Abstract
Qualitative case-study research follows a purposeful (rather than random) sampling strategy. Just what the ‘purpose’ behind sampling is, i.e. how cases get selected by authors practising qualitative research is not well understood, however. We focus on an extreme scenario, where authors select only one single case in 300 papers published in Organization Studies, Organization Science, Academy of Management, Administrative Science Quarterly, and Strategic Management Journal between 1999 and 2019. We find 102 papers that feature several sub-units of analysis within the case (e.g. multiple teams in the same organization). Unfortunately, few authors motivate their selection strategies on the case-level, let alone on the sub-unit level. Authors that do relay their selection strategies on both levels use different criteria. On the case level, criteria such as selection on constructs, context, and convenience of data-access provide for ‘in-depth’ insight, whereas the sub-units are selected to allow for a comparative logic based on differences or similarity between sub-units, to propose a process, or to undertake variance-based theorizing. Such comparisons, by design, are not possible at the case level. In light of these practices, we propose a 'two-level selection’ framework with the aim of promoting pluralistic case selection strategies allowing scholars to leverage the single case study research even more rigorously and transparently for theoretical advancement and discovery.
