Optimising Programming Skill Acquisition Using AI-Enhanced Problem Based Learning and Decision Support Models

Abstract

A significant number of first-year computer science students struggle to acquire programming skills, largely due to challenges such as varying levels of logical complexity, limited capacity for algorithmic thinking, and the absence of tailored feedback. In response to these issues, a mobile learning application was designed, incorporating the Problem-Based Learning (PBL) approach. This application integrated interactive exercises, automated feedback mechanisms, and collaborative elements to facilitate the learning of C# programming. The effectiveness of this intervention was evaluated using a quasi-experimental research design, involving 60 participants divided into control and experimental groups. Findings demonstrated that the experimental group achieved statistically significant gains in programming competence when compared with the control group, with notable improvements observed between pre-test and post-test scores. In addition to the empirical analysis, the study employed the Analytic Hierarchy Process (AHP) to inform instructional strategy selection. Three pedagogical approaches—traditional lecture-based learning, non-technological PBL, and mobile-supported PBL—were compared based on usability, learner engagement, skill enhancement, and scalability. Weights were assigned to each criterion, and AHP analysis revealed that the mobile-supported PBL approach ranked highest, indicating a pedagogical advantage in this context. The outcomes of the study provide valuable insights for educators, researchers, curriculum designers, and academic policymakers aiming to enhance programming instruction methodologies.