The paper evaluates the role of simulation software in improving the practical skills of students of Electrical – Automation and explains the pedagogical mechanism that helps turn "software hours" into "measurable competence". Mixed-methods research with 309 participants (287 students, 22 lecturers) at three campuses (School of Electrical and Electronic Engineering, Hanoi University of Science and Technology; School of Electrical and Electronics, Hanoi University of Industry; University of Industry – Thai Nguyen University). The five core platforms include MATLAB/Simulink, Proteus VSM, Siemens TIA Portal, Beckhoff TwinCAT 3, and Factory I/O. The quantitative block uses a tested scale (Cronbach's α=0. 81–0. 89), EFA (KMO=0. 91; 6 factors; extraction variance 68. 4%), CFA (CFI=0. 94; TLI=0. 92; RMSEA=0. 048) and SEM for direct/indirect impact testing; 5, 000-sample bootstrap and multi-group analysis (MG-SEM) test for durability and differentiation by school year/campus. Topic-coded qualitative blocks (interviews/observations) to collate results. The results show that the direct impact is consistent with the pedagogical function of each software: MATLAB/Simulink → modeling skills (β=0. 42), Proteus → embedded skills (β=0. 37), TIA Portal → PLC skills (β=0. 34), TwinCAT → distributed control (β=0. 28), Factory I/O → system integration (β=0. 26). Three intermediate channels are significant: Active Learning (AL) is the strongest (βᵢndirect=0. 14), followed by interactive learning (0. 12) and project-based teaching – PBL (0. 10). R²=0. 29–0. 47 suggests that simulation is an important lever but still needs support from real labs, mentoring, and appropriate learning organization. MG-SEM confirms measurement invariance (ΔCFI=0. 004; ΔRMSEA=0. 003) and records year 4 over year 3 in TIA→PLC (+0. 08) and Factory I/O→integrated (+0. 07). Qualitative data indicate onboarding for the first 2–3 sessions, capstone attached to industrial situations, and mentoring is the key to amplifying the effect.
Vu Huy Hoa (Tue,) studied this question.