Home › Posts tagged data analysis

data analysis

Trainee Showcase 2026: Part 1

Posted on by — No Comments ↓

If you’ve read previous blog posts, you might have seen us mentioning our trainee projects at various points. As an Oxford trainee you have the chance to take on a personal project outside your regular duties to help you explore different parts of library work and leave your mark on the Bodleian. This all leads to the Graduate Trainee Showcase at the end of the year, where you get to present your work and explain what you’ve been doing to a captive audience of librarians and colleagues.

The showcase this year was held at Christ Church College and organised by fellow trainees Heidi, Rebecca, Jules, and Amy – huge thanks to them for their planning and effort (and for some last minute tech fixes). Over the next month we’ll be posting short summaries of our projects and presentations for anyone who couldn’t make it. Writing about our own work is too easy though, so we’ve each picked someone else’s project to talk about. Now to get us started, here’s some Royal Regalia, medical paraphernalia, and a whole lot of library data. Enjoy!

Promoting Special Collections at New College Library & Archives

Leah, New College Library

Written by Catherine Birch

Leah’s project involved designing an exhibition to promote New College Library’s incredible special collections. Having had the chance to see her exhibition in-person, I thought it looked incredible, but I’d had no idea just how much work had gone into it over the year! 

Leah started by giving us some background on New College’s special collections, which house an astounding 384 historic manuscripts and over 30,000 rare books. Having specialised in medieval royalty and diplomacy during her MA, she wanted to highlight the wealth of royal history in the collection and work to improve public engagement with them. Identifying materials for display involved endless SOLO searching, trawling through various handlists, and consulting various colleagues, as well as every librarian’s secret weapon… a huge spreadsheet. Deciding which items to use in an exhibition is a very involved process, as Leah explained – you have to consider display space, transport logistics, and item condition before you even get into issues like audience interest and the exhibition theme. Having narrowed down which items she could practically include and which were most relevant to the topic, she picked her final items based on one key rule: are they fabulous?  

From there she wrote item captions and labels, proofread them, proofread them again (those pesky oxford commas…), and got to work producing promotional material. Some of this material included posters, banners, booklets, and postcards, which looked great and were very nice to keep as souvenirs. Leah also promoted the exhibition through social media, university mailing lists, local news pages, and word of mouth, trying to reach library staff, students, and the public in equal measure. Her hard work definitely paid off, as they had over 150 visitors, making it the most successful exhibition of the year! Leah herself was on hand to answer any questions and talk to visitors, which was a nice touch that really improved engagement and interest (as did the free refreshments next door).  

Even though the exhibition has finished, Leah is far from done getting people interested in New College Library’s special collections. She’s already planned an article about representations of King David in medieval manuscripts for an upcoming issue of New College Notes (the college’s scholarly journal), and she’s supporting a Curator’s Choice video highlighting a book published in 1687 to mark the coronation of James II. It was really interesting getting to see the full process of putting together an exhibition, and her display shone a spotlight on the sort of books and manuscripts that most librarians relish working with. 

A poster for the exhibition 'Royal Regalia: Signatures, Statecraft, and Sovereignty' at New College. The background has been made to look like a manuscript covered in cursive middle english. The foreground contains the date and location of the exhibition along with some images of items from the collection.

HML Grey Literature: Part II

Catherine, Bodleian Library (History Faculty Library)

Written by Olly Marshall

Catherine’s project revolved around a collection of grey literature from the History of Medicine Library, picking up the baton and expanding on the work done by previous HFL trainee Xanthe.  

Catherine first explained to us the complicated past of the HML, occupying a site on Banbury Road but being under the stewardship of the Radcliffe Camera, with most of the material having now moved to the Humanities Library at the Schwarzman Centre and some of it being stored in our offsite storage facility. This is fairly standard for a lot of collections in the Bodleian- everything has been moved between many libraries and has had many different names.  

The “grey literature” in question is far more interesting than the name suggest- it’s a mix of material produced by various medical institutions, but not controlled by a traditional commercial publisher like a book or journal article is. The HML collection includes a lot of organisational reports, speeches, statistics, training guides, public health pamphlets and other miscellaneous material important to understanding the history of healthcare and the work of significant institutions and professionals in the field.  

The bulk of Catherine’s project involved going through 41 boxes of the material and creating handlists for them- a list describing each item in a box – to make the contents identifiable and discoverable to researchers.  

The grey literature Catherine has been working with is now housed offsite and involved many requests for it all to be called up, looking through the contents and picking out important information such as institutions, authors, date ranges, topics, formats and other useful information for locating the right resource. This was frequently made difficult by material being in foreign languages, being in odd formats, using antiquated language or being catalogued poorly. By doing thorough research into the items, using templates for how to describe the items, and like any seasoned librarian – expecting and accepting a certain amount of chaos – Catherine developed a good system for listing the boxes consistently. All together there were approximately 450 items to describe and list, which is an amazing amount to achieve over the year!  

The contents of the boxes will ultimately be published in a LibGuide that Catherine has been making to promote their use, make the collection navigable and link to boxes on SOLO for people to call them up from CSF. I was really impressed by the skills Catherine got to develop during her project and her methodical approach, and I thought that the collection sounds fascinating. Work like this is really important for ensuring library collections are accessible and available to use, and the efforts Catherine went through will benefit readers and library staff alike. 

Managing & Adapting to Change

Rebecca, Bodleian Library (Reader Services)

Written by Jules McGee-Russell

Rebecca’s presentation focused on her research into different changes happening in the Old Bod and wider changes across the library service. This included the new Extended Lending rules at the Bodleian, the digital shift towards eBooks and AI usage, and how these things connect to the Climate Emergency. 

Rebecca demonstrated her points using journalistic headlines over time, data from Freedom of Information Requests, and statistical forecasting she did herself using Python. The messages I took away were that there is a year-on-year decline in borrowing requests from readers in the Bodleian. Even though the lending rules have changed to allow some items to be loanable from the Old Bod, the number of annual requests to borrow physical books is decreasing. In contrast, the demand for eBooks is huge and ever-growing, as are the profits publishers are making from eBooks sales. 1/3 of students surveyed are, concerningly, reaching for AI over primary sources.  

A map generated by ChatGPT, colour coded according to which countries it believes are smarter. The general trend is that Europe, North America, East Asia, and Australasia are considered smartest, while Africa, South America, and the Middle East are considered the least smart. This is designed to highlight ChatGPT's racial bias and inherent inequality.
‘GPT Opinion Map‘. 1

ChatGPT has been shown to amplify global inequalities by making assumptions about geographical locations that are not supported by data. AI “makes mistakes and tells people what they want to hear” according to research at the Oxford Internet Institute, influencing search strategies and resource discovery. Using these search strategies can reduce the “serendipity” of discovery, and can also dramatically change how users are interacting with libraries. 

Rebecca questioned the transparency of publishers’ connections with fossil fuel companies, and how skyrocketing profits from eBook sales could be funding carbon emissions. There have been walkouts from journal editors protesting unethical fees for these eResources, and there is growing concern regarding the use of AI within publishing. Purchased goods and services made up 42% of Gardens, Libraries, and Museums (GLAM) emissions in 24/25, which may not even fully capture the scope of emissions related to library purchasing of eBooks and eResources. The effectiveness and integrity of climate pledges across publishing are also questionable – who is holding them to account, and who is measuring their success? 

I was incredibly impressed with Rebecca’s research, and so were several others in the room. She was asked to share her slides with senior members of staff, and will be repeating the presentation to wider university staff groups. A great success, and it really shows how interesting and relevant her work was. These are concerns and developments that affect us all, and I truly admired the effort she put into excavating and illustrating them. 

The presentation closed by recommending the university’s internal Carbon Literacy training module, which anyone can attend. There are certainly more steps to be taken both as individuals and as an institution to investigate and mitigate these impacts. This presentation is one that really stuck with me, and increased my awareness about how all these issues connect.  

1 Kerche, F. W., Zook, M., & Graham, M. (2026).The Silicon gaze: A typology of biases and inequality in LLMs through the lens of place. Platforms & Society, 3. https://doi.org/10.1177/29768624251408919 (Original work published 2026)

Further reading for anyone who wants to learn more:

Trainee Showcase 2025: Part 4

Posted on by — No Comments ↓

This is the fourth and final post in our series on the 2025 Trainee Showcase. If you missed the previous three posts, you can find them directly below this one!

Jake Banyard – Improving user access: wayfinding and resource signposting at Teddy Hall

Written by Millie Krantz

Jake’s trainee project tackled a problem intimately familiar to many library staff: how can we make libraries as approachable and intuitive as possible for readers?

Teddy Hall has a beautiful but imposing college library – it’s housed in the converted medieval church of St Peter-in-the-East and retains many original features like fifteenth-century stained glass. When libraries are in historic buildings that were built to intimidate and impress, readers can understandably have apprehensions about using the space, which poses a problem when we as library professionals want to welcome students in and encourage them to use our resources. In addition, the fact that it isn’t purpose-built means that it can be a bit of a maze, even to users who are familiar with academic libraries. Since it tends to be difficult to knock down medieval buildings just to make them easier to navigate, Jake has had to settle on improving accessibility by reworking the map, improving signage, and ensuring that accurate and up-to-date information is circulated on the college website.

An image of a computer on a desk, accompanied by a keyboard and mouse, a map, a sign about searching the catalogue, and a small teddy bear
Jake’s new and improved map and book-finding guide alongside an important member of staff

Jake’s improved maps take a seemingly simplistic approach, but one that is impressively pulled off: stripping out excess information while retaining and adding things that are actually useful. Anyone who has spent hours trying to rework their library’s map in Canva will understand how hard it can be to include as much useful information as possible without everything becoming visually overwhelming, but Jake’s tactic of splitting up and naming the different sections makes navigation much easier. Jake’s use of icons to indicate amenities not only makes it easier for students to look up where their section might be in the library, but also provides a common language of locations so that giving directions becomes easier, and represents the amenities in a visually different way to reduce the amount of clutter on the map. Furthermore, the maps provide an idea of the building’s actual layout and structure in order to help students navigate their way to the books rather than getting stuck at dead-ends. Jake has made sure to display the map at points where students would need it, alongside other useful resources, like a basic guide to using SOLO. On top of this, his map is designed to function in tandem with improved signage that indicates where study and accessibility aids can be found, improving wayfinding in general.

Webpages and social media are also important ways of helping students use our services easily, especially in college libraries that remain open outside of staff hours. Jake’s useful updates to the Teddy Hall websites ensure that students have easy access to information about printing and accessibility aids, and that various different pages are better linked, enabling students to be more independent and confident in their library usage. Jake also made several posts on Teddy Hall Library’s Instagram explaining what the library has to offer; this direct social media outreach provides an invaluable counterpart to the wayfinding and signage improvements inside the library.

All in all, Jake has planned and executed a number of highly useful improvements to Teddy Hall’s reader provisions, ensuring that the library will better serve the needs of students and that they will get more out of everything the library can offer them.

Hannah Richmond – Data, displays and digital support: my year in review

Written by Ash Lammers

Hannah’s presentation took us through her three main projects this year, which are paradigmatic of the various ways in which we as trainees make lasting impacts on our libraries. 

Shortly after Michaelmas started, Hannah set out to collect and analyse data from the student induction tours that the Law Library runs for new (and returning) students. She aimed to make these tours more effective for staff and students in three key areas: engagement, satisfaction and workforce. For engagement recommendations, she reviewed keywords that appeared most often in positive feedback about the tours (helpful and informative, if you are wondering!), as well as collecting attendee numbers for each tour or induction session held by the library. By combining this data with the improvements suggested by attendees, she concluded that increasing the awareness of inductions among students would be useful to make sure that all who might benefit from an induction could attend, and to increase satisfaction among those already attending. For workforce recommendations, Hannah collated a spreadsheet which compared the number of staff members involved in giving the induction and tours, and came to the conclusion that a reduction in the number of tours/inductions given would be beneficial for both staff workload and student engagement. Naturally, this data came presented beautifully in highly readable pie charts, spreadsheets and diagrams – what’s not to love? 

A collection of books related to Green Action Week arranged into a book display

A second project that accompanied her all throughout the year was the setup of the themed book displays in the Law Library. By curating themed displays for events such as Black History Month and Disability History Month, Hannah was able to highlight the diversity of the law collection and specific intersections of marginalised identities with the law. If you are interested in Hannah’s process, our earlier Disability History Month blog post features some examples from not only the Law Library’s display, but also the Social Science Library and New College Library. While it is widely known that trainees’ hopes of books being taken off the display and read typically remain but a distant dream, Hannah noted significant interest in readers walking past, who would stop to browse. 

Her final project involved recording three bitesize videos on Panopto, in which she took students through the basic steps of how to access digital support materials to assist them in their studies while at Oxford. These videos are currently available online, and will hopefully help many more students in the years to come.

Lilly Wilcox – Fresco: managing web archiving in preparation for the move away from Oxford Mosaic

Written by Elena Brearley

Over the course of the year, Lilly has presented on a few occasions about her work as a Digital Archivist, and each time I have been so impressed by her knowledge and passion for the subject. On the day of the showcase, she explained to us that University IT services are currently transitioning from the web publishing platform Mosaic to the new platform, Fresco. Lilly told us how her work as a Graduate Trainee Digital Archivist working with the Bodleian Libraries Web Archive (BLWA) has been impacted by this change.

To begin with, Lilly confronted us with an intriguing question: ‘Why archive the Web?’. She argued that we should archive the web for the same reason we archive any other kind of material: web archives have legal and evidential value as well as cultural, social, and historical importance. They are a significant resource for research, including for use in data sets.

Lilly introduced us to some key tools and programmes that web archivists use to do their work. An important tool for web archiving is a ‘web crawler’, which is an automated bot that ‘crawls’ and browses through the internet, capturing and archiving websites as it goes. Crawlers are used in other contexts too: Google uses them to find and bring together resources relevant to a search, and AI companies use them to scrape information to feed to Large Language Models.

Another key term for those such as myself who are new to learning about web archiving is ‘Seed URL’. Seed URLs function as a starting point for web crawlers to begin their journey working through and collecting data from a website. Web archivists can schedule crawlers to visit seed URLs at different points in time, and each captured version of the site is then saved to a WARC (Web ARChive) file. This means that past iterations of the site can be compared to the live version.

A screenshot of the Bodleian's Archive-It page

Bodleian Libraries Web Archive, which Lilly works on, was started in 2011 and is primarily focused on archiving University of Oxford websites. Until recently, these websites had been supported by the content management system Mosaic, a system which has posed some challenges to web archivists, including content frequently missing from captures due to being pulled through Application Programming Interfaces (APIs).

With the transition from Mosaic to the new content management system Fresco, Lilly’s team had to quickly make some challenging decisions about which websites to document for the Bodleian Libraries Web Archive. With the limited time, resources and budget available to them before the migration took off, they appraised over 1000 sites and decided to begin archiving over 150 additional websites, mostly for lab groups, centres, institutes, and departments. As they were in this process, sites were already being removed, so they had to work quickly and ended up gathering an enormous amount of data.

Fresco is being rolled out to the university community slowly, so the BLWA team is still learning how archive-friendly Oxford Fresco will be. Lilly emphasised the importance of collaboration and communication with web archivists from the outset of projects such as this one and hoped that people might gain more awareness of web archiving when creating websites (including designing them to be crawler friendly).

Lilly closed with a profound message on the importance of digital archiving, especially in times of political change, pointing to the recent change in administration in the US earlier this year and the drastic changes to government websites that have ensued. A question from the audience prompted Lilly to talk further about the implications of AI crawlers and how some sites have banned crawlers altogether. Hopefully in the future site owners and web developers will be open to working with web archivists to allow permission for their crawlers to carry on their work of documenting the World Wide Web.

Lilly is an effective and persuasive advocate of Web Archiving. She has certainly given me new insight and perspective onto how impermanent and slippery the internet is, and how necessary it is to document its changes over time. I hope she continues to be a champion for this important work!

Gia Simmons – Working papers and a small ‘archive’

Written by Lilly Wilcox

Last but certainly not least, Gia Simmons gave us a look into some of the work that she has done this year on two unconventional collections at the Social Science Library: a small archive of papers from the former International Development Centre Library and a large donation of working papers from the University of Bradford. For her project, she helped the SSL make steps toward understanding the makeup of these collections, where they belong, and how to make them accessible in future.

A collection of books and archival boxes from the Queen Elizabeth House collection

The International Development Centre Library was originally based at Queen Elizabeth House, home to the Department of International Development. This collection of papers came into being from a precursor to the IDC called the Agricultural Economics Research Institute and was created between ca. 1900–1980s. The papers were eventually inherited in 2005 when the contents of the International Development Centre Library moved into the SSL. Because of the complicated chain of custody and time that passed between when the papers were created, acquired by the SSL, and when Gia began looking at them, it had become unclear what the collection comprised.

In this larger collection from the International Development Library, Gia discovered a series of papers from the House of Lords for which the SSL had no record of the contents. The four boxes of material comprise miscellaneous papers relating to the creation of the 20th Report of the select committee on the European community’s agriculture and the environment during parliamentary sessions from 1983–1984. These made their way into the archives through a former Oxford Lecturer in Agricultural Economics, Dr Rosemary Fennell, who served as an advisor to the committee and deposited the papers with Queen Elizabeth House. Gia documented the different archival material in the boxes, taking note of their titles, creators, publishers, and dates of creation, as well as the mysterious set of codes with which the documents were labelled. Equipped with a new understanding of what is in the collection, the SSL is now looking for an archive with which to deposit these papers so they can be made accessible for research.

The second collection was a donation of working papers relating to international development donated to the Bodleian by the University of Bradford, which is weeding their collections. To integrate this donated material with existing SSL collections, Gia researched the full institution names and series names behind the working papers (which weren’t always evident from existing records and the material) to determine whether that series and its papers were already held by the SSL. With this information she was able to either match the material up with an existing shelfmark and barcode them for physical processing, or work toward creating shelfmarks and records for materials that were completely new to the SSL collections.

Gia’s presentation was a fascinating look into the sometimes-unconventional ways that our libraries acquire new material and the massive amount of work that is required to understand and make available these acquisitions.

And Finally, Our Farewell

Sadly, with this post, our time as the 2024-2025 graduate trainees has come to an end. We want to thank all of you who supported us, read our blog posts, or simply put up with us as we tried to make sense of the wacky world of the University of Oxford libraries. Hopefully we will see you all again, but if not…

So long, and thanks for all the fish!