This week marked a shift from thinking about what we want to research to thinking much more critically about how we produce research that is robust enough to be useful. As we move into the data collection stage, I have found myself thinking much more critically about the reliability of evidence and the importance of developing systematic research methods.

The team began the week by completing the first draft of our literature review. My section focused on the impacts of social enterprises and why they matter. The literature repeatedly highlighted outcomes such as community resilience, improved wellbeing, social inclusion and environmental sustainability, reinforcing my view that social enterprises create value in ways that traditional economic indicators often fail to capture. However, the literature also raised new and challenging questions. If most of the impacts of these organisations are intangible and long term, how can they ever be fully measured? How can we compare them to what would have occurred without intervention? There seems to be an inherent tension between producing evidence that is measurable and recognising the complexity of social change.

As a group, we decided to adopt a mixed methods approach. Initially, I viewed this primarily as a way to collect more data. However, I have since come to appreciate that its real strength lies in allowing different forms of evidence to complement and challenge one another. Discussing our methodology with our supervisor, Mengchen, highlighted areas where our approach required stronger justification and reminded me that every methodological decision needs to be explained rather than simply assumed.

Perhaps the most significant learning experience this week came from our work constructing the social enterprise database. Our original preliminary database had been complied largely using a large language model, without a systematic collection strategy, and it only contained around 450 organisations. While this provided a useful starting point, it also highlighted the risks of relying too heavily on AI. Relevant organisations could easily have been missed, while some entries may have contained hallucinated information. This reinforced to me that while AI can be an excellent starting point for exploratory work, it cannot replace systematic evidence collection when the credibility of a research project depends on the accuracy and completeness of its data. Hence, we decided to rebuild the database using transparent and verifiable public data sources.

I identified and integrated multiple datasets, including the Charity Commission register, Companies House records, and Social Enterprise UK’s membership directory. I spent a considerable amount of time writing Excel scripts to automate data cleaning, and compare and merge sources. Surprisingly, I found this task genuinely engrossing! As the database became increasingly complex, I attempted to find new ways of organising the team’s workflow to maximise efficiency. I introduced my teammate Angela to the coding methods I had been using and created a shared task sheet to document changes and divide responsibilities. I also produced a project timeline for the team. In doing so, I began to appreciate how project management is an essential component of collaborative research, rather than an additional chore. Reliable research depends not only on imaginative questions and strong methods, but also on effective communication within the research team. Now that we have developed more efficient systems for sorting through our data, I can see how our database will, hopefully, be one of the most valuable outputs of our project, providing OSEP with a systematic resource to aid their policy advocacy.

Stepping away from the spreadsheets, the cohort also attended the ‘Uncomfortable Oxford’ tour, which encouraged me to reflect on the University’s relationship with the city in a way I had not previously considered. Before the tour, I had largely viewed the ‘Town Gown Divide’ as a light-hearted rivalry. However, learning about the violence, inequalities and exclusion that characterised this relationship made me realise it represents a much more significant social issue. It led me to question further what responsibility current students have to engage more meaningfully with the local community, rather than just remaining within the University’s own social and physical boundaries. Furthermore, coming from a college, St Anne’s, that was established by women of the local community, reminded me that the opportunities that exist for myself and others today exist because previous generations challenged deeply embedded systems of exclusion.

Looking back over the week, I was struck by a common theme: lasting change depends on strong foundations. Whether constructing a rigorous research database or challenging long-standing systems of exclusion, meaningful progress comes from careful, often unglamorous work that enables others to build further. As our project moves into data collection, I hope the systems we have developed this week will provide that same foundation for producing research that is both academically rigorous and genuinely useful to OSEP.