Transforming static self-instructional materials (SIMs) into AI-enabled learning environments to enhance learner achievement, engagement, and reduce dropout risk in open and distance learning (ODL)

Open and Distance Learning (ODL) has become an important approach to widening access to higher education while supporting Malaysia’s aspirations for flexible and lifelong learning under the Malaysian Qualifications Framework 2.0 (MQF 2.0). Despite its widespread adoption, conventional Self-Instructional Materials (SIMs) remain largely static, providing limited support for personalised learning, learner analytics, and timely instructional intervention. This study presents the design, development, and evaluation of an AI-enabled SIM prototype that integrates adaptive content sequencing, embedded learning analytics, and personalised feedback to enhance learning experiences in ODL environments. The prototype was developed using a Design Science Research (DSR) approach combined with Instructional Design-Based Development (IDD) and evaluated through a pilot study involving 40 postgraduate ODL course learners. The findings indicate that learners using the AI-enabled SIM achieved higher post-test scores than those using conventional SIMs (78.5% vs. 70.8%; Cohen’s d = 0.65), with greater normalised learning gains (48.2% vs. 32.2%), an 18% increase in learning engagement, and an overall satisfaction score of 4.23 out of 5. In addition, the embedded early-warning model predicted dropout risk with 82% accuracy (AUC = 0.81), enabling timely instructor intervention that eliminated actual dropout among learners in the treatment group. These findings demonstrate that integrating adaptive learning, learning analytics, and AI-enabled feedback into conventional SIMs can substantially enhance learner engagement, academic performance, and retention while supporting the flexible learning aspirations of MQF 2.0.