Application of Learning Analytics for the Study of the Virtual Campus Activity in an Undergraduate Fluid Mechanics’ Course in Bachelors of Engineering

Abstract Learning Analytics tools are employed to assess student engagement with the Virtual Campus in an undergraduate Fluid Mechanics course at university level in Spain. This is aimed at obtaining a diagnosis of the course problematics which include low attendance rates, poor performance on activity tests and exams and a high number of re-enrolments. A total amount of 67,902 records from the online platform are analyzed to understand activity trends and information distribution. Specific indicators like percentage gap are defined and correlated with evaluation data for deeper insights. Different profiles of students are identified as a function of the interaction level with the platform. Significant correlation has been found between the level of interaction and the academic results of the students. While proactive profiles exhibit remarkable performance rates, no low-activity profiles were capable of passing the course. The continuous evaluation was found neither correlated with the interaction nor with the final exam marks, so its redefinition is suggested as future work. It is also shown that the higher the average number of student enrolments, the lower the participation and performance in the course. No bias was observed with respect to the gender variable. Moreover, the daily analysis of the records has discovered overall trends about the students’ habits. Learning Analytics is demonstrated as a useful tool for the specified purpose. It is also proposed to evaluate the effectiveness of further actions for the improvement of the course in future academic years.