This research investigates students' perceptions of AI-enhanced pedagogy in higher education, focusing on dimensions of teaching quality and skill acquisition in Kazakhstan. A large-scale quantitative design was employed to gather data from 1107 students across various universities, surpassing the minimum sample size of 1058 determined by a priori power analysis, thereby ensuring sufficient statistical power. The scholars used Cronbach's alpha and the Rasch Rating Scale Model to make sure the measurements were reliable and valid. The scholars then used Structural Equation Modeling to test the proposed relationships. The structural model showed a good fit to the data, with comparative and absolute fit indices (CFI and TLI > 0.90; RMSEA and SRMR < 0.08) meeting the required levels. The findings suggest that usability, accessibility, trust in AI-generated learning materials, and AI-assisted skill development significantly influence students' perceptions of teaching quality and overall satisfaction. In contrast, algorithm-driven personalization and perceived improvements in human and diagnostic skills exert no notable impact. These results show that students care more about AI applications that are clear, dependable, and make sense from a teaching perspective than about features that automatically personalize things. The study enhances understanding of AI-enabled learning as a socio-technical process by presenting empirical evidence from a previously underexplored Central Asian context and provides context-specific insights for the development of inclusive, effective, and skill-oriented AI-supported pedagogical practices in higher education.
Potluri et al. (Thu,) studied this question.