Cornelius Schubert – “Coding Data in Software and in STS”

Comment by Miriam Bachmann

With Cornelius Schubert we take a deep dive to discover what experimentalist data studies may be as we learn about the challenges and fascinations of interdisciplinary collaborative research and the combining of different ways of processing data to design novel visualizations for rich data in healthcare.

In „Coding data in software and in STS“ Schubert traces collaborative data and design practices in an interdisciplinary project that brings together the disciplines of neurosurgery, sociology, and computer science. At each stage of the project and in each discipline, data play a central role as they come about in multiple „shapes and sizes“. While the neurosurgeon or nurses may jot down medical data on paper and pre-classified tables, the sociologist may collect data digitally from conversations amongst medical personnel, creating other forms of codes and classifications. And the design engineer may use digital information systems to visualize “the“ data. By referring to Annemarie Mol’s „The Body Multiple“ Schubert makes us aware that data as well is multiple, considering the three disciplines it is in fact data³. The project speaks of data³ because all three disciplines come from different data practices. While the data practice of medicine revolves around diagnosis and treatment, the data practice of sociology aims at participant observations and interviews, and the data practice of computer graphic revolves around developing digital visualizations. Across these disciplines, data has to be translated and transformed which presents itself as a comprehensive undertaking. Considering i.e. that clinical data such as the physical symptoms of feeling pain are difficult to visualize, the data has to be viewed through the specific lens of each discipline until it is visualizable.

This process requires a substantial effort in interdisciplinary conversation. As the understanding of data can be diverse throughout the disciplines, it is important to communicate one’s own perspective in the attempt to discover possible connections. But not all differences can be overcome.

Taking a look back to the 1980s and 1990s the terrain for interdisciplinary collaborative research has not always been easy and it is still not today. Researchers should make themselves aware of their intentions towards working collaboratively and consider the limitations of their work. In Schubert’s project the line was drawn between the redesigning or critiquing of a partnering discipline and their re-combining. While the former were avoided as an invasive practice, the latter was priorities, as it allows processes of reflection and collaboration.

Perhaps the working with differences is the essence of experimental data science. „One of the main experimental features of the program is that the different disciplines have different ways of finding solutions. Some push more for creation and others push more for design […]. How do we use the design as a mode for engagement“, Schubert asks.

First a look at the empirical field (all three disciplines) is necessary to distinguish certain characteristics such as data dense environment, space for collaboration, the practice of note taking, coding data, etc. It may become apparent that some of the working processes are able to be combined, such as in Schubert’s project the coding in sociology and the coding in computer science. Throughout the design process that follows, researchers should view their own discipline through the lens of the other and maintain an openness for collaboration as it requires continuous and frequent engagement. 

Interdisciplinary research is „qualitative research outside of your comfort zone“, states Schubert – and it comes with a lot of invisible work. But in the case of Schubert’s project, using the design of digital technologies in the visualization process, offered each collaborating discipline the opportunity to reflect on its own ways. The developing of digital visualizations partly from qualitative sociological data is an exiting example of experimentalist data studies.

The idea of interdisciplinary research comes with the romantic implication of collaborative problem solving – and in an ideal academic world this seems like a desirable way to conduct research. But the efforts that have to be undertaken to participate in this form of experimentalist data studies seem to collide with the precarious realities of academia and will exclude many researchers – especially younger researchers or female researchers – from the get go. I am left to hope that this project presents a unique example of academic passion. If experimentalist data studies executed as stated, yield results beneficial to academia, academia should level the playing field to make this kind of research more accessible to the scientific community.