Reducing Procrastination on Programming Assignments via Optional Early Feedback

TL;DR

The paper investigates whether optional early feedback deadlines can reduce procrastination on large upper-year CS programming assignments. Using a quasi-experimental design in a fourth-year AI course, the intervention increased the rate of early starts but did not by itself raise overall grades; however, among students who used early deadlines, grades were higher, and the analysis indicates that the added automated feedback is the key driver of performance gains. Interviews suggest perceived benefits in academic performance, mental health, and skill development, balanced against concerns about competing deadlines and perceived lack of graded value. The study highlights the potential and limits of non-graded early feedback to support self-regulated learning, with implications for designing scalable interventions in CS education.

Abstract

Academic procrastination is prevalent among undergraduate computer science students. Many studies have linked procrastination to poor academic performance and well-being. Procrastination is especially detrimental for advanced students when facing large, complex programming assignments in upper-year courses. We designed an intervention to combat academic procrastination on such programming assignments. The intervention consisted of early deadlines that were not worth marks but provided additional automated feedback if students submitted their work early. We evaluated the intervention by comparing the behaviour and performance of students between a control group and an intervention group. Our results showed that the intervention encouraged significantly more students to start the assignments early. Although there was no significant difference in students' grades between the control and intervention groups, students within the intervention group who used the intervention achieved significantly higher grades than those who did not. Our results implied that starting early alone did not improve students' grades. However, starting early and receiving additional feedback enhanced the students' grades relative to those of the rest of the students. We also conducted semi-structured interviews to gain an understanding of students' perceptions of the intervention. The interviews revealed that students benefited from the intervention in numerous ways, including improved academic performance, mental health, and development of soft skills. Students adopted the intervention to get more feedback, satisfy their curiosity, or use their available time. The main reasons for not adopting the intervention include having other competing deadlines, the intervention not being worth any marks, and feeling confident about their work.