The impact of individual differences in developing computational thinking and sensor data analytics skills in construction engineering education

The construction industry is a hazardous environment with a high prevalence of work-related musculoskeletal disorders, compromising workers’ physical and emotional well-being. Construction practitioners can leverage sensor-based safety assessment systems to track and identify workers’ awkward postures, preventing potential injuries. Educational sensor data practices with block programming can enable higher-order learning of the required computational skills for sensor data analytics. However, limited research exists on the factors influencing the acquisition of these skills in training graduating construction students. Through a sensor-based risk assessment intervention, this study explores how individual characteristics (demographics) influence students’ learning. Assessments included perceived self-efficacy of data analytics skills, analytical performance scores, and user acceptance of the educational platform. The results suggest: (a) women show higher self-efficacy gains, while Hispanic/Latino students and those without construction or programming experience report lesser gains, (b) students reach similar performance levels, but those with construction experience excel in reflection reports, and (c) students without construction experience perceive higher utility and lower risks, while Hispanic/Latino students show greater future intent to use the pedagogical tool. The findings contribute to Aptitude-Treatment Interaction Theory by highlighting how individual differences can impact the efficacy of pedagogical interventions in acquiring technical skills in construction education.