April 2025, Volume 28, Issue 2
Special Issue on "Reinventing pedagogies and practices of 3D MUVEs with the rise of blended learning"
Guest Editor(s): Dilek Doğan, Ömer Demir and Murat Çınar
Download Table of Content in PDF
Download Complete Issue in PDF
Full Length Articles
Fei Gao
Department of Applied Technology, Bowling Green State University, USA // gaof@bgsu.edu
Rong Liu
Department of Mathematics and Statistics, The University of Toledo, USA // rong.liu@utoledo.edu
ABSTRACT:
This study examined the relationship of learner profiles and the satisfaction of basic psychological needs among 109 undergraduate physics students participating in an online algebra training program. The study utilized latent profile analysis to identify groups of learner profiles and explored the role of students’ perceived needs satisfaction in predicting group composition using multinomial logistic regression. The results indicated that not all students benefited equally from the online training program. Four distinct groups of learner profiles emerged from the analysis: high achiever-high engagement, high achiever-medium engagement, low achiever-high engagement, and low achiever-low engagement. There were significant differences in perceived autonomy and competence among these groups, with students’ perceived autonomy and competence predicting the group composition. The study suggests the importance of considering motivational factors such as students’ perceived needs satisfaction when designing tailored online learning experiences.
Keywords:
Online learning, Perceived needs satisfaction, Adaptive learning, Latent profile analysis
Cite as:Gao, F., & Liu, R. (2025). Towards tailored online learning: Predicting learner profiles in an online learning environment with perceived needs satisfaction. Educational Technology & Society, 28(2), 1-14. https://doi.org/10.30191/ETS.202504_28(2).RP01
Submitted January 4, 2024; Revised August 15, 2024; Accepted September 30, 2024; Published December 12, 2024
Stella Xin Yin, Dion Hoe-Lian Goh, Choon Lang Quek and Zhengyuan Liu
Stella Xin Yin
Wee Kim Wee School of Communication and Information, Nanyang Technological University, Singapore // xin013@e.ntu.edu.sg
Dion Hoe-Lian Goh
Wee Kim Wee School of Communication and Information, Nanyang Technological University, Singapore // ashlgoh@ntu.edu.sg
Choon Lang Quek
National Institute of Education, Nanyang Technological University, Singapore // choonlang.quek@nie.edu.sg
Zhengyuan Liu
Institute for Infocomm Research (I2R), A*STAR, Singapore // liu_zhengyuan@i2r.a-star.edu.sg
ABSTRACT:
With the growing popularity of computational thinking (CT) classes in K-12 schools, it is important to investigate public perceptions of these initiatives. Analyzing public discussions and opinions provides valuable insights that can inform future educational policies and reforms. In this paper, we collected questions and answers related to CT education on the Quora platform. Next, we applied a topic modeling approach to find out major topics in online discussions. Through analysis, we identified 6 topics in questions and 14 topics in answers. Our findings revealed that people showed great interests but also uncertainty about CT education learning outcomes. Many people asked for suggestions on CT learning tools and platforms, but they struggled to identify appropriate information to support their learning needs. Among their answers, while people held positive attitudes toward CT education, they were concerned about the difficulties their children faced in the learning process and the problem of educational equity. Moreover, since CT practices cultivate information literacy skills for children in the 21st century, the benefits of CT education might be overestimated. These findings deepen our understanding of CT education, which could inform education policies and future research directions.
Keywords:
Computational Thinking, K-12 education, Public perception, Social media, Topic modeling
Cite as:Yin, S. X., Goh, D. H.-L., Quek, C. L., & Liu, Z. (2025). Mapping the public understanding of computational thinking education: Insights from social Q&A platform discussions. Educational Technology & Society, 28(2), 15-34. https://doi.org/10.30191/ETS.202504_28(2).RP02
Submitted May 18, 2024; Revised July 26, 2024; Accepted November 6, 2024; Published December 15, 2024
Seokmin Kang, Chun-Hao Chang and Sungyeun Kim
Seokmin Kang
The University of Texas Rio Grande Valley, USA // seokmin.kang@utrgv.edu
Chun-Hao Chang
National Taipei University of Nursing and Health Sciences, Taiwan // chunhao@gm.ntunhs.edu.tw
Sungyeun Kim
Incheon National University, South Korea // syk@inu.ac.kr
ABSTRACT:
Virtual tools are rapidly supplementing physical tools in classrooms. Although extensive research has compared the benefits of using virtual and physical manipulatives, few studies have systematically explored the learning benefits of using a physical and a virtual tool. As the major learning goal in science, technology, engineering, and mathematics (STEM) education is to help students better understand scientific and mathematical concepts and processes, this study reviewed empirical studies comparing learning performance between working with physical and virtual manipulatives published between 2000 and 2022 in Education and Psychology journals. The results revealed inconsistent patterns in learning comparison across studies even with similar interventions. The discrepancy was reinterpreted based on the amount of information provided by tools, and cognitive engagement calibrated by verbs from instructional directives in the implementation. It was found that learners showed better learning performance when a tool provided more concept-related information and when they engaged in their learning activities more deeply, not whether a physical or a virtual manipulative was used. The implementation guideline for teachers to consider was discussed when they create lesson plans with physical and virtual manipulatives.
Keywords:
Affordance, Cognitive engagement, Instructional directives, Learning, Physical, Virtual
Cite as:Kang, S., Chang, C.-H., & Kim, S. (2025). Working with physical and virtual manipulatives: A systematic review of affordance and cognitive engagement manifested through implementation. Educational Technology & Society, 28(2), 35-52. https://doi.org/10.30191/ETS.202504_28(2).RP03
Submitted April 26, 2024; Revised July 31, 2024; Accepted October 4, 2024; Published December 17, 2024
Jing Liu
Department of Linguistics and Modern Language Studies, The Education University of Hong Kong, Hong Kong SAR, China // School of International Education, Chengdu Polytechnic, Chengdu, China // s1151965@s.eduhk.hk
Qing Ma
Department of Linguistics and Modern Language Studies, The Education University of Hong Kong, Hong Kong SAR, China // maqing@eduhk.hk
ABSTRACT:
This meta-analysis evaluated the effectiveness of data-driven learning (DDL) among low-proficiency L2 English learners, addressing the mixed results found in previous meta-analyses. The study incorporated 38 studies involving 2085 participants, yielding 37 effect sizes from control-experimental (C/E) studies and 42 from pre- and post-test (P/P) studies. The findings demonstrated that DDL had a medium effect in C/E studies (g = 0.71) and a large effect in P/P studies (g = 1.43). The moderator analyses, based on the corpus-based language pedagogy (CBLP) framework by Ma et al. (2022), examined 7 pedagogical moderators. The results reaffirmed the efficacy of DDL in teaching lexicogrammatical items and suggested DDL’s curriculum flexibility; the duration of DDL did not significantly impact its effectiveness. Unique to this meta-analysis were findings that DDL was more effective for low-proficiency L2 learners of English when employing the following pedagogical strategies that cater to the cognitive-social nature of DDL: (1) utilizing paper-based concordancing to facilitate the pedagogical processing of corpus resources, (2) leveraging learners’ first language (L1) to improve comprehension of concordance meanings, (3) applying interactive communication with teacher verbal guidance or teacher verbal feedback attuned to learner responsiveness, and (4) providing teacher support in collaborative work to reduce the collaborative cognitive load on learners. Finally, this study proposed a holistic approach to CBLP design tailored to low-proficiency L2 learners, which presents an essential frontier for future research.
Keywords:
Data-driven learning (DDL), Corpus-based language pedagogy (CBLP), Low-proficiency L2 English learners, Cognitive load theory, Meta-analysis
Cite as:Liu, J., & Ma, Q. (2025). Examining corpus-based language pedagogy (CBLP) practices in data-driven learning (DDL) for low-proficiency L2 English learners: A meta-analysis. Educational Technology & Society, 28(2), 53-76. https://doi.org/10.30191/ETS.202504_28(2).RP04
Submitted August 1, 2024; Revised September 17, 2024; Accepted November 7, 2024; Published December 17, 2024
Gisu Sanem Öztaş and Gökhan Akçapınar
Gisu Sanem Öztaş
Department of Computer Education and Instructional Technology, Hacettepe University, Türkiye // gisuoztas@hacettepe.edu.tr
Gökhan Akçapınar
Department of Computer Education and Instructional Technology, Hacettepe University, Türkiye // gokhana@hacettepe.edu.tr
ABSTRACT:
This study aimed to develop a prediction model to classify students based on their academic procrastination tendencies, which were measured and classified as low and high using a self-report tool developed based on the students’ assignment submission behaviours logged in the learning management system’s database. The students’ temporal learning traces were used to extract the features used in the prediction models. The study participants were 51 students enrolled in the Database Management Systems course, which was conducted online using the Moodle learning management system. The study compared the performance of different machine learning algorithms in predicting students’ academic procrastination tendencies, analysed the important features of prediction models, and examined whether there is a difference between the academic performance of low and high academic procrastinators. Logistic regression was found to outperform other classification algorithms and reached 90% accuracy in classifying low and high academic procrastinators. Students’ regular and early access to course activities were found to be important features in predicting their academic procrastination tendencies. In terms of academic performance, the findings support the existing literature. Students with low academic procrastination tendencies got significantly higher final grades than those with high academic procrastination tendencies. These findings show that students’ academic procrastination tendencies can be predicted with high accuracy using online learning trajectories. Such a model will be important in the development of intervention methods for preventing academic procrastination.
Keywords:
Learning analytics, Academic procrastination, Online learning, Time management, Prediction
Cite as:Öztaş, G. S., & Akçapınar, G. (2025). Predicting students’ academic procrastination tendencies using online learning trajectories. Educational Technology & Society, 28(2), 77-93. https://doi.org/10.30191/ETS.202504_28(2).RP05
Submitted January 24, 2024; Revised October 22, 2024; Accepted December 5, 2024; Published December 23, 2024
Chih-Yueh Chou
Department of Computer Science and Engineering, Yuan Ze University, TaoYuan, Taiwan, R.O.C.
// cychou@saturn.yzu.edu.tw
Wei-Han Chen
Department of Computer Science and Engineering, Yuan Ze University, TaoYuan, Taiwan, R.O.C. // s1086035@mail.yzu.edu.tw
ABSTRACT:
Studies have shown that students have different help-seeking behavior patterns and tendencies and furthermore, that students with certain help-seeking behavior patterns and tendencies may have poor performance (i.e., at-risk students). This study applied an educational data mining approach, including clustering and classification, to analyze students’ problem-solving and help-seeking data in a computer assisted learning system to identify student help-seeking behavior patterns and tendencies. First, nine observable problem-solving and help-seeking features for identifying help-seeking behavior patterns were established. Second, this study applied the k-means clustering method and identified three well-known help-seeking behavior patterns: executive, avoidant, and instrumental help-seeking. The results further identified two new help-seeking behavior patterns. One was static instrumental help-seeking and the other was static instrumental and executive help-seeking. Third, executive help-seeking and static instrumental and executive help-seeking patterns could be used as at-risk predicators of poor performance. Fourth, the study applied clustered and identified results to build a minimum distance classifier to identify help-seeking behavior patterns in new data. The study also investigated the accuracy of the classifier in early identifying help-seeking behavior patterns from early-stage data. The early identification accuracy was 61% for the first three minutes and 75% for the seven-minutes of early-stage data, respectively. Fifth, this study identified three help-seeking tendencies: independent problem-solvers, executive help-seekers, and static instrumental and executive help-seekers. In summary, the study showed the feasibility and effectiveness of applying an educational data mining approach, including clustering and classification, to build data-driven student models to identify student help-seeking behavior patterns and tendencies.
Keywords:
Help-seeking behaviors and tendencies, Educational data mining, Clustering, Classification, Data-driven student model
Cite as:Chou, C.-Y., & Chen, W.-H. (2025). Identifying student help-seeking behavior patterns and help-seeking tendencies from student problem-solving and help-seeking behavior data: An educational data mining approach. Educational Technology & Society, 28(2), 94-110. https://doi.org/10.30191/ETS.202504_28(2).RP06
Submitted January 16, 2024; Revised November 2, 2024; Accepted November 5, 2024; Published January 19, 2025
Progression of argumentative reasoning and the relation with visual attention in an interactive learning environment
Fang-Ying Yang, Yuan-Li Liu, Shih-Chieh Chien and Yi-Wen Hung
Can STE(A)M education nurture creativity? A meta-analysis
Shu-Hsuan Chang, I-Cheng Lin,Po-Jen Kuo,Chia-Chung Kuo,Tsung-Han Tsai,Pin-Chien Liu,Yan-Ling Hsu and Pei-Ling Chien
Measuring teacher assessment literacy in digital environments: Development and validation of a scenario-based instrument
Zahra Banitalebi, Masoomeh Estaji and Gavin T. L. Brown
Drivers of academic achievement in high school: Assessing the impact of COVID-19 using machine learning techniques
Ana Beatriz-Afonso, Luísa Canto e Castro, Frederico Cruz-Jesus, Catarina Nunes, Mauro Castelli and Tiago Oliveira
Systematic review of risks hidden in VR-based learning environments: Research frameworks and evidence from the literature
Yu-Ju Lan and Kai-Yu Tang
Effects of web-based multimedia annotation on the performance, self-regulation, and cognitive load of students
Chin-Yuan Lai and Li-Ju Chen
Theme-Based Articles
Generative artificial intelligence in education: Theories, technologies, and applications
Shu-Jie Chen, Xiaofen Shan, Ze-Min Liu and Chuang-Qi Chen
Shu-Jie Chen
CountryCollege of Teacher Education, Wenzhou University, Wenzhou, China // csj@wzu.edu.cn
Xiaofen Shan
CountryCollege of Teacher Education, Wenzhou University, Wenzhou, China // xiaofenshan1207@163.com
Ze-Min Liu
Department of Educational Information Technology, East China Normal University, Shanghai, China // iszemin.liu@gmail.com
Chuang-Qi Chen
Institute of Linguistics (IoL), Shanghai International Studies University, Shanghai, China // rico.chen9521@gmail.com
ABSTRACT:
The introduction of programming education in K-12 schools to promote computational thinking has attracted a great deal of attention from scholars and educators. Debugging code is a central skill for students, but is also a considerable challenge when learning to program. Learners at the K-12 level often lack confidence in programming debugging due to a lack of effective learning feedback and programming fundamentals (e.g., correct syntax usage). With the development of technology, large language models (LLMs) provide new opportunities for novice programming debugging training. We proposed a method for incorporating an LLM into programming debugging training, and to test its validity, 80 K-12 students were selected to participate in a quasi-experiment with two groups to test its effectiveness. The results showed that through dialogic interaction with the model, students were able to solve programming problems more effectively and improve their ability to solve problems in real-world applications. Importantly, this dialogic interaction increased students’ confidence in their programming abilities, thus allowing them to maintain motivation for programming learning.
Keywords:
Large language models, Generative artificial intelligence, Debugging skills, Computational thinking, Self-efficacy, Programming education
Cite as:Chen, S.-J., Shan, X., Liu, Z.-M., & Chen, C.-Q. (2025). Employing large language models to enhance K-12 students’ programming debugging skills, computational thinking, and self-efficacy. Educational Technology & Society, 28(2). https://doi.org/10.30191/ETS.202504_28(2).TP01
Published December 15, 2024
Hao-Chiang Koong Lin, Chun-Hsiung Tseng and Nian-Shing Chen
Hao-Chiang Koong Lin
Department of Information and Learning Technology, National University of Tainan, R.O.C. // koonglin@gmail.com
Chun-Hsiung Tseng
Department of Electrical Engineering, YuanZe University, R.O.C. // lendle.tseng.archive@gmail.com
Nian-Shing Chen
Institute for Research Excellence in Learning Sciences and Program of Learning Sciences, National Taiwan Normal University, R.O.C. // nianshing@gmail.com
ABSTRACT:
In recent years, learning programming has been a challenge for both learners and educators. How to enhance student engagement and learning outcomes has been a significant concern for researchers. This study examines the effects of AI-based pedagogical agents on students’ learning experiences in programming courses, focusing on web game development using JavaScript and Phaser. We developed two pedagogical agents: a debugger that provides context-sensitive assistance and a chatbot that offers guidance based on pre-configured Phaser knowledge. The experiment involved 60 sophomore students from a university in southern Taiwan, and they were randomly assigned to control and experimental groups. The study measured changes in students’ self-efficacy (creative, persuasive, and change dimensions), JavaScript proficiency, debugging efficiency, and overall engagement. Results show significant improvements in all self-efficacy dimensions and JavaScript proficiency for the experimental group. Debugging log analysis showed that students who used the pedagogical agents were able to fix bugs more quickly and more effectively. Qualitative analysis of student reflections indicated more positive learning experiences and deeper engagement with learning content in the experimental group. These findings suggest that integrating AI-based pedagogical agents can enhance students’ learning experiences in programming courses.
Keywords:
Pedagogical agent, Learning experience, Self-efficacy, Programming education, Artificial intelligence
Cite as:Koong Lin, H.-C., Tseng, C.-H., & Chen, N.-S. (2025). Enhancing programming education: The impact of AI-based pedagogical agents on student self-efficacy, engagement, and learning outcomes. Educational Technology & Society, 28(2). https://doi.org/10.30191/ETS.202504_28(2).TP02
Published January 16, 2025
Hui-Chun Chu, Chia-Ying Hsu and Chun-Chieh Wang
Hui-Chun Chu
Department of Computer Science and Information on Management, Soochow University, Taiwan // carolhcchu@gmail.com
Chia-Ying Hsu
Department of Computer Science and Information on Management, Soochow University, Taiwan // chia314314@gmail.com
Chun-Chieh Wang
Department of Education, National Pingtung University, Taiwan // jie809@gmail.com
ABSTRACT:
In ancient poetry courses, students are expected to learn literary skills and the ability for interpreting ancient poems, from which they can gain insight into the historical context and appreciate the beauty of the poems. However, due to the gaps between ancient and modern living contexts, students often encounter difficulties in comprehending ancient texts. Scholars have suggested that visualizing learning scenarios could enhance learners’ comprehension and knowledge retention, implying that engaging students in learning by illustrating the context of ancient poems could encourage them to explore the contexts in that historical background and comprehend the contents. Therefore, a generative Artificial Intelligence (GenAI) drawing mode is proposed in the present study. In this learning mode, students describe the characters, events, and scenes in the poems in words, and GenAI creates visual representations to immerse students in the contexts of the ancient poems. An experiment was conducted with two classes of students from a high school. One class was the experimental group consisting of 31 students, who used GenAI drawing mode to assist their learning of ancient poems. Another class was the control group consisting of 29 students, who used the conventional drawing mode. The experimental results showed that the experimental group had better learning achievement than the control group. Moreover, for the students in the experimental group, the higher creative ones showed better self-efficacy than the lower creative ones. Additionally, the drawings by the experimental group had rich contents with more details, including emotions, contexts, objects, and expressions.
Keywords:
GenAI drawing, Learning by drawing, Ancient poetry, Situated learning
Cite as:Chu, H.-C., Hsu, C.-Y., & Wang, C.-C. (2025). Effects of AI-generated drawing on students’ learning achievement and creativity in an ancient poetry course. Educational Technology & Society, 28(2). https://doi.org/10.30191/ETS.202504_28(2).TP03
Published December 19, 2024
Chih-Hung Wu, Ting-Sheng Weng and Chih-Hsing Liu
Chih-Hung Wu
National Taichung University of Education, Taiwan // chwu@mail.ntcu.edu.tw
Ting-Sheng Weng
National Chiayi University, Taiwan // politeweng@mail.ncyu.edu.tw
Chih-Hsing Liu
National Kaohsiung University of Science and Technology, Taiwan // Ming Chuan University, Taiwan // phd20110909@gmail.com
ABSTRACT:
With the growing attention directed towards ChatGPT and its applications in education, this study explored its impact on various variables pertaining to student learning. Specifically, an integrated theoretical framework was used to investigate the factors that influence student problem-solving and critical thinking abilities when using ChatGPT for educational purposes. Data was collected from 687 students by questionnaire survey. The research model was evaluated using both first order and second-order factors through partial least squares structural equation modeling analysis. The results revealed significant relationships and mediating and moderating effects within the proposed conceptual framework. These findings underscore that learning motivation directly affects learner engagement and problem-solving skills. Moreover, learner engagement was found to positively influence problem-solving and critical thinking tendency. Problem-solving skills have also emerged as pivotal and significant factors influencing critical thinking tendencies. The study also confirms the moderating roles of students’ mental attributes, which amplify the relationship between learning motivation, learner engagement, and problem-solving. The partial least squares multigroup analysis method was used to ensure the generalizability of our model. The concluding section discusses both the theoretical and practical implications for educational settings.
Keywords:
ChatGPT in education, Motivation and engagement, Problem-solving, Critical thinking, Cognitive load theory, Artificial Intelligence (AI), Large Language Model (LLM), Higher-order thinking (HOT)
Cite as:Wu, C.-H., Weng, T.-S., & Liu, C.-H. (2025). Exploring ChatGPT’s potential to enhance problem-solving and critical thinking in education. Educational Technology & Society, 28(2). https://doi.org/10.30191/ETS.202504_28(2).TP04
Published December 29, 2024
Lei Tao, Hao Deng and Yanjie Song
Lei Tao
Department of Mathematics and Information Technology, The Education University of Hong Kong, Hong Kong SAR, China // s1147997@s.eduhk.hk
Hao Deng
Department of Mathematics and Information Technology, The Education University of Hong Kong, Hong Kong SAR, China // dengh@eduhk.hk
Yanjie Song
Department of Mathematics and Information Technology, The Education University of Hong Kong, Hong Kong SAR, China // ysong@eduhk.hk
ABSTRACT:
Information and communication technologies have transformed education, driving it towards intelligent teaching and learning. With the rise of generative artificial intelligence (AI), represented by tools such as ChatGPT, there is also a growing body of literature on generative AI in education. In this study, we searched the Scopus, ERIC, and Web of Science databases as well as the proceedings of the International Conference on Artificial Intelligence in Education and the International Conference on Educational Data Mining. These searches yielded 1,158 papers that we subjected to topic modelling analysis, the Mann-Kendall trend test, and keyword analysis to comprehensively review the evolving trends and dynamics of generative AI in education, active publication sources and authors, major research topics in generative AI education, and potential directions for cross-thematic research. We analysed the trends of keywords and relevance during different sub-periods. Topic modelling enabled us to classify all abstracts of the reviewed papers into 12 topics and identify the tendency of each topic. The topics included predicting student performance, learning tutoring systems, generative AI and AI literacy, writing/essay automated grading, chatbot-based learning/assessment, gamified/game-based learning & AI learning environments, emotional engagement, automatic feedback, question generation, generative AI use in peer assessment, personalised recommendation system, and simulation-based study environment. Finally, a heat map and a hierarchy cluster provided information about the correlations between topics and the potential integration of different research directions for generative AI in education, offering a reference for future research.
Keywords:
Bibliometric analysis, Generative AI, Co-occurring network, Topic modelling, Major research topics
Cite as:Tao, L., Deng, H., & Song, Y. (2025). Generative artificial intelligence in education: A topic‑based bibliometric analysis. Educational Technology & Society, 28(2). https://doi.org/10.30191/ETS.202504_28(2).TP05
Published January 20, 2025
Xin Cui, Zhi-Qiang Ma, Xin-Ya You, Yuchen Chen, Jia-Jia Yao and Yun-Fang Tu
Xin Cui
School of Design, Jiangnan University, Wuxi City, China // cuixin960220@163.com
Zhi-Qiang Ma
Jiangsu Research Center of “Internet Plus Education”, Jiangnan University, Wuxi City, China // mzq1213@jiangnan.edu.cn
Xin-Ya You
Jiangsu Research Center of “Internet Plus Education”, Jiangnan University, Wuxi City, China // 1114989283@qq.com
Yuchen Chen
Sydney School of Education and Social Work, The University of Sydney, Sydney, Australia // yuchen.academic@gmail.com
Jia-Jia Yao
Jiangsu Research Center of “Internet Plus Education”, Jiangnan University, Wuxi City, China // yjjyoka@jiangnan.edu.cn
Yun-Fang Tu
Department of Educational Technology, Wenzhou University, Wenzhou, China // sandy0692@gmail.com
ABSTRACT:
Collaborative argumentation allows groups to express, criticize, and integrate arguments to achieve the co-construction of collective knowledge. However, students often face challenges when proposing diversified arguments, gathering evidence, and rebutting others reasonably. Incorporating generative conversational agents (GCAs) into collaborative argumentation has been demonstrated to effectively broaden the perspective of the argument and to stimulate the generation of new ideas. For this study, we designed rhetorical argumentation customized strategies (RACS), dialectical argumentation customized strategies (DACS) for collaborative argumentation, and a mixed-strategy (RACS+DACS), and compared their effects on the quality of argumentation mappings and argumentation discourse patterns. A total of 121 first-year postgraduate students were enrolled: 33 for the control group, 33 for the RACS group, 27 for the DACS group, and 28 for the mixed-strategy group. Results found that: (1) Regarding the quality of argumentation mappings, DACS could help students select high-quality evidence and learn the logical skills from evidence reasoning to claims. In addition, the mixed-strategy could help students search for multiple types of evidence to support their position; (2) Regarding argumentation discourse patterns, in the characteristics of the structural dimension, DACS could help students use evidence to support higher-order claims during argumentation. The mixed-strategy could help students use evidence to rebut others’ arguments in group discourses. However, no significant differences were detected among the four groups in the characteristics of the social dimension.
Keywords:
ChatGPT, Collaborative argumentation, Generative conversational agents (GCAs), Agent customization, Ordered network analysis (ONA)
Cite as:Cui, X., Ma, Z.-Q., You, X.-Y., Chen, Y., Yao, J.-J., & Tu, Y.-F. (2025). Incorporating generative conversational agents into collaborative argumentation: Do different customization strategies matter? Educational Technology & Society, 28(2). https://doi.org/10.30191/ETS.202504_28(2).TP06
Published December 17, 2024
Special Issue Articles
Guest editorial: Reinventing pedagogies and practices of 3D MUVEs with the rise of blended learning
Dilek Doğan, Ömer Demir and Murat Çınar
Starting from Volume 17 Issue 4, all published articles of the journal of Educational Technology & Society are available under Creative Commons CC-BY-ND-NC 3.0 license.