Abstract
This study investigates the effects of Information and Communications Technology (ICT)-supported learning models and students’ behavioral activity on academic performance in professionally oriented secondary education. Using a quasi-experimental pre-test/post-test design, three instructional models were compared: traditional instruction (Control Group (CG)), blended learning with digital tools Inquiry-Based Learning (IBL), and an ICT-supported production-task–based model Digital Learning Tasks (DLT). The study involved 156 students in grades 9–10 over a 16-week intervention. Academic achievement, critical thinking, and professional self-efficacy served as outcome variables, while online and offline behavioral indicators (learning management system activity, classroom engagement, and production-task reports) were analyzed as explanatory factors. The results show that both ICT-supported models significantly outperformed traditional instruction in academic achievement (Analysis of Variance (ANOVA), p < 0.001), critical thinking, and self-efficacy (p < 0.001). Regression analyses revealed that classroom engagement and the quality of production-task reports were the strongest predictors of academic performance (β = 0.56–0.71, p < 0.01), while baseline knowledge was significant only in the traditional model. The novelty of this study lies in reconceptualizing professionally oriented schooling as a behaviorally mediated learning process and empirically demonstrating how production-based tasks and behavioral indicators explain the effectiveness of ICT-supported instruction.