Reducing Procrastination on Programming Assignments via Optional Early Feedback

Alice Gao · 2025

Read Reducing Procrastination on Programming Assignments via Optional Early Feedback by Alice Gao (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-16052.txt.

Introduction

Alice Gao and Victoria Sakhnini test whether free early feedback can break student procrastination on large programming assignments. Delay is common among undergraduate computer science students and is linked to worse grades and worse well-being. The harm grows in upper-year courses, where assignments are large and complex enough to punish a late start.

Their intervention sets early deadlines worth no marks that return extra automated feedback to anyone who submits by them. Comparing a control group with an intervention group, far more intervention students start early. Grades overall do not differ between groups, yet intervention students who use the early route outscore those who skip it. Interviews show gains in performance, mental health, and soft skills, with curiosity and spare time driving uptake. Competing deadlines, zero marks, and confidence explain most non-use.

Catalog record

CREATORAlice Gao
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE40 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2510-16052

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Reducing Procrastination on Programming Assignments via Optional Early Feedback Alice Gao Abstract 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. Keywords time management, procrastination, optional early feedback, programming assignments, self-regulated learning 1 Introduction Introduction Academic procrastination is prevalent among undergraduate computer science (CS) students. Several meta-analyses have documented the negative consequences of procrastination, including poor academic performance [17, 24] and well-being [14, 27, 30]. Procrastination can force students to prioritize achieving good grades with minimal effort over genuine learning, potentially creating challenges for future academic pursuits and professional careers. Upper-year undergraduate CS students are especially prone to the harmful effects of procrastination. Many upper-level CS courses require students to complete large and complex programming assignments. These assignments provide valuable opportunities for students to apply the theoretical concepts learned in lectures. However, anecdotal evidence suggests that many students don’t start working on these assignments until a few days before the deadline. Such procrastination leads to insufficient time to debug and test their code, a lack of opportunity to seek help from course staff, and a spike in physical and mental health issues. One may attribute such procrastination to a failure of time management and self-regulation. While numerous interventions such as extra-credit opportunities, reflective exercises, email reminders, and varying deadline structures have been employed to reduce academic procrastination in computer science courses, their effectiveness has been mixed and often context-dependent. In particular, previous work has either tied early deadlines to grading incentives or provided limited feedback, leaving unanswered questions about the role of flexible, non-graded early feedback in shaping students’ time management and learning outcomes. Moreover, there is a lack of targeted research on procrastination mitigation in advanced undergraduate programming assignments, where the stakes, complexity, and self- regulatory demands are heightened. This study addresses these gaps by investigating whether optional early feedback deadlines, unlinked to grades but offering substantive automated feedback, can motivate timely engagement and improve academic performance among upper-year computer science students [2, 20]. We designed an intervention to combat academic procrastination on programming assignments. We applied the intervention to an upper-level Introduction to Artificial Intelligence course with four substantial programming assignments, each available for three weeks. Drawing inspiration from prior work [7, 26], we designed our intervention to consist of two optional early feedback deadlines. These early deadlines were not worth any marks. However, if students submitted their work before an early deadline, they would receive extra feedback in the format of unit test results. These unit tests evaluated the quality of the submission and helped students identify mistakes in their submissions. The early feedback deadlines can mitigate procrastination in several ways. First, these early deadlines divided each assignment into smaller, manageable chunks, potentially making it easier for students to get started and reducing students’ stress. Second, early feedback was exclusively available during that particular week, motivating students to start their assignments early. Third, detailed test results in early feedback could help students correct mistakes in their programs and improve their grades on assignments. We applied this intervention in the Fall (Sep-Dec) 2021 offering of the course and evaluated the effects of this intervention using the Spring (May-Aug) 2021 offering of the course as a control group. Our work addresses the following research questions. RQ1: Do the early feedback deadlines motivate students to start their assignments early? RQ2: Do the early feedback deadlines improve students’ assignment grades? 2 Related Work In this section, we review the literature relevant to our study, focusing on three key areas: academic procrastination, interventions designed to mitigate procrastination and self-regulated learning theory. These topics provide the foundation for understanding the mechanisms and potential impacts of our proposed early feedback intervention. 2.1 Academic Procrastination According to Steel [28], academic procrastination refers to voluntarily delaying an intended course of study-related action despite expecting to be worse off for the delay. Academic procrastination negatively affects students’ academic performance and well-being [28]. Some negative consequences include lower grades, health issues (e.g. stress and sleep problems), affective consequences (e.g. anxiety and anger), and problems in their private lives (e.g. lack of social networks, negative reactions from others) [13, 16–19, 25]. Unfortunately, procrastination is widespread among undergraduate students, affecting more than 70% of them [22, 25]. Numerous factors cause procrastination, such as a lack of interest in the task [27], the task’s urgency or deadline proximity [27], depression, fatigue, and low self-esteem [10, 27], perfectionism [12], locus of control [5, 11], goal management abilities [15], neuroticism, openness to experiences, agreeableness, extraversion, conscientiousness [28], lack of academic self-efficacy and poor self-regulation skills [18], fear of failure, task aversiveness or laziness, lack of assertiveness, resentment towards control from others, and a desire to take risks [23], and excessive use of social networking sites for interactive information seeking, discussion, and sharing [29]. 2.2 Procrastination Interventions Prior research has studied many creative interventions to combat procrastination in CS courses, including extra credits [1], reflective writing assignments and schedule sheets [20], email reminders [20, 31], and early deadlines with or without feedback [7, 26]. Several interventions did not have significant effects on students’ behaviour. For instance, Allevato and Edwards [1] showed that extra credits had no significant effects on when students started or completed the programming assignments in an introductory CS course. Similarly, Martin et al. [20] demonstrated that reflective writing assignments and schedule sheets did not significantly impact students’ behaviour on programming assignments in a junior-level data structures course. Email reminders led to some positive results. On the one hand, email reminders resulted in a significant increase in the number of early assignment submissions and a significant decrease in the number of late assignment submissions in a junior-level data structures course [20]. On the other hand, email reminders did not encourage early submissions while increasing homework attempts in a CS1 course

Read the complete text