Modelling Students’ Source Code Evolutions To Provide Personalised Guidance In Introductory Programming Courses

Programming skills are fundamental in cultivating critical thinking, problem-solving, and creativity, which are integral to computational thinking competencies. These competencies are increasingly recognised as prerequisites for success in the future workforce. To fulfil this demand, higher education institutions have integrated technology-driven tools, such as Learning Management Systems (LMS), into programming education. However, despite the wealth of data collected through these systems, there is an underutilisation of this information to enhance students' learning experiences, especially in introductory programming courses, which often exhibit high failure rates and attrition. The challenge lies in the absence of personalised guidance to help students overcome obstacles in the learning process, resulting in students frequently becoming stuck in states of uncertainty.

This research aims to address these challenges by exploring the application of Markov Models to analyse and predict students' problem-solving behaviours through the modelling of their source code evolutions while carrying out formative practical assessment activities. A mixed-method approach coupled with experimental design, is adopted to provide a comprehensive understanding of the complexities surrounding the thought process, decision-making, and problem-solving approaches of students in introductory programming courses.

The intended outcome of the study is a personalised guidance tool based on the Markov models designed to provide students with targeted and high-level guidance throughout their engagement with practical programming tasks. This tool will leverage differences in source code among students to inform its approach and offer recommended changes to students' code structure. The application of Markov Models presents a promising avenue for addressing these challenges and potentially transforming introductory programming courses into more engaging and effective learning experiences with a focus on enhancing the overall introductory programming education.

Keywords: Programming skills, Problem-solving, Learning Management Systems (LMS), Introductory programming courses, Personalised guidance, Markov Models, Source code evolutions, Educational technology.