Exploring the impact of emergency risk information on construction workers' emergency information behavior: insights from confirmatory and exploratory PLS-SEM analyses.

Drawing on risk perception and information adoption theories, this study develops a structural equation model to examine the factors influencing emergency information behavior (EIB). The model includes risk information characteristics (RIC) and information dissemination channels (IDC) as independent variables, risk perception (RP) and information self-efficacy (ISE) as mediating variables, and EIB as the dependent variable. Data were collected from a questionnaire survey of 569 construction workers in Hangzhou and analyzed using Partial Least Squares SEM (PLS-SEM). The results indicate that RIC and IDC have a significant positive influence on RP, ISE, and EIB. Risk perception directly promotes EIB and indirectly influences it by enhancing information self-efficacy. Information self-efficacy, in turn, is also a significant driver of EIB, with the complete model explaining 55.5% of its variance (R 2 = 0.555). This study concludes that delivering high-quality risk information through multiple channels effectively enhances workers' ability to retrieve and use information during emergencies. This is achieved by improving their risk perception and information self-efficacy. The findings provide valuable empirical evidence and strategic recommendations for emergency management in the construction industry.