Integration in a mixed-method case study of construction phenomena: From data to theory

Purpose: This study demonstrates integration within a mixed-methods case study of construction phenomena, whilst ensuring reliability and validity. This is in view of the established philosophical challenges in theory generation, whereby qualitative and quantitative methods are underpinned by divergent, almost incompatible, paradigmic assumptions. Design/methodology/approach: The study uses a sample case study research on the phenomenon of cost overruns, supported by a coherent flow of well-articulated philosophical arguments to idealise the logic of integration. Issues of reliability and validity were resolved along these lines, by incorporating applicable criteria from both the qualitative and qualitative strands. A detailed outline and rationalisation of the stepwise approach to achieving integration are provided, from the point of design conceptualisation, data collection, analysis and further down to theory generation. Findings: The study generated two level-1 theories by collecting numerical data on cost overruns, geotechnical index parameters and textual data on the geotechnical practices. Another level-1 theory was generated in reflexive adaptation to unanticipated social constructs emerging from the qualitative data. All level-1 theories from the quantitative and qualitative strands were triangulated to yield two “level-2 theories”: A log-regression model and a cognitive map. The approach to integration is thus explanatory sequential, and concurrent (at the second stage of transformation in the generation of level-2 theories). Research limitations/implications: The study empirically reinforces that ontological flexibility, achievable through the use of thoughtfully designed integrated mixed-methods case studies, permits the investigation of multidimensional construction phenomena in innovative ways, relevant to provide holistic theoretical and practice-based contributions. Originality/value: The study practically signposts a bespoke stepwise approach to integration, in a mixed-methods case study of construction phenomena, against the contextual backdrop of its relative novelty and lack of studies delving in-depth into the theoretical nitty-gritty.