Proceedings of The 7th International Conference on Advanced Research in Teaching and Education
Evaluation of Student’s Performance Using Item Response Theory and Path Analysis
Rita Azerhouni, Prof. Dr. Amine Amar
Successful educational experience and future success depend on student progress. It is indispensable for the academic institution, students, and other stakeholders to keep records of it and evaluate it, thus it can be used to test the student’s ability to acquire knowledge and skills over time. Students can stay on track and make considerable progress with the support, feedback, and regular assessments they receive. Nevertheless, student progress assessment is a daunting task. It can be influenced by different features such as age, major, learning strategies and socioeconomic background. In addition, the relationship between student progress and aforementioned features is not linear. In this context, learning analytics can provide help to clarify this complicated relationship and to extract information to make good decisions. In this paper, we suggest using Path Analysis (PA) method to model the relationship between the student progress and different features and Item Response Theory (IRT), to measure and analyze students’ abilities and traits, based on their responses to a set of items and questions more accurately than traditional scoring methods. IRT can help to identify which items are more informative or discriminatory. To implement our approach, we have come up with an online survey to gather required data about students from different universities and schools. The size of our sample is good enough to ensure statistical inference. It is worth to note that the quality check stage and other steps were implemented using a suitable software.
keywords: item response theory, learning analytics, path analysis, scoring methods, student progress