Optimising Pedagogy in the Digital Age: A Comparative Empirical Study of AI-Driven and Traditional Teaching and Evaluation Methods

This study compares perceptions of artificial intelligence (AI)-driven and traditional teaching and assessment practices across six educational domains: learning and engagement, accuracy and curriculum alignment, scoring and grading, timeliness of feedback, practicality, and time and workload management. A mixed-methods design combined a 19-item, five-point Likert-scale survey of educators and learners with qualitative interviews. The reported mean scores generally favoured AI-assisted approaches for personalisation, grading speed, scalability, and workload reduction, whereas traditional methods received higher ratings for motivation, engagement, and the identification of misconceptions. Of the 19 items, 12 had higher numerical means for AI-driven methods, although the direction of some items requires careful interpretation. Participants also identified limited access to devices and unreliable electricity as practical obstacles to adopting AI tools in the North West Region of Cameroon. The findings suggest that a context-sensitive hybrid approach may combine the operational advantages of AI with teachers’ relational, motivational, and pedagogical expertise. Inferential conclusions cannot be established from the available summary data alone.