Which digital research tools can we use to enrich ethnographic methods? At the end of March, we held an introductory session on the network analysis tool Gephi, which was guided by this question. Gephi can be used to visualise tabular data sets as graphs. The programme is typically used to analyse large data sets, which are often collected automatically. In our workshop, together with Mace Ojala from Ruhr University Bochum, we considered whether and how Gephi can also be used effectively for qualitative research with ethnographic methods. Both students and researchers from the ‘Fixing Futures’ graduate college took part. A big advantage for everyone: the software is open source and therefore freely available.
The graph, Gephi’s central function, has the typical form of a network of nodes and connecting lines, known as ‘nodes’ and ‘edges’. Network semantics has become established in almost all areas of society over the last few decades: the metaphor of the network can be used to describe the relationships between discrete instances (e.g. individuals, organisations or objects). It thus offers a possible perspective on the world that has a long tradition in science and technology studies, for example in actor-network theory or in relational approaches. The research software Gephi was developed to enable the visual representation and computer-assisted analysis of networks on the basis of a database.
In our workshop, we first clarified some basic epistemological assumptions of network thinking in order to develop conditions for the meaningful use of Gephi and comparable tools for ethnographic research. These are characterised by the fact that they mostly examine practices and focus on the relationships and entanglements of their empirical objects. The metaphor of the network can help to recognise the connections between human and other entities in the field that arise through practice and in practice, to transfer them into a symmetrical scheme and thus make them accessible to scientific analysis. Guided by Mace Ojala, we were able to jointly explore this underlying logic of network analysis and consider which aspects of our various research projects we would like to digitally map in this way. We also discussed the difficulties of translating ethnographic material into the network logic of Gephi: observations from the field cannot always be transferred into the tabular form required for the creation of a graph. For example, the intensity or relevance of a contact can be visualised using numerical weightings or colour codes. However, many decisions must be made in advance, such as how intensity or relevance manifest themselves in the field and whether they can be quantified appropriately in all cases.
The introduction was therefore a good opportunity to look at the various STS research projects at Goethe University in Frankfurt with new methodological questions and to discuss the potential and obstacles of digital research tools. What we take away: Working with the Gephi software tool not only allows us to generate clear graphs, but also helps us become aware of our own scientific working methods and make implicit ethnographic assumptions explicit. A valuable practice for our projects!