Selecting Cases for in-Depth Study from a Survey Dataset: An Application of Ragin's Configurational Methods

Abstract

While ‘establishing the phenomena’, to use Merton's phrase, is an important part of the sociological enterprise, in then accounting for such empirical regularities, theoretical models are required to understand causal processes. Both regression analysis and configurational methods applied to large datasets can establish patterns of relationships. Following a realist view, we assume that causal mechanisms have generated such patterns, and sound theoretical models are required to understand them. In-depth case studies can contribute to advancing such causal knowledge. We describe how, in the particular context of the configurational mode of analysis that characterises Ragin's Qualitative Comparative Analysis (QCA), we have selected individuals for in-depth study with the eventual purpose of advancing causal or explanatory understanding of conjunctural empirical regularities concerning educational careers. While forms of regression analysis seek to establish the net effects of ‘independent’ variables, QCA, employing Boolean algebra, analyses the conjunctions of conditions sufficient and/or necessary for an outcome to occur. QCA aims to preserve, holistically, the features of cases and is therefore well-suited to case selection. We use QCA both to undertake an initial large scale cross-case analysis and to subsequently select cases to develop theoretical understanding via within-case analysis. Using QCA's measures of consistency with relations of sufficiency and necessity, we can classify cases as typical and deviant, with these two types of cases playing different roles in testing and developing theory. Drawing on analyses of the German SOEP dataset undertaken as part of a larger study which is applying case-based configurational methods to English and German survey datasets while undertaking subsequent in-depth interviews with selected cases, we demonstrate how QCA can be used to select cases for interview in a systematic and theoretically informed manner.

Using Rational Action Theory and Bourdieu’s Habitus Theory Together to Account for Educational Decision-making in England and Germany

“…Thus, we were able to select cases which could be expected to prove useful in exploring potential generative mechanisms. Typical cases can help confirm supposed mechanisms, and deviant cases can suggest additional factors involved in producing the outcome (for more details see Glaesser & Cooper, 2011b;. For this paper, we draw on all those interviews with young people whose parents are either both members of the service class or are both working class.…”

Pathways from marine protected area design and management to ecological success

“…While beyond the scope of the current study, it would be possible to delve more deeply into context at MPAs that have been flagged as having potentially anomalous performance given their NEOLI configuration and develop hypotheses about which additional conditions, if empirically present, may or may not lead to outcomes of interest. Individual cases identified as logical remainders or contradictions in QCA can provide rich insights to help interpret Boolean solutions and can complement statistical analyses even in large- n studies ( Glaesser & Cooper, 2011 ) by helping advance theoretical insights on determinants of MPA success. They can also provide direction for future sampling strategies—for example, it would seem to make sense for set theoretic analyses to have more sites that combined ecological isolation with various combinations of the other NEOLI conditions (recall Fig.…”

A qualitative comparative analysis of well-managed school sanitation in Bangladesh

“…Schools were selected purposively based on Save the Children monitoring data, rather than randomly, to ensure variation of the outcome between cases, as suggested in QCA literature [ 28 , 29 ]. All schools were located in Meherpur Sadar sub-district, all within about a one hour drive from each other in a similar geographical setting, with student populations between 72 and 287 per shift.…”