I have previously blogged about my work in helping DMU Library to create a framework to move towards a more data-driven process for online resources renewals. The context for this work is to ensure that DMU Library is making well informed, evidence-based decisions relating to its electronic resources subscriptions. It is vital for DMU Library to continually maximise the value of its annual investment in online resources and provide access to relevant electronic content to support learning, teaching and research across the university.
Subscribed v Non- subscribed titles
One electronic resource that I concentrated on in the summer of 2016 was Elsevier’s ScienceDirect full text journal collection. Like many UK HE institutions, DMU Library subscribes to ScienceDirect via the national Jisc/NESLi agreement. This is a long-standing subscription cost that constitutes a major spend from the library’s annual online resources budget. The construction of the ScienceDirect e-journal agreement means that DMU Library connects to content in two ways:
- Subscribed (or core) titles. These are journals in the collection that have been selected by library subject teams to best support learning, teaching and research at the institution.
- Non-Subscribed (or non-core) titles via ScienceDirect Freedom Collection content. These are non-core publications that the library purchases as top-up content to the Subscribed titles in the collection.
As the Jisc/NESLi agreement with ScienceDirect was to be re-negotiated for 2017, July 2016 was a suitable time for the library to re-evaluate the value and relevancy of its ScienceDirect Subscribed titles. The existing Elsevier agreement allowed UK universities (should they wish) to substitute Subscribed journals for Non-Subscribed journals in the collection, as long as the institution’s contractual spend in the deal was maintained. If the ScienceDirect journal usage data for DMU highlighted that individual titles did not continue to support academic faculty interests, then the library would look to act to increase the value of its Subscribed titles before the new 2017 journal agreement was in place. The review also made sense when looking at maintaining post-cancellation access rights for DMU Subscribed titles should the library decide not renew the Elsevier agreement.
After several days of planning, the objectives for the ScienceDirect review were agreed. These were:
- Identify Subscribed ScienceDirect journal titles with low usage with a view to replacing them with high use Non-Subscribed titles. Any title substitutions would have to be made in line with the institution maintaining its annual contractual spend with the provider.
- Base the review on COUNTER JR1 usage data exported from the JUSP service and 2016 journal list prices from Elsevier. This usage and cost data would then be combined to provide a cost-per-use (CPU) metric for each reviewed Subscribed and Non-Subscribed title. As libraries can mark “core” titles in the JUSP service (via access to KB+), DMU Subscribed titles could be easily filtered for display in the exported COUNTER file.
- Usage and cost analysis to be collated by the library’s Content Delivery Team and disseminated to subject librarians in Academic Liaison for evaluation.
- Outcomes from the evaluation process to be captured by Content Delivery and candidates for substitution communicated back to the provider.
Ready, steady, go…
I have mapped out the ScienceDirect review workflow in more detail below:
- Exported Elsevier JR1 COUNTER usage data from JUSP to Microsoft Excel. The statistics covered the period between Jan 2012 to Jun 2016. As DMU Library had already identified existing Subscribed (or core) titles in the JUSP service via KB+, I was able to filter the usage data to display Subscribed titles only and then sorted the data from low to high usage.
- To ensure the review was more effective and up-to-date, the Elsevier usage data was refined further to cover usage between Jan 2015 – Jun 2016. The data was resorted: Subscribed titles from low to high usage, Unsubscribed titles from high to low usage.
- Received breakdown of 2016 journal costs from Elsevier to aid cost analysis in the review – costs for Subscribed titles in the agreement, and list prices for top 25 most used Unsubscribed titles.
- Two Excel files to work with – usage data exported from JUSP and journal costs from Elsevier. Used VLOOKUP function in Excel to merge the two datasets into one spreadsheet. Ran VLOOKUP match on journal title ISSNs.
- Created CPU metric for current Subscribed titles and top 25 most used Unsubscribed titles. Created proposed lists of Subscribed title with high CPU (to be removed) and Unsubscribed titles with low CPU (for addition). Analysis sent to subject librarians in Academic Liaison for evaluation.
- Based on the CPU metric, subject teams selected proposed Subscribed titles for substitution. Proposed titles sent to Elsevier. Elsevier ratified substitutions and made relevant changes to DMU Library’s Subscribed journals list for 2017.
CPU: what constitutes good value?
CPU provides a basic appraisal of library user activity in line with the financial investment made by the library. CPU may well represent a starting point for library staff discussion with regards the value of an online resource or collection. CPU does require context though, and one way to potentially achieve this is to set up some usage indicators. I blogged about creating usage indicators to help shape resource renewal or cancellation in my Aug 2016 post asking “How do academic libraries value their electronic resources?“.
Whilst not strictly applied by subject teams at DMU in this example of a journals review, the general usage indicators below could be applied by libraries when approaching this type of journal substitutions work. Obviously, other factors may need to be considered, but it does provide a framework for a move towards a more data-driven decision making process:
- CPU < £1: excellent value, automatically retain title.
- CPU between £1-£5: good to fair value, recommend retaining title.
- CPU between £5-£10: fair to poor value, investigate reasons for low use, potential substitution.
- CPU > £10: automatic substitution.
Outcomes of the review
After the review was completed, and the revised Subscribed titles list was approved by Elsevier and Jisc, DMU Library had removed approximately 20 low use, high CPU Subscribed titles and replaced them with a similar number of high use, low CPU Non-Subscribed titles for the new 2017 agreement. When analysing DMU’s revised Subscribed titles list, Jisc reported (based on 2015 full year COUNTER usage) that the substituted titles would increase the proportion of DMU core title usage in the Elsevier deal from around 18% to 29%.
I found the ScienceDirect review process to be a worthwhile activity. The process was not without its challenges, but it did provide an opportunity for library staff collaboration and discussion. It also gave the directorate a chance to engage with useful library user activity data and make online resource selection decisions that will hopefully provide more relevant content and support for students and staff at the institution.