Germane Cognitive Load and Programming Self-Efficacy as Mediators of Teacher Effects on Student Engagement in Vocational Programming Education

This study integrates social support theory, cognitive load theory, and social cognitive theory to address a persistent pedagogical challenge: bridging the gap between programming’s high abstraction and logical rigor and vocational students’ often weaker foundational skills, which typically leads to low engagement. Using structural equation modeling and multi-group analysis of survey data from 310 secondary vocational programming students, our findings revealed that perceived teacher support significantly and positively predicted programming learning engagement ( β  = 0.105, p  = 0.005); computer programming self-efficacy served as a significant single mediator ( β  = 0.085); germane load and self-efficacy sequentially mediated the effect of teacher support ( β  = 0.042); gender moderation effects were significant, with direct and serial mediation effects pronounced in the male group, whereas for female students, the influence operated indirectly through reduced extraneous load, with germane load demonstrating a significant negative effect ( β = −0.135, p  = 0.026). Taken together, these findings clarify the core mechanism of “environmental support — cognitive processing — motivational beliefs — learning behaviors”. The study provides an evidence-based framework for designing gender-differentiated instructional strategies, advancing both pedagogical effectiveness and equity in vocational STEM education.