Big Data Analytics and Electronic Resource Usage in Academic Libraries: A Case Study of a Private University in Kenya

Abstract

The purpose of the study was to apply Big Data analytics as a tool for evaluating electronic resources usage in the academic library setup in Kenya with reference to the library of one private university. Log files of postgraduate students were mined from the server where the offsite access platform (ezproxy) has been installed. Descriptive statistical techniques such as mean, standard deviation and percentages were computed. Data was transferred to the Statistical Package for Social Sciences (SPSS) software was aided in the analysis. Results revealed that in terms of usage intensity, total URL count was 2,352, the highest user made 283 downloads and the mean URL count was 49 downloads. Further findings revealed that no user utilized more than 5 databases over a period of one year. The mean usage intensity score for respondents who were trained or orientated on e-resource usage was above average at 69.0 while those who had not received training were below average at 29.8. It was concluded that big data analytics is a necessary and powerful tool for investigating electronic resources seeking and usage trends and patterns within Kenyan university libraries. Through big data analysis and data mining, usage patterns and trends such as usage intensity that might not have accurately been revealed through other tools are unearthed. Big data analytics has revealed user preferences and intensity of utilization of various databases and helped in detection of redundant databases. From the usage patterns, it was clear that the level of utilization of the University library’s e-resource platform was very low. Most of the databases accessible through the platform were redundant. Further, only two databases namely e-book central and ebscohost were the most popular among users while the rest were barely being utilized if at all. For most students, just one or two databases were sufficient in meeting their research needs. An integrated data analytics model for investigating university library’s e-resources usage is proposed