In this study, we delve deeper into retention at the level of individual courses. We analyze three courses offered as part of a large online CS graduate degree program taught at a major research university in the United States. While the average course drop rate for the examined courses in the program was 7%, variations in this rate could be predicted by a number of different attributes: for instance, we observed that women, native English speakers, and students with less prior education were all more likely to complete a course that they had started.
Online education offers unparalleled convenience and accessibility, allowing students from diverse backgrounds to pursue higher education from anywhere in the world. However, retention rates in online programs, including Computer Science (CS) programs, have emerged as a significant challenge, with students more likely to withdraw from individual classes or the program as a whole compared to their counterparts in traditional settings. This study aims to delve deeper into the issue of retention at the level of individual courses, examining the factors that contribute to students' persistence in an online CS graduate degree program.
To better understand the determinants of retention in online CS courses, we conducted a comprehensive analysis of three courses offered as part of a large online CS graduate degree program taught at a major research university in the United States. Over 12 semesters, we collected voluntary data from students, including information on gender, prior CS experience, anticipated workload for the course, and other relevant attributes. By analyzing these variables in the context of course retention, we sought to identify patterns and predictors of students' likelihood to persist in their studies.
Our findings revealed that the average course drop rate for the examined courses in the program was 7%, but variations in this rate could be predicted by several different attributes. For instance, we observed that women, native English speakers, and students with less prior education were all more likely to complete a course they had started. These findings highlight the complex interplay of factors that contribute to retention in online CS programs and provide valuable insights for educators, administrators, and policymakers looking to improve student persistence in these settings.
Importantly, our study also demonstrates that many of these predictors can be measured before the class begins, supporting the implementation of early intervention strategies. By identifying students who may be at higher risk of withdrawing from a course or program, educators and institutions can tailor support services, resources, and interventions to address the unique challenges faced by these individuals. This proactive approach not only has the potential to improve retention rates but also to foster a more inclusive and supportive learning environment for all students in online CS programs.