Susie and I were delighted to have been invited to the University of Sussex this week to talk about our project at an NCRM event on case histories. Organised by case study experts, Rachel Thomson and Julie McLeod, the event brought together an international and interdisciplinary group of researchers working with, and negotiating, ‘cases’ in qualitative longitudinal research.
An overriding theme was the tricky question of what makes a case ‘a case’. A case, it seems, can be many things. They can be selected theoretically or pragmatically; they can represent the ‘typical’ or the extraordinary; and they can relate to, and shape, both process and outcome. As noted by Rebecca Taylor, a case study design can help the qualitative researcher ‘frame’ their research, enabling a stronger analytical grip, particularly when handling large or complex data. But whilst the case can provide the means through which to make sense of the messiness of our data, it doesn’t necessarily have to structure it. The case in itself can be temporal. Its boundaries can shift (and be shifted) over time, responding to events and actors in the field, new points of comparison and emergent theoretical possibilities.
Two points made at the training spoke to our research. The first is that despite their fluidity, the very nature of case studies is reductionist: decision making about how your cases should be bounded, of course, results in some things being included, and other aspects being excluded. Second, is that while a case may be bounded in some way (a case can be a town, an organisation, a family, an individual), it should always seek to illuminate something bigger, and be larger than the sum of its parts.
If you have been following our work, you will know that we are working with secondary data from six of the projects archived under the Timescapes initiative. Together, this dataset amounts to 1,000 documents, and 165 ‘sets’ of data (cases?) relating to individuals and families. We have been using the metaphor of the archaeological excavation as a means of approaching our dataset. Working as ‘aerial archaeologists’ we began by completing a ‘surface survey’ of the dataset, looking at its contours and texture. A ‘geophysical survey’ followed, utilising new computational methods through which to conduct a corpus analysis of the dataset. With this big ‘aerial’ view in place, we began the process of digging into our data. Initial ‘shovel test pits’ involved the selection, and analysis, of a small number of selected cases. We are now planning ‘deep excavations’. These will involve a larger number of case studies, but focused on a narrow research question.
Our process is very much work in progress. Key questions for moving forward are how should we both define a case, and in turn, how our cases should be selected. In essence our aim is to demonstrate the ways in which corpus analysis (the aerial view if you like) can assist in the analysis of large volumes of qualitative data, whilst retaining the forms of analysis which provide biographical depth, and insight into the complex micro dynamics of how, why, and in what circumstances, change happens. Most critically we want to demonstrate how these approaches can be brought into dialogue – enriching each other, rather than sitting in opposition.
We continue to develop our work and will be reporting our progress here. We have a forthcoming event in London on the 18th October where we will discuss our archaeological metaphor in more detail. Further training events are being planned for the months ahead.
We will finish by saying thank you to all the presenters and participants at the event. It was incredibly valuable, and has given us many, many ideas for developing our project. We look forward to sharing our explorations with you.