The quantification of qualitative data
This session continues the discussion about the nature of data and focuses on the quantification of qualitative data. Quantification may be necessary to make different data types, scales and levels (somewhat) uniform and comparable, but it is also needed when one works with formal models and simulations. Quantification is hard, as qualitative data typically evades counting (and countable qualitative data is essentially quantitative already). How does that work and what are the pitfalls of the quantification qualitative data? Particular attention will be given to calibration, a key step in many methods that interrogates the change of state of textual data into set membership scores.
- Dr. Sofia Pagliarin (EUR)
- Prof. Roel Rutten (Tilburg University)
About the series
The nature of contemporary society is such that many scholars call for interdisciplinary, multidisciplinary and transdisciplinary research. This is easier said than done. There are all sorts of technical problems – from data collection to congruence between different types of models – but there are deeper, more fundamental issues underneath those. Disciplinary differentiation and the solidifying into highly-specialised niches means that the scientific landscape has sacrificed holism for single-field and single-method expertise. Tell-tale signs include the convention of dichotomizing data into quantitative and qualitative data, the widely-held belief that quantitative research is superior to qualitative research, and an entrenchment of methods in different schools. While differentiation is inevitable, necessary and relevant, and has brought us many good things, it also created and maintained a sort of sectarism along the lines of epistemological and methodological cliques. Expertise has become a dogma to defend and to evangelise as the only possible way. Besides hampering interdisciplinary research, it also narrows opportunities for a healthy dialogue, debate and cooperation across different disciplines and expertise. Multi-data, comparative, multi- and mixed-method research, integrated methods is a sort of antidote to the extreme specialisation of disciplines and methods in the (social) sciences.
We are hence faced with the task to put the pieces together in the face of pressing contemporary issues requiring us to plan and implement a type of research that can look beyond differences. This requires a rediscovery of the fundamentals of each position and the attempt to mitigate the persistent incompatibilities that hinder modelling, data collection and (empirical) analysis to find a space for dialogue. This series "The Practice of Mixed Methods and Mixed Data Research" aims to articulate the various dimensions of the problem and to come up with tentative methodological solutions, that we explore in a series of seminars.
Sessions take place on Fridays, 13:00-14.00 (online):
- 3rd December 2021: Opening session
- 10th December 2021: The qualitative nature of quantitative data
- 14th January 2022: The quantification of qualitative data
- 28th January 2022: The nature of data
- 11th February 2022: One, many, thousands… big
- 25th February 2022: Syntactic and semantic structures
- 11th March 2022: Causality
- 25th March 2022: Doing mixed methods
- 29th April 2022: Teaching mixed methods
Each session will consist of a discussion kicked off by 2-3 panelists and will be followed by a Q&A session with a moderator.
Please register for each session separately, as we aim to keep the group to a manageable size, with a maximum of 25 participants.