The flipped classroom enhances learning but requires strong self-regulated learning (SRL) skills to ensure successful learning outcomes (van Alten et al., 2020). However, many students lack self-regulation in flipped learning (Shyr & Chen, 2018). Additionally, existing research on flipped learning has been predominantly outcome-oriented, focusing mainly on cognitive outcomes while overlooking the SRL process (Jiang et al., 2022).
Gamification enhances motivation and performance in flipped learning and is increasingly being explored to support SRL. However, studies on gamified SRL (e.g., Qiao et al., 2022; Zhao et al., 2024) show mixed results, often relying on self-reported data that fail to objectively capture its dynamic nature. To address this gap, a more robust and data-driven approach is needed to analyze students’ SRL processes.
Education process mining (EPM), a technique used to extract meaningful patterns from event logs recorded by educational systems (Cerezo et al., 2020), has been widely applied in various learning contexts. However, despite its potential, only a limited number of studies have explored its application in SRL research (Cerezo et al., 2020), and, to date, no study has employed EPM to examine SRL behaviors in a gamified flipped learning environment.
Association rule mining, a foundational technique in data mining, identifies relationships between data items within large datasets. This method generates association rules (e.g., X → Y), where the presence of X suggests a likelihood of Y (Zhang & Zhang, 2023). Given its computational efficiency and suitability for real-time data processing, the Apriori algorithm is well-suited for uncovering patterns in students’ SRL behaviors. By applying this technique, this study aims to provide educators with actionable insights into how gamification influences SRL processes, thereby enabling the design of targeted interventions to enhance students’ self-regulation skills. This study seeks to address the following research question:
RQ: How does gamification influence students’ SRL behaviors in a flipped learning environment, as revealed through association rule mining?
This study integrated a flipped learning approach with a gamified SRL framework to enhance students’ active engagement in SRL behaviors. To support SRL, gamification elements such as points, badges, and achievements—aligned with the GAFCC model (Huang & Hew, 2018)—were incorporated to facilitate the forethought, performance, and self-evaluation phases of SRL. Additionally, scaffolding mechanisms, including prompts, task calendars, performance metrics, and a dedicated SRL dashboard, were implemented to track students' SRL and provide personalized recommendations.
A quasi-experimental design was used to examine SRL behaviors in two intact classes randomly assigned to a control group (n = 86) or an experimental group (n = 91). Both used the same SRL platform, but only the experimental group experienced gamification. Trace data were collected over 12 weeks.
Participation was voluntary, with informed consent obtained online. Behavioral data included SRL strategies such as goal setting, reflection, and dashboard checks. The experimental group’s use of the gamified dashboard was also tracked. The Apriori algorithm (support ≥ 0.1, confidence ≥ 0.8) was applied for sequence pattern mining of SRL behaviors.
Some of the results (top 10 rules for each group) are presented below.
The non-gamified group exhibited a linear and task-oriented SRL approach (see Table 2), where students concentrated primarily on completing assigned tasks, such as submitting learning tasks, setting learning goals, and conducting self-evaluations, while engaging in task-related preparatory activities, such as reviewing task requirements and studying course materials.
Specifically, the rules generated in Table 2 reveal a strong association between “View learning tasks” and other task-related behaviors, including “Submit learning tasks,” “Submit learning goals,” “Submit self-evaluation,” and “Study course materials.” This suggests that students who actively monitored tasks (e.g., by viewing the required tasks) were also highly likely to complete task-related activities. These patterns indicate that students in the non-gamified group focused primarily on task fulfillment, emphasizing task monitoring and completion over reflective or adaptive learning strategies. This approach underscores a structured and goal-driven process, with limited evidence of iterative adjustments or deeper engagement in self-regulated learning practices.
Table 1
Apriori results of non-gamified groups’ SRL behaviors
Rule | Confident | Support |
|---|---|---|
{Submit learning tasks} -> {View learning tasks} | 0.8118 | 0.3617 |
{Submit learning goals} -> {View learning tasks} | 0.8313 | 0.2713 |
{Submit learning goals} -> {Study course materials} | 0.8153 | 0.2661 |
{Submit self-evaluation} -> {View learning tasks} | 0.8933 | 0.2634 |
{Submit self-evaluation} -> {Submit learning tasks} | 0.8889 | 0.2621 |
{Study course materials, Submit learning tasks} -> {View learning tasks} | 0.8584 | 0.2621 |
{View learning tasks, Submit self-evaluation} -> {Submit learning tasks} | 0.8955 | 0.2359 |
{Submit learning tasks, Submit self-evaluation} -> {View learning tasks} | 0.9000 | 0.2359 |
{Submit self-evaluation} -> {Submit learning tasks, View learning tasks} | 0.8000 | 0.2359 |
{Submit learning goals, View learning tasks} -> {Study course materials} | 0.8647 | 0.2346 |
The gamified group exhibited a dynamic and integrated SRL approach, facilitated by gamification features that emphasized monitoring, reflection, and iterative adjustments.
First, a strong association was observed between “Check SRL dashboard” and “View gamification progress” (Confidence = 0.9674, Support = 0.5135). This indicates that students frequently utilized gamification tools, such as the SRL dashboard, to monitor their performance and evaluate their SRL competence. Compared to the non-gamified group, these patterns suggest that the gamified environment actively encouraged students to interact with SRL scaffolds, fostering greater self-awareness of their learning processes. Additionally, students who combined “View learning tasks” with “Check SRL dashboard” were highly likely to “View gamification progress” as well, demonstrating that gamification tools were seamlessly integrated into students' learning workflows.
While the gamified group exhibited some task-oriented patterns similar to those of the non-gamified group, the gamified environment extended these patterns through gamification features. For example, students frequently combined “Submit learning tasks” and “View gamification progress” alongside “View learning tasks” (Confidence = 0.836, Support = 0.161). This highlights a multi-dimensional approach, in which task completion was closely tied to task review and progress monitoring via gamification tools.
Table 2
Apriori results of gamified groups’ SRL behaviors
Rule | Confident | Support |
|---|---|---|
{Check SRL dashboard} -> {View gamification progress} | 0.9674 | 0.5135 |
{Submit learning tasks} -> {View learning tasks} | 0.8038 | 0.2359 |
{View learning tasks, Check SRL dashboard} -> {View gamification progress} | 0.9888 | 0.1803 |
{Submit learning tasks, View gamification progress} -> {View learning tasks} | 0.8360 | 0.1610 |
{Submit learning goals} -> {View learning tasks} | 0.8704 | 0.1574 |
{Study course materials, Submit learning tasks} -> {View learning tasks} | 0.8852 | 0.1493 |
{Submit learning goals} -> {Study course materials} | 0.8056 | 0.1457 |
{Submit self-evaluation} -> {Submit learning tasks} | 0.8636 | 0.1452 |
{Submit self-evaluation} -> {View learning tasks} | 0.8515 | 0.1431 |
{Study course materials, Check SRL dashboard} -> {View gamification progress} | 0.9886 | 0.1330 |
This study analyzed SRL behaviors in a flipped, gamified learning environment using process mining techniques. The results indicate that gamification fosters a more dynamic and reflective SRL process, while non-gamified learners adopt a linear, task-focused approach. These findings highlight gamification’s role in enhancing self-regulation and the effectiveness of data-driven approaches in capturing SRL behaviors. The study contributes to research by advancing understanding of SRL in gamified settings and offers practical insights for educators. Future research could explore the long-term impact of gamification on SRL development across diverse learner profiles and subject areas. Additionally, investigating the integration of adaptive gamification strategies tailored to individual SRL needs could further enhance personalized learning experiences.
This work was supported by the Research Grants Council of Hong Kong Research Fellow Scheme (Reference no: RFS2223-7H02).
EdTech Archives