Scholarly journal usage: The results of deep log analysis

Purpose - To present the latest results of research conducted at University College London as part of the Virtual Scholar Research Programme, investigating the impact of the digital roll-out of information services to academics and researchers. This is the second study to look at the information seeking behaviour of academics and researchers in regard to digital journal libraries, and concentrates on the users and usage of Blackwell Synergy. Design/methodology/approach - Nearly a million users making ten million item requests were investigated employing deep log methods, developed by the authors to provide robust and big picture analyses of digital information consumers and their behaviour. Findings - Usage data has been embellished with user data (for 500,000 people), so enabling comparisons to be made between the information seeking behaviour, for instance, of students and staff, academics and practitioners, scientists and social scientists. We believe this is the first time this type of analysis has been attempted with logs. Of particular note is the "repeat visitor" evaluation and the analysis of one and a quarter million search sessions which categorised sessions in terms of how "busy" they were for a whole range of user groups. Research limitations/implications - Demonstrates a powerful and new method, deep log analysis, for mapping and evaluating information seeking behaviour. Practical implications - Important data for publishers to enable them to target their services more effectively Originality/value - Probably the first analysis of its type, hence showing an aspect of information seeking not previously seen.