Rasch-Based Validation of Agreement and Usage Instruments for Integrated Project Quality Management Systems in Construction Projects

Integrated Project Quality Management Systems (IPQMS) require robust measurement instruments to assess how project quality processes are recognised and practised by key construction stakeholders. Although quality management systems, project quality management and integrated management systems have been widely discussed in construction research, limited empirical attention has been given to validating instruments that measure IPQMS processes across stakeholder roles. This paper validates agreement and usage instruments for IPQMS processes using the Rasch Measurement Model. The study analysed six datasets representing three stakeholder groups: client representatives, consultants and contractors. A total of 67 IPQMS process items were examined, comprising 20 client representative items, 22 consultant items and 25 contractor items. Each item was measured across two dimensions: agreement and usage. Rasch analysis was conducted using WINSTEPS to evaluate rating scale functioning, reliability and separation indices, item and person fit, unidimensionality, differential item functioning and item hierarchy. The findings demonstrate that all six instruments showed ordered rating scale functioning, acceptable to excellent reliability, meaningful separation indices, productive item fit and essential unidimensionality. Several items with elevated fit statistics were retained due to their theoretical relevance and contribution to the IPQMS construct. The results confirm that the IPQMS agreement and usage instruments are psychometrically sound and suitable for subsequent critical process identification and structural model development. The study contributes to construction quality management research by demonstrating that role-specific IPQMS processes can be empirically validated through Rasch measurement, thereby providing a defensible measurement foundation for future critical process identification and logit-based structural positioning.