As two separate research teams - one based in Business and the other in Law - we recently undertook SCiLAB-funded projects using methods with similar names: a World Café and an Evidence Café. Despite the shared language of informal, conversational enquiry, use of the two approaches revealed important differences in both design and purpose.
In this blog, we reflect on our experiences of running these sessions with students - one face-to-face and exploratory, the other online and evidence-led - and consider what each method offered, where each struggled, and what we might do differently next time.
The World Café (Lohr et al., 2020) brought together a cohort of young management apprenticeship students for a face-to-face research day. The students rotated around tables, responding to open questions about their experiences, writing directly onto tablecloths, and building on each other’s ideas in successive rounds. The emphasis was on participation, reflection, and shared sense-making.
In contrast, the Evidence Café (Clough and Adams, 2017) was conducted online with a smaller group of level one law students. Participants first completed a poll about their views on generative AI, then engaged with a series of expert inputs - including practitioner perspectives and institutional data - before taking part in a facilitated discussion. The same poll was repeated at the end, with the intention of capturing any shifts in perception.
Following the Cafés, the two research groups met together to discuss what we did. We recorded the meeting and used the transcript – with some help from Microsoft CoPilot – to pull together this comparative blog.
One of the most striking similarities was that both methods were highly structured in design. However, the approaches played out in very different ways.
The Evidence Café was structured and directive: participants were given curated evidence and invited to respond to it. In contrast, the World Café was structured but open-ended: participants were guided through a process, but not given content to shape their views.
This highlights a fundamental difference: The Evidence Café feeds ideas in; the World Café draws ideas out.
The Evidence Café was designed to explore whether exposure to evidence might shift student perspectives, in this case around the use of generative AI. The inclusion of pre- and post- polls reflected this intention. By contrast, the World Café deliberately avoided leading participants. The aim was to surface lived experience around their apprenticeship programme, without influencing responses.
In practice, this meant that the key purpose of the Evidence Café was pedagogic in nature, students learning was subtly developed as they engaged. The World Café offered a slightly different nuance, prioritising the capture of authentic voices - and where shared learning was often between participants.
Interestingly, both Café research methods highlighted the challenges of data collection and analysis.
In the Evidence Café, quantitative polling was intended as a key output. However, this approach met with practical challenges (such as students struggling to access the polls when using an external log in). This meant that the richest data instead came from the ensuing discussions, rather than the measurement as anticipated. In contrast, the World Café data was generated through written notes, visual traces, artefacts (such as drawings) produced by learners, and plenary discussion. This created a rich but messy dataset, which proved challenging for analysis.
A shared reflection we had was that it seemed that the more creative the method, the harder it is to analyse data generated.
Both methods created space for meaningful student engagement, but in different ways.
The World Café enabled quieter students to contribute more comfortably, by writing before discussing. It also worked well where participants already knew each other, creating a safe and collaborative environment.
Despite being online and involving participants unfamiliar with one another, the Evidence Café also generated rich and thoughtful discussion. The provision of evidence appeared to support engagement by giving students something concrete to respond to, and this was enhanced by the experience and skill of the Associate Lecturer (OU tutor) facilitating the discussion.
From a practical perspective, the methods differed significantly.
The World Café required extensive preparation, careful facilitation, and significant logistical input ahead of the face-to-face event. In contrast, the Evidence Café was simpler to organise, particularly online, but arguably less effective in that format than it would have been face-to-face. It did however allow students to join easily from across the regions and nations and the transcription is now available for future thematic analysis.
Both approaches also raised questions of value generated for input required. The World Café was resource-intensive, but delivered strong developmental benefits for participants, alongside time spent together in generating ideas and group bonding. The Evidence Café proved a more efficient use of staff and student time, but given the limited time available, the requirement for careful planning and a tightly focused agenda meant less opportunities for informal learning and ideas exchange.
Across both projects, some shared lessons emerged:
Rather than identifying a ‘better’ method, our experience suggests that each approach serves different purposes. We can see value in using a World Café if you want to explore experiences, surface perspectives, and encourage collective reflection. On the other hand, we might use an Evidence Café if the aim is engaging participants with information, testing their reactions, or exploring changes in understanding. Both methods can generate rich discussion, but neither yields quick or easy routes to data analysis.
.png)
Despite their shared use of the term ‘Café’, the World Café and Evidence Café represent two distinct ways of engaging participants in research projects. One privileges openness and emergence; the other introduces direction and influence. Used thoughtfully, both offer valuable ways of working with students - not just as research participants, but as active contributors to learning and enquiry.
This blog is based on two different research projects using Café techniques that were supported and funded by SCiLAB – The OU Scholarship Centre for innovation in online Legal and Business education. Project references SCiLAB252607 and SCiLAB252610.

Sarah is a Senior Lecturer in Work and Organisational Learning and the Director of Undergraduate Apprenticeship Qualifications in the Open University Business School.
Her research and practice focus on how individual and collective managerial effectiveness can be improved in the workplace, recognising that each work situation is unique.

Hayley is a Senior Lecturer working across both the Business and Law programmes at the Open University. Hayley's scholarship interests focus on developing online pedagogy and issues of student retention, progression and success in a distance HE setting.
Her current research interest is focused on the changes to space and place in the blended post pandemic HE setting. Hayley is a Senior Fellow of the Higher Education Academy.

Helen is an Associate Lecturer in the Open University Business School, teaching undergraduate and postgraduate programmes, including degree and senior leader apprenticeships.
A practitioner with commercial experience, she is passionate about bridging academic theory and professional practice.

Fred is a Senior Lecturer in Law and a Fellow of the Higher Education Academy.
Fred’s scholarship interests relate particularly to assessment, in particular innovative and authentic assessment that replicates and tests practical skills alongside knowledge.

Fran's research interests include storytelling and mythmaking in public and organisational discourse, how individuals develop and craft their experiences at work and in society.
Fran has a PhD in Political Economy from the University of Manchester and is a Senior Fellow of the Higher Education Academy.
This blog includes content enhanced with the assistance of Microsoft Copilot (GPT-4), a generative AI tool developed by OpenAI and integrated into Microsoft services.
