Predicting student final examination result using regression model based on student activities in learning management system

Background: Learning analytics (LA) is the measurement, collection, analysis, and reporting of data about learners and their contexts to understand and optimize learning and the environments in which it occurs. Most teaching and learning (T & L) data was obtained from learning management systems (LMS), such as Moodle platform. However, this data is not utilized for teaching purposes, for example, learning how students’ activities can influence the final exam marks. Methods: Therefore, this study aims to find a correlation and develop a prediction model between students’ activities, independent variables known as LMS factors, against dependence variables, which are students’ final results. Besides, four non-LMS factors (race, sponsorship, admission requirements and final coursework marks) were also included in the research to obtain the best model. The regression analysis, models are used to predict the outcomes by evaluating the accuracies of testing data. Results: The findings reveal that the best model utilizes Simple Linear Regression (SLR) with coursework as an independent variable, resulting in an average error difference (AED) of only 1.8. The remaining experiments produced AED results ranging from 2.74 to 7.58 using Multilinear Regression (MR). Conclusions: in summary, this study provides a significant finding that demonstrates the potential of utilizing LMS activities to predict final marks, enabling lecturers to enhance their students’ results.