Construction Risk Measurement and Driving Mechanisms in Old Residential Community Renovation Projects: A Combined Correlation–SEM Approach

Old residential community renovation projects face intense socio-technical risks driven by complex construction environments, diverse stakeholders, and management frictions. To optimize risk governance, this study develops a systematic risk measurement framework by integrating correlation coefficient analysis and structural equation modeling (SEM). Based on multi-source empirical data, five critical risk dimensions are identified: personnel, materials and equipment, technology, management, and environment. Following indicator screening via correlation metrics, an SEM path analysis is performed using risk probability, consequence severity, control difficulty, and exposure degree as endogenous variables. The results demonstrate that the model achieves robust statistical goodness-of-fit, effectively uncovering a tripartite driving mechanism encompassing risk triggering, amplification, and constraint across varying dimensions. This study delivers a replicable quantitative tool and an actionable decision-making basis for the refined safety management and sustainable implementation of complex urban renewal projects.