INTRODUCTION
Higher education in the UK is facing a financial crisis with reduced incomes and increased costs across the sector. The Office for Students identified 93 higher education providers reporting a financial deficit in 2022/23, which is expected to increase to 108 providers in 2023/24.Footnote 1 The national press reports budget cuts, course and department closures, and redundancies across the sector.Footnote 2 Even for those universities not named in the newspapers, the financial challenges in higher education are real. For librarians, university budget cuts mean library budget cuts and the scrutiny of every purchasing decision.
The story of this research project began in Summer 2023, with a proposal to massively increase the cost of the annual subscription to a major law database at the University of Warwick. It prompted a review of all law databases, which included a deep dive into their usage statistics and perceived value to the library and law school. The story almost ended with the cancellation of a major law database, but this was avoided after months of negotiation with our supplier.
The data revealed a surprising decline in law database use at the University of Warwick, which prompted me to share our story at the BIALL conference in June 2024.Footnote 3 This article explains methods of data collection and analysis, the limitations of the data, and discuses trends in law database use. It also brings the lightning talk up to date with complete data for the 2023/24 academic year.
RESEARCH METHODS
The research data comprised database usage statistics (for Lexis+ UK, Westlaw UK, HeinOnline and Law Trove) and student numbers (of full-time equivalent law students) for each academic year (from 1 August to 31 July). The research initially examined the previous five academic years, from 2018/19 to 2022/23, but is now brought up to date to 2023/24. The exception to this is Law Trove, because our subscription only started in 2020/21, which means that earlier data is not available. The research calculated the average database use per student and analysed trends in database use across the academic years.
STUDENT NUMBERS
Universities report official annual statistics on students, staff, graduates and finances to the Higher Education Statistic Agency (HESA),Footnote 4 but data on student numbers was more easily acquired internally from the university’s Strategy, Planning and Analytics Office.Footnote 5 The data is a snapshot of student enrolments on 31 October in any given academic year. For this research, the total number of full-time equivalent students enrolled in the Warwick Law School in academic years from 2018/19 to 2023/24 was recorded. This included all undergraduate, postgraduate taught and postgraduate research law students, but excluded some students on joint degree programmes (who undertake law modules), because their degree programmes were administered by other departments in the university.
DATABASE USAGE
Usage statistics for law databases were more difficult to acquire than student numbers. Lexis+ UK and Westlaw UK usage data was available on request from our account managers. Data was supplied in a PDF report, which also included a detailed breakdown of usage by month, content type and practice area. The unit of measurement for database use was Activities for Lexis+ UK and Usage for Westlaw UK. Both databases had undergone significant platform changes during the research period, which complicated the collection and comparison of usage statistics between the current and legacy versions.
HeinOnline usage data was available on request from its Technical Support team and was then supplied as a scheduled monthly report thereafter.Footnote 6 Data was supplied as text / table in a HTML email, and the units of measurement were Hits, Articles, Visits, Page Views and Searches. Articles was selected as the measure for this research because it was the most similar to those used by other law databases.
Law Trove usage data was available from the Journal Usage Statistical Portal (JUSP),Footnote 7 and also on request from the Oxford University Press account manager. JUSP provides access to COUNTER compliant usage statistics for electronic books, journals and databases from participating publishers.Footnote 8 A TR-B1 Book Request report was run for Oxford Academic. The data was exported as a CSV file, and the spreadsheet limited to the Law Trove platform only. The unit of measurement was an Item Request, and the report provided monthly and total usage data by book title and across the platform.
RESEARCH DATA
The research data was recorded in a simple table in Excel, with columns for total annual database usage and number of students, and rows for academic year (Figure 1). The database use per student was calculated using a simple average (mean). For each academic year, the annual database usage was divided by the number of students and rounded to one decimal place (Figure 2). Finally, the combined database usage was calculated by adding the total usage per student of Lexis+ UK, Westlaw UK and HeinOnline for each academic year (Figure 3). Law Trove was omitted from this calculation because data was not available for the entire reporting period.
Figure 1: Count of law database use and student numbers at University of Warwick
Figure 2: Average law database use per student at University of Warwick
Figure 3. Combined count of law database usage per student and student numbers
CHALLENGES OF THE DATA
Access to usage statistics for law database use was challenging. Only Law Trove provided self-service statistics via the JUSP portal. With JUSP, librarians can run reports and download statistics directly, without reliance on any third parties, so access to data is instant. Usage statistics for Lexis and Westlaw were requested via the account manager, and there was a delay (normally of days but sometimes weeks) in the provision of data.
The format of data was also varied: Law Trove and Westlaw supplied numerical data in CSV and XLS formats, but HeinOnline and Lexis reports supplied textual data in TXT and PDF formats respectively. Where numerical data was supplied in textual format, it needed to be copied, pasted and converted into a suitable numerical format. Where data was supplied in a numerical format, in CSV or TAB format files, it was more easily read and analysed in Excel software.Footnote 9
There was no standard unit of measurement, which meant that usage data was not comparable between databases. Lexis counted activities, Westlaw usages, HeinOnline articles and Law Trove item requests. At Warwick in 2023/24, Westlaw was the most used database (84,808 usages) and Lexis was the least used database (60,430 activities). But it is unclear what was being counted. Is an ‘activity’ more than, less than or the same as a ‘usage’? Is an ‘article’ more than, less than or the same as an ‘item’? Only Law Trove uses an industry standard unit of measurement. The item request (TR-B1) report is a COUNTER compliant metric, which is a global standard for measuring and reporting content usage through normalised metrics.Footnote 10
Additionally, usage data was not always standardised or comparable within a single database. Both Lexis and Westlaw had undergone product and platform changes during the research period, the most notable of which being the Lexis+ UK product launch in 2022, but the agile development of technology means ongoing changes to platforms. LexisNexis advised that usage data was not directly comparable between the legacy (Lexis Library) and new (Lexis+ UK) databases.Footnote 11
After the conference presentation, I had many interesting conversations with librarians and vendors about changes to law databases and their implications for usage data. Technology providers have improved the usability and functionality of their products in the past few years. Databases have become increasingly efficient and effective, and users are more likely to find content first time, which may result in fewer searches and lower usage. The size of law databases is not static: new content is added on a daily basis, and other content (e.g. new practice area or jurisdiction) can also be added or removed from database subscriptions. These combined changes to the products and their content mean that usage data may not be comparable over time.
TRENDS IN LAW DATABASE USE
When the data was first reviewed in Summer 2023, we were surprised to see a declining trend in the use of all law databases at the University of Warwick. Over the previous five years, the use of Westlaw decreased by 50% (from 145,853 to 73,443 usages), HeinOnline decreased by 31% (from 81,959 to 56,667 articles) and Lexis decreased by 18% (from 40,974 to 33,656 activities). Law Trove experienced a very small decrease of 2% (from 84,555 to 82,473 item requests) over a three-year period (Figures 1 and 4).
Figure 4: Chart of law database use at University of Warwick. (From the top-left of chart down the lines are Westlaw, Law Trove (partial), HeinOnline, Lexis)
In 2021/21, there was a recovery in the use of law databases: Lexis increased by 18%, HeinOnline plateaued, and Westlaw’s decrease slowed to only 6% on the previous year (Figures 1 and 4). With universities in lockdown due to the pandemic, libraries were forced to embrace the digital shift to electronic resources and remote working.Footnote 12 This was the year the University of Warwick started its subscription to Law Trove. But the recovery in law database use was slight, not a spike as might be expected given the conditions, and the declining trend in usage continued in 2021/22.
Changes in student numbers are the most likely reason for changes in the use of library resources. Logically, increased student numbers equal increased usage, and decreased student numbers equal decreased usage. However, law student numbers at the University of Warwick were relatively stable, averaging 1007 law students (Figure 1). The average use per student showed a similar declining trend. Over the five years until 2022/23, the average use of Westlaw decreased by 56% (from 155.2 to 69.0 usages per student), HeinOnline decreased by 39% (from 87.2 to 53.3 articles per student), Lexis decreased by 27% (from 43.6 to 31.6 activities per student), and Law Trove decreased by 5% (from 82.0 to 77.5 item requests per student) (Figures 2 and 5). We can conclude that student numbers did not drive changes in the use of law databases.
Figure 5: Chart of law database use per student at University of Warwick (From the top-left down lines are Westlaw, Law Trove (partial), HeinOnline, Lexis)
In Summer 2023, the clear declining trend in law database use triggered some very serious internal conversations at the University of Warwick about the value of law databases to the library and law school. This decrease in law database use contradicted the increased use of electronic library resources in general. Over a similar period, the University of Warwick reported a 41% increase in the use of ebooks and 21% increase in the use of ejournals.Footnote 13 Where libraries measure the value of electronic resources on a cost per use basis, and where prices are increasing and usage is decreasing, the data raised serious concerns about the value for money and long-term sustainability of our subscriptions to law databases.Footnote 14
The story did not end in Summer 2023. The most recent data shows a surprising reversal in the usage trends. In 2023/24, both total and average use of law databases increased for all law databases. The total use of Westlaw increased by 15% (up to 84,808 usages), HeinOnline increased by 33% (up to 75,250 articles) and Lexis increased by a massive 80% (up to 60,430 activities) (Figures 1 and 4). The average usage statistics increased by an even higher percentage due to a slight decline in student numbers. When examined by average use per student, Westlaw increased by 22%, HeinOnline increased by 40% and Lexis increased by 89% (Figures 2 and 5). Given the declining trend of law database use, what explanations are there for the recent increase in usage, and is it likely to be sustained as an increasing trend in law database use?
EXPLAINING THE TRENDS
When seeking to explain the trends in law database use, it is helpful to consider a timeline of key events, on local, national and global scales, which affected the University of Warwick (Figure 6).
Figure 6: Timeline of key events at the University of Warwick
COVID AND LIBRARY RESOURCES
The move from print to digital library resources began many years ago, but the pandemic accelerated the pace of change in library services and resources.Footnote 15 Early responses to lockdown focused on physical problems including the closure of buildings and access to print library resources. In time, the focus shifted to digital library resources, including digitisation of print resources, and purchase or licensing of electronic books, journals and databases.Footnote 16
Library expenditure on and use of digital resources increased massively in 2020/21. Across Russell Group universities in 2021/22, library expenditure on digital resources increased by 24% on ebooks and 28% on serials, and use of digital resources increased by 50% for books and 6% for journals.Footnote 17 Law Trove was one of the free resources offered by publishers in the early days of lockdown. It was very popular with students and quickly became embedded in teaching and learning in the Warwick Law School. The digital doors to Law Trove opened in 2019/20, and the University of Warwick started a subscription in 2020/21. In the first year, Law Trove was used 84,555 times, and this usage has remained constant over the following years (Figures 1 and 4). Investment in new ebook collections like Law Trove account for much of the post-pandemic increase in the use of digital library resources.
The other law databases (HeinOnline, Lexis and Westlaw) were established well before Covid. They were existing, not new, digital library resources. As such, the patterns in the use of these law databases did not change in the same dramatic way. These law databases had previously followed a declining trend in usage, and the effect of Covid was more to pause the decline, than reverse it entirely (Figure 3). This reflects the SCONUL library statistics, where the use of electronic journals increased by only 6% in 2021/22. Other studies have found that use of digital library resources fell during the Covid period, although some of this decline is attributed to the use of free resources in lockdown not being counted (as library users could access them without authentication, and therefore without usage being recorded).Footnote 18
ROLE OF LAW LIBRARIAN
Covid had a major impact on the wider economy, triggering changes to work and the workforce. Over 300,000 people became economically inactive in England after the pandemic, including 100,000 people who took early retirement.Footnote 19 The rush to retirement occurred in the higher education sector too, including at the University of Warwick, and the library faced a mini-staffing crisis in 2021/22 due to the retirement of multiple library staff. The previous law librarian retired in August 2021, and I was appointed in April 2022. The post was vacant for nine months, and the University of Warwick was without a law librarian (and therefore promotion of and training in law databases) for most of the 2021/22 academic year. This coincided with the lowest point in the use of law databases in the reporting period.
In September 2022, the library introduced new subject guides to library resources. Their purpose was to make key subject specialist books, journals and databases more discoverable, as well as direct students to support from their subject librarian. The new law library guide was well received by the law school’s student-staff liaison committee, and has had a consistently high number of page views. The appointment of a law librarian and introduction of library subject guides coincided with an increase in the use of law databases in 2023/24. This suggests that librarians can play an important role in promoting the use of library resources through advocacy, inductions, teaching and training, and marketing.
NEW LAW CURRICULUM
September 2022 also heralded the start of a great change to legal education and training in the UK. The Solicitors Regulation Authority (SRA) introduced a new qualification and training route for solicitors in England and Wales. Prospective solicitors are now required to hold a degree level qualification in any subject, pass the Solicitors Qualifying Examinations (SQE), and undertake two years of legal work experience.Footnote 20 Perhaps controversially, solicitors are no longer required to have a law degree, and law degrees are no longer required to meet Qualifying Law Degree (QLD) standards, although are still required to meet the Subject Benchmark Statement standard.Footnote 21
Without the QLD, universities now have more flexibility in the design of their law degree programmes, and many have been redesigned to diversify the academic curriculum or accommodate preparation for SQE. At the University of Warwick, foundation subject modules were offered at lower credit values, which opened space for additional optional academic modules and a compulsory dissertation in the final year.Footnote 22 The first cohort of students on the new law degree programme started their studies in 2022/23 and will complete their final year in 2024/25.
The new law curriculum means new modules and new recommended reading. Reading lists are increasingly important, and they drive student engagement with and use of library resources. Studies show that only 20% of students complete assigned reading,Footnote 23 and that students read strategically, focusing on essential reading only.Footnote 24. Where there are changes to recommended books or journals, or legislation or case law, there will inevitably be changes to the use of library databases. The change to the law curriculum may be one reason for the increased use of law databases, particularly in the second year (2023/24), as law students began to take more of the new optional modules.
FUTURE TRENDS
I am currently supporting the first 300 students of the new law degree in their final-year compulsory dissertation. The dissertation allows students to follow personal research interests and encourages the development of independent legal research skills. Alongside new final year optional modules, the dissertation could have a very significant impact on the use of law databases. I am hopeful the recent increasing trend in law database use will continue in 2024/25.
The availability of open access publications is increasing. In the legal sector, parliamentary, governmental and legal publications are openly available with Open Government and Open Justice licenses. The National Archives now publishes legislation and case law in open access databases. In the academic sector, journal articles must be published openly to be eligible for the Research Excellence Framework (REF). The number of journal articles published via gold open access (unlocked on the publisher’s website) in the UK has increased from 23% in 2018 to 39% in 2023.Footnote 25 These open access publications are easily discoverable and accessible via Google and other internet search engines, and they have the potential to reduce recorded use of library databases in the future.
There is much research on Google v The Library, but Perruso’s longitudinal study offers hope to the librarian and law database provider. While undergraduate students preferred Google in the first year of their study (70% favouring Google), their information seeking behaviour developed over time (due to librarian instruction and faculty requirements), and they preferred library databases by their final year (70% favouring library resources).Footnote 26
COUNTING ON COUNTER
Law Trove was the only law database to offer COUNTER usage statistics. COUNTER offers a global standard for measuring and reporting the use of electronic library resources through normalised metrics. COUNTER includes a range of reports to measure the use of electronic books, journals and databases, including searches, requests and denials.Footnote 27 COUNTER statistics are not infallible, and as well as not counting usage of open access resources, there is evidence that databases can be designed to inflate usage statistics by double-counting HTML access and PDF downloads.Footnote 28
That said, COUNTER metrics are common in academic libraries, and they are adopted as the sector standard for reporting the use of electronic library resources, for example in SCONUL’s annual library statistic return. Currently, 130 suppliers offer COUNTER metrics via the JUSP statistical portal. Notably, this excludes key law publishers and databases: LexisNexis (Lexis+ UK), Thomson Reuters (Westlaw UK and Practical Law), HeinOnline and Wolters Kluwer are all on the JUSP supplier wish list.Footnote 29
In my talk at the BIALL conference, I called on law database suppliers to provide COUNTER compliant usage statistics, and that remains my strong recommendation. While the journal (TR-J1) and book (TR-B1) request metrics may not be appropriate for complex law databases (which include multiple types of legal information), the platform usage (PR-P1) and database search (DR-D1) metrics may be more suitable measures of law database use. The adoption of COUNTER compliant statistics would go a long way to making law database usage data more accessible, accurate and comparable.
CONCLUSION
This research project has been a timely exercise in measuring and providing a narrative around the use and value of law databases at the University of Warwick. The pressure on university and library budgets has never been greater, and our resources are under increasing scrutiny. While not perfect, due to the challenges of data collection and reliability, it provides a simple framework for others to assess use of their databases, in law and other disciplines. It would be interesting to see how trends in law database usage compare across the academic sector, to understand if Warwick’s experience is unique or reflective of general trends across the sector.
Open access
