Evaluating students as surrogates in virtual reality-based construction safety training: A comparative performance analysis

Abstract

High rates of fatal and non-fatal occupational accidents worldwide have made construction safety a global concern. Despite the proven effectiveness of construction safety training in reducing these rates, current safety implementations often fail to enhance construction workers’ safety awareness and knowledge. Recently, virtual reality (VR) has been increasingly employed in safety training to enhance its effectiveness and contribute to occupational safety. While construction workers and engineers are usually the intended target audience for safety training, students are frequently used as surrogates during the development of training tools. However, the appropriateness of this practice has not been thoroughly investigated in the literature. This study utilizes the V-SAFE v.2 training tool in the experiments to investigate this phenomenon by comparing the performances of students, construction workers, and engineers. The paper-based test scores of students, construction workers, and engineers before and after receiving safety training in the virtual environment are analyzed using appropriate statistical analysis approaches, including paired T-test, Wilcoxon signed-rank test, and two-factor mixed ANOVA. Results indicate that V-SAFE v.2 has significantly improved the learning outcomes of all groups. Additionally, the study reveals that students can serve as surrogates for construction workers during the development of virtual safety training tools. However, students’ performances significantly differ from those of engineers, indicating they are not suitable surrogates for this subject group.