made available by the Edith Cowan University Library (Perth, Australia) to assist with the study. Deep Log Analysis (DLA) was used to explore the data. With statistical methods, further investigation yielded insight into whether an equation for identifying academic e-book power users within transaction log data could work at an appropriate confidence level. Identifying and isolating academic power e-book users in transaction logs for study presents some methodological challenges, for DLA targets large datasets requiring new skills and a commitment to learning new methods. This study has met this challenge by modelling academic e-book power users in transaction logs.