2013 Annual Meeting
Montréal, Québec, Canada | November 1-5, 2013
Daniel Coughlin, Penn State University
Mark Campbell, Penn State University
Bernard J. Jansen, Penn State University
This research uses the methodology of web analytics to examine the usage of subscription databases at a major academic library. Our research goal is the development of key performance indicators from which academic libraries can evaluate the business value of their content collections. There are 1,447 databases to which this academic library provides access, and these databases received nearly 2.5 million customer visits in 2012 via the library’s meta-search application, which is used for searching these databases. As such, these visits represent a substantial subset of the total traffic to the university’s academic databases. The first level analysis shows that the top 20 most used databases represent over half the traffic to these academic databases. The second level analysis compared these heavily used databases (20) categorizing them by provider, and quantifying them with the remaining databases (428) from these providers. These results show the inequality of traffic generated by the top databases relative to the remaining databases from these providers in the context of search. The implications of this inequality illustrate the extreme usefulness of select databases and the possibility of the dispensability of less popular databases. The third level of analysis is a temporal evaluation of demand of databases over the course of two semesters (spring and fall 2012). This evaluation displayed the lack of increased demand throughout a semester beyond the top 300 databases. We used this analysis as the beginnings of the formulation of a set of web analytic metrics tailored for academic libraries.