EMBEDDING LEARNING ANALYTICS IN TARGETED RETENTION PROGRAMS TO ENHANCE THE UNIVERSITY EXPERIENCE FOR COMMENCING STUDENTS

S. Kutieleh

Flinders University (AUSTRALIA)
Student attrition from higher education is a problem worldwide. It signifies a personal loss to the individual; it carries a social and economic cost to the community; and it results in a significant financial cost to each university. While differences exist in the way universities have approached this issue, the literature shows the need for a holistic strategy that aims to facilitate and enhance student engagement with their institutions. Flinders University developed and implemented a centrally run program called the Student Success Program (SSP) to contact students who are likely to drop out in the first year of their enrolment. The aim of the SSP is to maximise retention and improve the student experience at Flinders University. The relevant literature and institutional experiences led us to consider all data sources, both static and dynamic, that are readily available for all students in scope Predictions from machine learning models, built using big data, were utilised to target such students. At least three call-out campaigns per semester, which are consistent with the transition timeline of commencing students, were conducted each semester with each running for at least four weeks. About 14270 students were contacted by their peers over a period of three years to elicit information about their level of preparedness for their studies. The findings demonstrate reduced student attrition, improved appreciation and effective utilisation of student support services. Data analysis also shows that the SSP has saved directly a significant number of students and has contributed to the enhancement of the university experience for many more.