Dear Arshad Haroon,

I suggest you to see links and attached file on topic.

https://journals.sagepub.com/doi/abs/10.1177/1525822X09332633?journalCode=fmxd

http://methods.sagepub.com/book/configurational-comparative-methods/n4.xml

Article Manipulating measurement scales in medical statistical analy...

Best regards

Dear Arshad Haroon, abstractly, It does. csQCA and mvQCA require characteristics to be coded as binary and multi-valued (more than two discrete values, usually three) variables, respectively, while fsQCA allows a characteristic to have any continuous value from 0 to 1. This methods brief focuses on fsQCA. the links provided by dear Mohamed-Mourad Lafifi are more direct.

Dear Arshad,

Thanks for sharing this question.

Generally, it is possible to combine all three set types in a single analysis. The generalization of this procedure, where even multi-value variables become fuzzy I have termed "Generalized-Set QCA" (gsQCA), introduced here: Article Unifying Configurational Comparative Methods: Generalized-Se...

The problem(s): first, gsQCA has not been put and tested in software yet (see also this related post on ResearchGate: https://www.researchgate.net/post/gsQCA_requirements_for_condition_variables ), and second, fuzzy sets really have no well-fitting theory of causation. Both csQCA and mvQCA (the former is just a special case of the latter), are well-grounded in a theory of causation and have correspondingly well-developed algebrae underlying them. But what does it mean that some cause is a cause to degree 0.34 under fuzzy sets? That has neither been clarified nor is it intuitive. So what fsQCA does in the end is to act again as if everything was binary. That's why fsQCA solutions always look like a straightforward csQCA solution, and people also continue to interpret them in a binary fashion.

The most solid thing to do, from my perspective, is to transform your interval-scale variables into multinomial variables with well-chosen clusters, and use mvQCA. That method is, in many respects, much more powerful than fsQCA because, as mentioned above, fsQCA still has to treat all variables and causation as a binary phenomenon in the end, whereas mvQCA relaxes this restriction, and provides you with more fine-grained answers. The price you have to pay, however, is that the more variables/values per variable you have, the more data you need to reach relatively clear-cut solutions. But that's something which is, of course, not peculiar to QCA, but any method of data analysis. The more/better insights you want, the more data you have to provide.

I hope this helps.

Best wishes,

Alrik

Thanks Professor.