To meet the needs of researchers and other library users, librarians must analyze data from a multitude of sources, including usage reports. Usage reports illuminate the who, what, when and where of access to electronic resources, allowing librarians to make more informed decisions about how to best serve their user communities.
At the Université de Lille 3, I led the EPEF research project (see sidebar) on French students' and researchers' use of e-journals, e-books and search databases. The three-member team of librarians and researchers in information science and librarianship also studied how French university libraries use statistics. With an impressive 37% response rate to the questionnaire, we were able to put forward preliminary yet significant results around this issue.
Amazingly, of the libraries surveyed, only 25% of their publishers provide usage statistics. The majority of French and Francophone publishers - often in humanities and social sciences - do not provide usage statistics. Therefore, libraries dedicated to law, letters and social sciences cannot rely on statistical usage data for their purchase strategies. Science, technology and medical (STM) libraries are less affected as they subscribe to international scientific publishers that do provide this data. Therefore, STM librarians are able to further develop important skills regarding usage statistics and integrate them into their purchasing decisions.
Usage statistics are now crucial in the daily work of many librarians — 68% of them collect data every month. The main reasons for collecting are: 96% for reporting to the French higher education ministry, 90% for purchasing decisions and prioritizing spending, and 71% to justify costs to the university. Only 21% of professionals use the statistics for training, communication or a better understanding of the usage of their scholars.
Most respondents (93%) use Counting Online Usage of Networked Electronic Resources (COUNTER) reports to exploit the data in Excel or CVS formats. The JR1 is the most used report, confirming its status as a key indicator for the use of electronic journals in libraries. At the same time, some librarians believe the JR1 does not allow a nuanced analysis of usage. JR1 is therefore paired with the JR1a (backfiles) and the JR5 (JR1 per year). In 67% of cases, using the COUNTER reports allows professionals to compare data among publishers and to study cost per download (CPD). Thanks to this, professionals claim, they can better argue their purchasing choices.
“Amazingly, of the libraries surveyed, only 25% of their publishers provide usage statistics.”
Many of the respondents (39%) also received usage data that were not COUNTER-compliant (e.g., sessions, requests, views). These respondents commented on the data's lack of uniformity among publishers, and they questioned the statistics' reliability based on the instability of the platforms and packages.
To obtain detailed metrics on cost and preferred items and collections, more than half of the respondents (67%) combine usage data from publishers with other data, such as:
- The size of the target audience
- Local statistics (log files) for improved web analytics
- Ulrich's data, to identify usage by discipline
- Costs, to specify CPD, but also shared costs among libraries that subscribe together to the same resource
- Bibliometric data to balance the status of journal titles through usage
The major obstacle to exploiting these statistics was time. During the 2009–2010 academic year, librarians spent an average of only 72 hours on statistics. The category labeled as “statistics” for electronic resources was ranked as “very time-consuming,” adding to the pile of tasks performed by librarians.
The study outcomes clearly showed that statistics are becoming a daily reality for French academic libraries. COUNTER has become the standard for collecting and exploiting data, though many French publishers still need to supply COUNTER-compliant statistics. Librarians would benefit from having more information and training on the COUNTER report, including interpreting the statistics for practical purposes. No survey participants used Enterprise Resource Management (ERM) or Standardized Usage Statistics Harvesting Institute (SUSHI) protocols; therefore, librarians may want to explore a range of tools to manage their daily work on statistics.