Fuzzy Set Qualitative Comparative Analysis (fsQCA)
Date: 8 September 2017
Instructor: Wendy Olsen
Fee: £195 (£140 for those from educational, government and charitable institutions).
CMI offers up to five subsidised places at a reduced rate of £60 per course day to research staff and students within Humanities at The University of Manchester. These places are awarded in order of application. In some instances, such as for unfunded PhD students, we may be able to offer free or bursary places.
Please note: this is not guaranteed and is considered on a case by case basis. Please contact us for more information.
Qualitative Comparative Analysis is a systematic method of studying data on multiple comparable cases from about N=8 through to large datasets of N=10,000 etc. The QCA methods firstly involve casing, i.e. delineating cases; secondly organising a systematic data matrix (we will show these in NVIVO and in Excel); thirdly examining sets of cases known as configurations; fourth interpreting these in terms of ‘necessary cause’ and ‘sufficient cause’ of each major outcome of interest. We demonstrate the fsQCA software for QCA. A fuzzy set is a record of the membership score of a case in a characteristic or set. A crisp set is a membership value of 0 (not in the set) or 1 (fully in the set), and thus is a simplified measure compared with a fuzzy set. Fuzzy sets or crisp sets, and combinations, can be used in QCA. All the permutations of the causal factors, known as X variates, are considered one by one. We test whether X is necessary, or sufficient, or both, for an outcome Y. We then augment the standard measures of ‘consistency’. We show that one can generate both within-group and sample-wide consistency levels for testing sufficient cause.
This one-day training course will attract those doing case-study research, those doing comparative research, and those who want to extend their skills in fuzzy set analysis from beginner to intermediate levels. It will suit qualitative as well as quantitative and mixed-methods researchers; all are welcome.
Learn to compare nested cases, or isolated but comparable cases such as countries.
Learn to measure fuzzy and crisp sets.
Learn to test a pair of X and Y variates for X being sufficient or necessary for Y.
Learn some basics of Boolean algebra (not, or, and, intersection, and superset).
Examine how measures can indicate whether a pattern of data appears to be consistent with sufficient causality.
Examine and run the fsQCA software (freeware available from fsqca.com).
Consider matters of sampling and population-wide descriptive statistics for the data, and whether to statistically test fsQCA results.
- As an intermediate course, it presumes that you have either of the following backgrounds – either some use of statistical regression; or some experience with verbatim transcripts or document analysis as forms of qualitative analysis of detailed data.
- To ensure all participants reach an intermediate level, about 2-3 hours should be spent on the preliminary readings, and one hour looking at examples on this website: www.compasss.org (Bibliography section). Notice that the latter does include many data sets.
- Marx, A. and G. van Hootegem (2007). "Comparative configurational case analysis of ergonomic injuries." Journal of Business Research 60(5): 522-530.
- Kent, R. (2005) ‘Cases as configurations: using combinatorial and fuzzy logic to analyze marketing data’, International Journal of Market Research, 4, 2, pp. 205-228
- Olsen, W.K. (2012) Data Collection: Key Trends and Methods in Social Research, London: Sage, sections on case-study research.
- Olsen, W.K. (2009), Non-Nested and Nested Cases in a Socio-Economic Village Study, chapter in D. Byrne and C. Ragin, eds. (2009), Handbook of Case-Centred Research Methods, London: Sage.
- Ragin, C. (2008). Redesigning social inquiry: Set relations in social research. Chicago: Chicago University Press.
- Snow, D. and D. Cress (2000). "The Outcome of Homeless Mobilization: the Influence of Organization, Disruption, Political Mediation, and Framing." American Journal of Sociology 105(4): 1063-1104.