Dear ResearchGate-community,
As a beginner in Qualitative Comparative Analysis (QCA), I would like to ask you whether I should apply and report analysis of necessity, i.e. necessary conditions, in my paper in which I want to infer causal relations between my exogenous factors and the outcome. I am asking because I found contradictory suggestions.
In favor of the analysis of necessity are, in my field of research, for instance, Fiss (2011) and Leppänen et al. (2019). Against it are, to my understanding, for instance, Thiem (2017, p. 427-428) or Thiem (2018, p. 626).
Thank you in anticipation.
Best regards,
Philip
References
- Fiss (2011): Building better causal theories: A fuzzy set approach to typologies in organization research
- Leppänen et al. (2019): Qualitative Comparative Analysis in Entrepreneurship: Exploring the Approach and Noting Opportunities for the Future. 10.1108/S1479-838720190000011010.
- Thiem (2017): Conducting Configurational Comparative Research With Qualitative Comparative Analysis. 10.1177/1098214016673902
- Thiem (2018): Improving the use of qualitative comparative analysis for inferring complex causation in development and planning research. 10.2166/washdev.2018.149
Dear Philip,
a very good question! You say, you want to infer causal relations in your data. By analyzing merely necessary conditions, as practiced in a lot of QCA research, you can't get at causes. John S. Mill, and later also John L. Mackie, had already shown about 200 years ago that the idea that causes are necessary conditions is hopelessly inadequate.
The idea that causes are necessary conditions had been formulated by David Hume more than 250 years ago in his famous "An Enquiry", where he says that "(...) we may define a cause to be an object, followed by another, and where all the objects similar to the first are followed by objects similar to the second. Or in other words, where, if the first object had not been, the second never had existed."
In the mid 19th century, however, John S. Mill had already shown in his voluminous "System of Logic" that Hume's definition does not get us very far - causal relations are of a considerably more complex structure. In particular, he already developed the notions of conjunctural causation and the notion of equifinality. Concretely, he says in Book III that "The cause indeed may not be simple; it may consist of an assemblage of conditions" (conjunctural causation). And he further argues that "One fact may be the consequent in several invariable sequences; it may follow, with equal uniformity, any one of several antecedents, or collections of antecedents" (equifinality).
In "The Cement of the Universe", John Mackie (1973) later takes up these original ideas again and further formalizes his advancements using modern tools of propositional logic, particularly in chapters 2 and 3 of his book. The upshot of Mackie's work has been the famous INUS Theory of Causation, and, as put by Ragin himself: in QCA "causes are viewed as INUS conditions: insufficient but necessary components of unnecessary but sufficient combinations of conditions" (Ragin and Strand, 2008, 431-432). Let's put all the above on a more formal footing:
According to the Boolean definition of necessity, P is necessary for Q if the conjunction of the presence of Q and the absence of P is not the case. In formal notation, necessity is defined as follows:
P P ≡ P + ¬Q ≡ ¬(¬P*Q)
where "¬" (negation), "+" (disjunction), "*" (conjunction), and "->" (implication) are the Boolean operators for denoting the logical concepts "not", "or", "and", and “if, then" respectively. QCA researchers often use the terms "condition" and "cause" interchangeably, but it is important to realize that the above definition does not entail that P is a necessary cause of Q.
That, of course, does not mean that INUS causes cannot also be necessary conditions. In fact, John Mackie already argued that "it is not part of the definition of an INUS condition that it should not be necessary" (Mackie 1965: 246). Thus, necessary causes are INUS conditions which - in addition - are also necessary conditions.
To put all this more succinctly: If you want to do antiquated causal data analysis in line with ideas shown to have been hopelessly inadequate about 200 years ago already, you can still analyze plain necessary conditions with QCA. But if you want to carry out causal data analysis in line with the scientific state-of-the-art in configurational causal inference, you need to search for INUS structures.
All the best
Alrik
References
- Mackie, John L. 1965. "Causes and Conditions." American Philosophical Quarterly 2 (4):245-64.
- Mackie, John L. 1973. The Cement of the Universe: A Study of Causation. Oxford: Oxford University Press.
- Ragin, Charles C., and Sarah Ilene Strand. 2008. "Using Qualitative Comparative Analysis to Study Causal Order: Comment on Caren and Panofsky (2005)." Sociological Methods & Research 36 (4):431-41.
Dear Alrik,
Thank you very much for the elaborated answer covering different historical and state-of-the-art perspectives. I appreciate this mixture.
Best wishes, Philip
- Badges
- Science topic
- Similar topics
- Psychology
- Mental Processes
- Learning