In the research project Mapping the digital transformation and hybrid education in Brazilian public networks, we have reached an important moment. Until now, Miro has been our main recording space: there we have been laying out categories, notes, excerpts from documents, and connections between elements that emerge from the field of the Innovation Network for Hybrid Education (RIEH). The image below shows the current state of this map, and it speaks for itself. The number of nodes, sticky notes, and connecting lines has grown so much that reading it has become practically unfeasible. The map, which was supposed to help us see, has started to conceal.
This led us to a new discussion: which software can help us organize the data while also contributing to the analysis process? We don't just want a place to store information. We want tools that help us follow flows, perceive relationships, and bring forth what we are not yet able to see.
Yesterday and today I ran some exploratory tests. I tried Obsidian, which works with interlinked notes and allows the connections between them to be visualized as a graph, something that resonates strongly with the rhizomatic logic of cartography. I also tested Gephi, a tool for network analysis and visualization, which opens up interesting possibilities for examining the density and centrality of the relationships we are mapping.
I also ran a test with artificial intelligence, using Claude to transform the transcripts of a few interviews into a corpus compatible with IRaMuTeQ, and ChatGPT to generate a similarity graph from that material. It was just an exercise, to understand what these tools can offer (and what precautions they require), but it has already given us clues about possible paths for lexical analysis.
Take a look at the result:
The next step is to discuss with the group which paths, software, and procedures we will adopt. And here is something I consider central: this discussion is not a "technical" stage separate from the research. In the cartographic method, there is no predefined path that we simply carry out. The journey itself traces its own goals. Thinking about how we produce, record, and store data is also part of the investigation, because the tools we choose shape what we are able to see.
And so the cartographic process continues: mapping the field and, at the same time, mapping ourselves as we research.
If you have worked with any of these tools, or others, in qualitative or mixed-methods research, please share your experience in the comments. Every contribution is welcome!
P.S.: Since the project provides for the exclusion of sensitive data and compliance with the LGPD (Brazil's General Data Protection Law), the transcripts used in the test were anonymized before being processed by the AI tools.
Nenhum comentário:
Postar um comentário