Our approach to research emphasises cross-project collaborations and transdisciplinary thinking. But what does this mean, in practical terms, for the work that we do and for our participants? Here, we consider these questions in relation to sharing interview transcripts among our team. We explain how we developed practices to create shareable data, and explore its implications for participants and the way we do research.
Nexus thinking invites us to dwell on the connections between science, welfare and human health in animal research, and think about overlaps between the topics that we research, from species to sites and practices to people. At the same time, we aim to explore ways of working that allow us to pool the range of perspectives that each team member brings to this project. Therefore, we have been reflecting, since the beginning, on our research practices and what our collaborative nature can mean in terms of methodology, research ethics and data.
Among these questions, the possibility of sharing interview transcripts among the research team stands out as particularly important. Qualitative interviews are rich forms of data and can yield valuable insights into a variety of issues, even when the discussion is narrowly focused. They often meander between different topics, and at any point can bring up data that are relevant to multiple projects. Moreover, we all, as analysts, can read and interpret data in different ways.
Therefore we have developed joint working practices as part of our planning process, in order to address transcript sharing from the start. The development of data management practices is an important part of good research practice according to the Concordat on Open Research Data, and will enable our data to be part of the UK Data Archive in the future.
The AnNex team discussed what data could and should be shared, and under what circumstances; and agreed that we could create useable and shareable interview transcripts from interview audio recordings by carefully removing identifiers. This strategy sets out how we can approach the creation and processing of interview data with a view to sharing it among us systematically and with ease. It also maintains a degree of flexibility so it can meet the requirements of our individual institutions and disciplines. This is important since, for example, anonymisation is standard practice in the social sciences but not for oral history interviews, where the identity of interviewees and other such information are important.
The first defining characteristic of this strategy is that we ask our participants, at the time of interview, for their explicit consent for the transcript to be shared. The second is that we have an AnNex-wide approach to identifying interviews and assigning pseudonyms to participants and sites where we do research. We have assigned specific letters of the alphabet for each project that can be used to pick pseudonyms. We have also got a common list for pseudonymised place names. This enables us to process data as it is produced, creating a data set that can be pooled across projects. This and other information is captured on a standardised transcript cover sheet that we have developed. It is designed to give future readers some “metadata” about the context in which the interview was produced.
We have begun to see the potential of this approach to data production for organising our interviews along more collaborative lines, too. Where participants’ experiences cut across topics, we may choose to hold interviews with researchers from different projects, or pass on a question to be asked on our behalf where appropriate. In so doing, we seek to streamline our interview process and reduce the likelihood that participants are asked for multiple interviews. This allows us to make the most of the time that our stakeholders have to engage with us. This development is an example of the ways in which our interest in incorporating data sharing (and protection) into our project from the start is engendering new possibilities for experiments in collaborative research methods.
We have been applying the common identification practice with success since the beginning of the year, and look forward to sharing the first transcripts in the coming months, once the IT is in place. We are very grateful to our participants for their insights, and look forward to developing joint analyses that can be as rich as the conversations that we have had with them and each other since we have begun this project.
Image credit: “Vintage File Cabinet” by victoriabernal is licensed under CC BY-NC-SA 2.0 via Flickr