Methodological Insights Theory development in qualitative management control: revisiting the roles of triangulation and generalization

Abstract

Purpose This research paper seeks to advance the techniques of “within-paradigm” triangulation and theoretical generalization adopted in qualitative field studies for new theory development in management control. Design/methodology/approach Using the evidence gleaned from three positivist case studies along with some published interpretivist studies, this paper identifies three triangulation approaches and links them with three forms of theoretical generalization in qualitative field studies. Findings While the “literal” approach to triangulation has its place in the extant literature, this study finds that two other approaches, labeled “convergent” and “normative” triangulation, which are relevant in many circumstances to link multiple pieces of evidence in order to build credible explanations. In theory development, while the extant forms of theoretical generalization (constructive and contextual) are useful for identifying new solutions for both practical and theoretical concerns, this paper finds that a third form, namely “transposed” logic, is relevant in identifying new control problems that can benefit by adopting the observed unusual solutions. Originality/value Within triangulation, while the extant literature endorses the role of literal triangulation in obtaining consistent evidence, including how verifying inconsistent responses helps improve the validity of the obtained evidence, this paper advances two new triangulation approaches that can enrich the extant literature. Within theoretical generalization, while the extant literature deems constructive and contextual forms as “rhetoric”, this paper (1) expands their status to “logic” by clarifying their theoretical purpose and (2) introduces one new form, namely “transposed” logic that helps identify a generalizable range of management control problems that can adopt the observed unusual solution.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 28 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.