Sirui Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Sirui Chen has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Wenlan Zhang
Faculty of Education, Shaanxi Normal University, Xi’an 710062, China
2 shared publications - Deliang Wang
Faculty of Education, The University of Hong Kong
1 shared publication - Fan Gao
Institute of Higher Education, Hebei Finance University, China
1 shared publication - Jiawei Chen
Faculty of Teacher Education, Chongqing Normal University, Chongqing, China
1 shared publication - Meifen Chen
School of Digital Media, Shenzhen Polytechnic University, Shenzhen 518055, China
1 shared publication - Mingze Sun
School of Educational Technology/Research Center for Knowledge Engineering, Beijing Normal University, China
1 shared publication
Latest Publications
- Open Access
- 22 May 2026
- Published by MDPI AG in Education Sciences
- Vol. 16 (6) , 816
Can Generative AI Feedback Effectively Enhance Learning Outcomes? A Meta-Analysis of 36 Experimental and Quasi-Experimental Studies
Although generative artificial intelligence (GenAI) feedback shows promise for educational applications, its actual impact on learning outcomes and the factors influencing its effectiveness remain unclear. This study conducted a systematic review and meta-analysis to evaluate the effectiveness of GenAI feedback and identify key moderating factors. Following the PRISMA 2020 guidelines, we reviewed 36 experimental and quasi-experimental studies published between 2023 and 2025, yielding 72 effect sizes. The results revealed that GenAI feedback had a moderate positive effect on academic achievement (g = 0.61), with significant moderation by contextual factors. Subgroup analysis revealed that teaching methods significantly moderated the effectiveness of GenAI feedback, with stronger effects observed in learner-centered environments promoting active construction than in teacher-centered, receptive instruction; whereas educational level, disciplines, intervention duration, and GenAI role showed no significant moderation. A three-level random-effects model was employed to account for effect size dependencies, correcting for the underestimation of standard errors typical of conventional two-level models. Outcome dimension analysis showed that GenAI feedback had the strongest impact on cognitive outcomes, with promising but less established benefits for metacognitive development, and modest effects on non-cognitive outcomes. Future research should further clarify the roles of metacognitive and non-cognitive outcomes in GenAI feedback. In practice, GenAI feedback should serve as complementary scaffolding within constructivist pedagogies to support metacognitive development, while teacher emotional support should be preserved to foster students’ non-cognitive development.
- 22 May 2026
- Published by MDPI AG in Education Sciences
- Vol. 16 (6) , 816
Open AccessCan Generative AI Feedback Effectively Enhance Learning Outcomes? A Meta-Analysis of 36 Experimental and Quasi-Experimental Studies
Although generative artificial intelligence (GenAI) feedback shows promise for educational applications, its actual impact on learning outcomes and the factors influencing its effectiveness remain unclear. This study conducted a systematic review and meta-analysis to evaluate the effectiveness of GenAI feedback and identify key moderating factors. Following the PRISMA 2020 guidelines, we reviewed 36 experimental and quasi-experimental studies published between 2023 and 2025, yielding 72 effect sizes. The results revealed that GenAI feedback had a moderate positive effect on academic achievement (g = 0.61), with significant moderation by contextual factors. Subgroup analysis revealed that teaching methods significantly moderated the effectiveness of GenAI feedback, with stronger effects observed in learner-centered environments promoting active construction than in teacher-centered, receptive instruction; whereas educational level, disciplines, intervention duration, and GenAI role showed no significant moderation. A three-level random-effects model was employed to account for effect size dependencies, correcting for the underestimation of standard errors typical of conventional two-level models. Outcome dimension analysis showed that GenAI feedback had the strongest impact on cognitive outcomes, with promising but less established benefits for metacognitive development, and modest effects on non-cognitive outcomes. Future research should further clarify the roles of metacognitive and non-cognitive outcomes in GenAI feedback. In practice, GenAI feedback should serve as complementary scaffolding within constructivist pedagogies to support metacognitive development, while teacher emotional support should be preserved to foster students’ non-cognitive development.
- Top Cited PapersOpen Access
- 19 July 2025
- Published by Springer Nature in Humanities and Social Sciences Communications
- Vol. 12 (1) , 1-11
The impact of project-based learning on EFL learners’ learning motivation and academic performance: an empirical study in a Chinese rural school
- 19 July 2025
- Published by Springer Nature in Humanities and Social Sciences Communications
- Vol. 12 (1) , 1-11
Top Cited PapersOpen AccessThe impact of project-based learning on EFL learners’ learning motivation and academic performance: an empirical study in a Chinese rural school
- Open Access
- 26 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 47-52
A Novel Two-Stage Personalized Learning Path Recommendation Approach for E-learning
In the context of e-learning, learners often struggle to make informed decisions about what and how to learn when they have access to a vast array of learning resources. To address this issue, several approaches have been proposed from different perspectives, aiming to generate personalized learning paths for e-learners. These approaches include learner-based, knowledge-based, and hybrid recommendation methods. Among them, hybrid methods have emerged as a promising solution for personalized learning path recommendations, as they combine the strengths of both learner-based and knowledge-based approaches. However, existing hybrid methods typically employ exhaustive techniques to determine optimal paths by extracting all possible learning paths. This approach can be time-consuming and computationally expensive. To overcome this challenge, we propose a novel two-stage personalized learning path recommendation approach that integrates concept map and an improved genetic algorithm. We conducted computational experiments using diverse simulation datasets to assess the effectiveness of our proposed method. The experimental results indicate that our approach surpasses other competing methods in terms of both performance and robustness.
- 26 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 47-52
Open AccessA Novel Two-Stage Personalized Learning Path Recommendation Approach for E-learning
In the context of e-learning, learners often struggle to make informed decisions about what and how to learn when they have access to a vast array of learning resources. To address this issue, several approaches have been proposed from different perspectives, aiming to generate personalized learning paths for e-learners. These approaches include learner-based, knowledge-based, and hybrid recommendation methods. Among them, hybrid methods have emerged as a promising solution for personalized learning path recommendations, as they combine the strengths of both learner-based and knowledge-based approaches. However, existing hybrid methods typically employ exhaustive techniques to determine optimal paths by extracting all possible learning paths. This approach can be time-consuming and computationally expensive. To overcome this challenge, we propose a novel two-stage personalized learning path recommendation approach that integrates concept map and an improved genetic algorithm. We conducted computational experiments using diverse simulation datasets to assess the effectiveness of our proposed method. The experimental results indicate that our approach surpasses other competing methods in terms of both performance and robustness.