This paper investigates the problem of analyzing student academic performance in an e-learning course using multiple regression. E-learning continues its development in the face of continuously emerging digital technologies. LMS Moodle and e-courses allow to automatically collect data about students’ learning actions and results. It helps to optimize the educational process, the main aspect of which is student academic performance as an indicator of the quality and effectiveness of learning. In order to be able to improve the academic performance of students it is necessary to deeply analyze the factors that affect the final grades of students to the greatest or least extent. Thanks to the functionality of LMS Moodle, teachers can easily obtain information about the number of attended lectures, completed practical and independent assignments, posts in the forum and chat of the e-course. These were the factors we selected as the key criteria for evaluating academic performance. Using the multiple regression, we were able to determine the relationship between the student’s final grade and the factors that can influence it and to what extent. The results of the conducted research, in which students of four groups participated, showed that the number of completed assignments has the greatest impact on the final grade and performance of students. Next in terms of the strength of the impact we noted the attendance at lectures and participation in chat rooms and forums of the e-course. The obtained data allowed us to identify the actions that can be applied by teachers in order to improve student academic performance such as stimulating students to actively perform practical and independent assignments and participate in discussions.