The challenge
Data science continues to evolve rapidly, yet the composition of its workforce does not fully reflect the diversity of the communities it seeks to serve. This limits the sector’s ability to generate research that captures the complexity of real‑world experiences and contexts.
At the Leeds Institute for Data Analytics (LIDA), the Data Scientist Development Programme (DSDP) recognised that traditional recruitment routes were not reaching the full breadth of talent available. For example, Black data scientists represented only 4% of appointments between 2016 and 2020, despite clear evidence of strong interest and capability across underrepresented groups nationally.
LIDA identified that addressing this disparity required organisational, not individual, change. Building a research culture grounded in equity, diversity and inclusion (EDI) is fundamentally about representation, belonging and access to opportunity.
When early career researchers see themselves in a role, they are more likely to recognise it as a space where they can thrive. Diverse teams bring a wider range of perspectives, experiences and analytical approaches, strengthening the relevance and quality of research.
The challenge was therefore to evolve recruitment and culture so that talented data scientists from all backgrounds could access, succeed in and shape the field.
The approach
Introducing Positive Action (PA) within the DSDP marked a strategic step to widen access to opportunities and ensure that exceptional candidates from underrepresented groups could see the programme as a place where they belong.
PA, permitted under the Equality Act 2010, enables an employer to appoint a candidate from an underrepresented group when candidates are of equal merit. It is a proactive, principled way to broaden participation, strengthening recruitment by expanding reach and representation.
The first PA initiative reserved a role for Black data scientists, supported by DATA‑CAN (the health data research hub for cancer) and informed by local and national workforce data.
The recruitment process maintained full merit‑based shortlisting, with assessors blind to PA eligibility, ensuring fairness and rigour throughout. Since the success of this approach in 2022, the DSDP has successfully written cases for, and delivered PA recruitment to support, the underrepresentation of global majority, women and candidates from low socio-economic backgrounds.
PA recruitment is a useful tool, but perhaps its key benefit is in sending a clear message to applicants that this is an inclusive programme where diversity is not only welcomed but valued as a research strength.
Beyond recruitment, the DSDP embeds values‑based practices that emphasise community, belonging, wellbeing and shared learning. Candidates are encouraged to articulate what “data science for public good” means to them, and cohort‑building activities during induction foster an environment where early‑career researchers can develop both technical confidence and leadership skills.
This approach enables diverse researchers not only to enter the sector but to influence its future direction.
The impact
The introduction of PA immediately broadened the visibility and accessibility of DSDP opportunities. In 2021, appointments of Black applicants increased to 33%, a significant shift demonstrating the effectiveness of widening reach and offering relatable pathways into data science careers.
Since 2023, the DSDP has also maintained a 50% appointment rate of women on the programme.
The DSDP community is rich in its diversity as a result. These cohorts are producing research that is more reflective, critical and socially grounded.
Staff stories
Precious-Gift Alele: gaining confidence in data science
For me, the Data Scientist Development Programme was a project‑based data science development programme where you got to work on real‑world data tailored to your areas of interest and abilities. You get to experience how data can influence decision‑making, patient care, or health outcomes.
While I was doing my Master’s degree at the University of Leeds in Health Informatics with Data Science, I was really focused on gaining real‑world experience. I have a health background, and I wanted to see how data could influence decision‑making in healthcare and potentially improve patient outcomes.
They weren’t looking for people who had all the experience in the world – they were looking for people who could grow and learn, which I was really happy about.
I didn’t have the technical background, but I had the medical and clinical background.
Through the programme, I worked on health‑related data, including data relating to giant cell arteritis, a rare vasculitis disease. Exploring how to improve early diagnosis for these patients was the most meaningful part for me.
Because it is a cohort‑based programme, you have a community to ask questions, share ideas, and help each other. The diversity of backgrounds – nationalities, academic paths, and experiences – created a beautiful mix. We supported each other and genuinely formed friendships.
Completing the programme gave me the ability to really believe in myself. I learned that I can learn on the job in a supportive environment where everyone is rooting for you. It opened my eyes to the fact that I can do anything I want to do.
Now, going into clinical practice, I can think about better ways to use both my clinical and data science skills to improve outcomes for patients and clinicians alike. It truly empowered me.
Jacob van Alwon: pivoting into data science
I’d been looking at data science while I was finishing my PhD because I learned to code during my PhD and really enjoyed it. After nearly two years working in engineering, I realised I missed writing code. I wanted coding to be a main part of my career.
My experience was mostly in engineering, and although the work was interesting, I didn’t have hands‑on data science experience. I’d applied for other data science roles and wasn’t getting interviews, which was understandable. This programme seemed like the ideal way to gain that experience and transition into more general data science roles.
Through the Data Scientist Development Programme, you learn what you need to do in an academic research‑type setting but with industry partners. You still get exposure to industry environments, but with academic roots.
That mix worked well for me because I had an academic research background before joining.
I think having six months to work on each project gives you more time than you might get in industry. Especially if you’re new to coding, the six‑month structure gives you time to find your feet.
My first project involved using smart water meter data to predict tourism trends. That work eventually became a journal paper that I’m now a co‑author on, and I also presented the findings at a conference.
My second project was with Morrisons. Although the start was rocky, I raised my concerns and was able to explore a different problem. I worked on delivery vehicle routing optimisation, and it turned out to be a great project. By the end of the programme, they offered me a job, which I still have now.
During the programme, I also experienced EDI support first‑hand. When I suspected I might have ADHD, I was supported and directed to resources. It definitely changed my awareness of EDI.
Toluwani (Tolu) Osabiya: accessing data science fairly
A major challenge I faced was that because I am not British – I’m an immigrant – employers weren’t always willing to hire someone on a student visa. Even when they knew I would later have a postgraduate visa, they still had to decide whether to sponsor or not.
The Data Scientist Development Programme stood out because those requirements were removed. You didn’t need sponsorship, and you didn’t need indefinite leave to remain before joining the programme. That was one motivating factor.
The other was that the application process suggested I genuinely had a chance. It focused more on what you knew and what you could do, rather than just previous experience. From the interview process alone, I could tell it was a development programme, and since I was early‑career, it felt like a good fit.
The programme felt like a platform where, as someone early in their career, you could get a job not necessarily because of extensive previous experience, but because you were interested, had some knowledge, and were eager to grow. It really was a platform to learn and earn at the same time.
In terms of inclusion, it helped that I worked with people from different cultural backgrounds. I didn’t feel out of place because I wasn’t the only non‑British person on the programme. Kylie was really helpful in bringing people of different ethnic backgrounds together — she made everyone feel at home.
On my first project, I brought forward ideas that felt unfamiliar at first. I stood my ground and also tried to understand their reservations. I showed how a different approach could work, structured the solution so it could be reused, and they were eventually happy with the work. In fact, my supervisor even took an interest in learning Python afterwards.
I’m very grateful I went through the programme. Even though I later left the UK due to visa issues, I can now engage confidently with more experienced data scientists. I don’t feel out of place and learning through real‑life projects made all the difference.
Further information
Find out more about the Data Scientist Development Programme.
For any enquiries about visa eligibility for the DSDP, email Programme Manager Kylie Norman via k.r.norman@leeds.ac.uk.