Tal Yifat – “Generalization from Case Studies: a Critique of Social Scientific Theory and Practice”
In-depth case studies are widely recognized in the social sciences as useful for the study and theorization of social processes. However, a key challenge to case studies has been the establishment of general theoretical claims. A main line of argument for the generalizability of case studies suggests that they allow for generalization to theoretical propositions, as opposed to statistical generalization from a sample to a population. This claim for generalization to abstract theoretical presuppositions, I argue, glosses over the fact that theoretical generalizations, like statistical ones, in fact generalize from a smaller set of studied cases (either a single case or a sample of cases) to a larger set of theoretically relevant cases. Once this is recognized, some aspects of generalization from case studies emerge as problematic, especially if we assume that the observed world may behave probabilistically. I conclude that case studies are particularly useful for generating and illustrating the application of new theoretical possibilities, but are limited in their ability to establish the validity of theoretical propositions for a set of relevant cases. I will discuss some variations of the argument for theoretical generalization from case studies, including multiple case study, comparative case study and even some types of sample-based generalization.
Monday Nov 30th, Room SS302, 12-1
*Please arrive at 11.45 for light refreshments*