In the previous chapter, the problem of internal validity was discussed from an experimental perspective. That is, the case study method was understood as an attempt to satisfy the methodological criteria that define a well-designed experiment. To the extent that a single case, or a small number of cases, exemplify a quasi-experimental design, the case study method is vindicated.

However, few case studies are truly experimental, in the sense of having a manipulated treatment. This is because a manipulable treatment is usually easy to replicate across multiple cases, thus providing a large-N cross-case research design. Moreover, among observational case studies, perfect “natural experiments” are rare. The observational world does not usually provide cases with both temporal variation (making possible “pre” and “post” tests) and spatial variation (“treatment” and “control” cases) across variables of theoretical interest, while holding all else constant. Usually, there are important violations of the ceteris paribus assumption.

What this means is that case study research usually relies heavily on contextual evidence and deductive logic to reconstruct causality within a single case. It is not sufficient simply to examine the covariation of X1 and Y, because there are too many confounding causal factors and because the latter cannot usually be eliminated by the purity of the research design or by clever quantitative techniques (control variables, instrumental variables, matching estimators, and the like). Thus, a “covariational” style of research is usually insufficient to prove causation in a case study format.