Early alert systems (EAS) are an important technological tool to help manage and improve student retention. Data spanning 16,091 students over 156 weeks was collected from a regionally based university in Australia to explore various microeconometric approaches that establish links between EAS and student retention outcomes. Controlling for numerous confounding variables, significant relationships between the EAS and student retention were identified. Capturing dynamic relationships between the explanatory variables and the hazard of discontinuing provides new insight into understanding student retention factors. We concluded that survival models are the best methods of understanding student retention when temporal data is available.

The main objective of this study is to evaluate the relationship between Early Alert Systems (EAS) and student retention. Specifically, the study aims to: (i) examine the effects of demographic, institutional and learning environment variables on student retention, (ii) examine the effects of EAS on student retention, and (iii) assess the financial implications of the interaction between EAS and student retention. Selected microeconometric models were estimated using data for 16,124 undergraduate students extracted from a case study university. The data was captured over three years between 2011 and the beginning of 2014.

Key findings of this study show that demographic, institution, student performance and workload variables all exhibit statistically significant relationships with retention measures at the case study institution. Furthermore, the EAS had a positive effect on increasing students’ length of enrolment. Females are more likely to discontinue, but are also more likely to complete their course. Aboriginal and Torres Strait Islander (ATSI) students are more likely to be retained than non-ATSI students. Institutional factors such as the type of course, the school a student enrols in, or mode of enrolment all affect student’s retention rate. Variables capturing student performance and workload further affect retention. Periods of inactivity during students’ enrolment was one of the strongest factors affecting measures of student retention. The study also finds that demographic, institution, learning environment and EAS variables are subject to significant temporal effects. Using weekly observations, temporal effects were captured up to 156 weeks (3 years) of student enrolment, yielding a total of 1,119,170 observations. Using survival modelling, the study provides an unprecedented degree of accuracy in estimating the relationship between explanatory variables and the hazard of discontinuing over time.

Finally, the financial implications of the EAS was evaluated using treatment effects modelling. On average, students identified by the EAS for targeted support remained enrolled for an extra 14 weeks than students not identified by the EAS. The additional revenue in tuition fees caused by EAS identification is estimated to be $4,004 per student. It is concluded that early alert systems have significant financial benefits, initiating support services that positively impact on student outcomes.