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

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 

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

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 

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

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

Fang-Ying Yang

Graduate Institute of Science Education, National Taiwan Normal University, Taipei, Taiwan // fangyang@ntnu.edu.tw

Yuan-Li Liu 

Graduate Institute of Science Education, National Taiwan Normal University, Taipei, Taiwan // denyeyetooth@gmail.com 

Shih-Chieh Chien

Graduate Institute of Science Education, National Taiwan Normal University, Taipei, Taiwan // yoman023591@gmail.com

Yi-Wen Hung

The Affiliated Senior High School of National Taiwan Normal University, Taipei, Taiwan // v3256bear@gmail.com 

ABSTRACT: 

In this study, an interactive science learning app on the topic of plate tectonics was developed for tablets to promote argumentative reasoning. The app guided learners through learning stages that required them to propose arguments, identify relevant evidence, acquire background knowledge, and engage in argumentative reasoning in different scenarios. Visual attention during learning was examined using the Tobii Classes 2.0 eye tracking system and analyzed in relation to reasoning performance. Thirty undergraduates (21 female and 9 male) aged 20 to 23 participated in the study. Argumentative reasoning performance was assessed through responses to prompting questions, while visual attention was measured using fixation-based eye movement measures in areas of interest corresponding to different knowledge representations. Descriptive analyses were conducted to illustrate students’ argumentative reasoning performance and visual attention distributions, while correlation and regression analyses were performed to explore associations between visual attention and reasoning performance. The results indicated that students’ ability to use evidence improved over time, with attention shifting from textual to graphical information following exposure to supporting information provided by the learning app. Higher visual attention to data-related information was linked to better reasoning performance.

Keywords: 

Argumentative reasoning, Interactive learning, Mobile-based learning, Visual attention, Eye tracking

Shu-Hsuan Chang

Department of Finance National Changhua University of Education, Taiwan // shc@cc.ncue.edu.tw 

I-Cheng Lin

Department of Industrial Education and Technology, National Changhua University of Education, Changhua County, Taiwan // icliniclin@cc.ncue.edu.tw

Po-Jen Kuo

Department of Finance National Changhua University of Education, Taiwan // pc7938@icloud.com

Chia-Chung Kuo

Department of Mathematics, National Kaohsiung Normal University of Education, Kaohsiung County, Taiwan // abc0919303857@gmail.com

Tsung-Han Tsai

Department of Finance National Changhua University of Education, Taiwan // kyokofukada70324@gmail.com

Pin-Chien Liu

Department of Finance National Changhua University of Education, Taiwan // d0731009@gm.ncue.edu.tw

Yan-Ling Hsu

Department of Industrial Education and Technology, National Changhua University of Education, Changhua County, Taiwan // s612101@stu.chsc.tw

Pei-Ling Chien

Engineering, Kyushu University // chien.peiling@gmail.com

ABSTRACT: 

Even though many experimental studies have considered nurturing creativity as an essential advantage in implementing STE(A)M (Science, Technology, Engineering, [Arts], and Mathematics) education, there is currently a lack of meta-analysis research on the effect of STE(A)M education on creativity to confirm that STE(A)M education can improve creativity. Accordingly, the present study used a meta-analysis to examine the effect of STE(A)M education on students’ creativity, analyzing 29 effect sizes from 29 eligible international journal articles published between 2013 and 2023. Furthermore, the study explored multiple moderator variables and their impacts on creativity, including educational stage, school location, subject domains, core subject, arts domain, weekly teaching hours, information technology support, and problem/project-based learning. The research results showed that the overall mean weighted effect size was 1.11, indicating that STE(A)M education had a large positive effect on improving students’ creativity. Moreover, subject domains, core subject, weekly teaching hours, and information technology support influenced the mean effect size. The study recommended future research and practices on STE(A)M education to enhance students’ creativity.

Keywords: 

STEM education, STEAM education, Creativity, Meta-analysis, Interventions

Ana Beatriz-Afonso

NOVA Information Management School (NOVA IMS), Universidade NOVA de Lisboa, Campus de Campolide, 1070–312 Lisboa, Portugal // aafonso@novaims.unl.pt

Frederico Cruz-Jesus

NOVA Information Management School (NOVA IMS), Universidade NOVA de Lisboa, Campus de Campolide, 1070–312 Lisboa, Portugal // fjesus@novaims.unl.pt

Catarina Nunes

NOVA Information Management School (NOVA IMS), Universidade NOVA de Lisboa, Campus de Campolide, 1070–312 Lisboa, Portugal // cnunes@novaims.unl.pt

Mauro Castelli

NOVA Information Management School (NOVA IMS), Universidade NOVA de Lisboa, Campus de Campolide, 1070–312 Lisboa, Portugal // mcastelli@novaims.unl.pt

Tiago Oliveira

NOVA Information Management School (NOVA IMS), Universidade NOVA de Lisboa, Campus de Campolide, 1070–312 Lisboa, Portugal // toliveira@novaims.unl.pt

Luísa Canto e Castro

Faculdade de Ciências, Centro de Estatística e Aplicações e Fundação Francisco Manuel dos Santos, Universidade de Lisboa, Lisbon, Portugal // lloura@ffms.pt

ABSTRACT: 

Education is crucial for individual and societal growth. However, it was significantly impacted by the COVID-19 pandemic, with long-lasting effects. Estimates suggest that students’ learning decreased by up to 50% compared to a typical year, though the full impact remains unclear. This paper evaluates primary AA drivers to guide efforts addressing pandemic-related educational inequities. Using government data from virtually all public high school students in a European country, we applied advanced data science methods— Multiple Linear Regression, Decision Trees, Neural Networks, Support Vector Machines, Random Forest, and Extreme Gradient Boosting—to analyze AA determinants before and during the pandemic (2019 and 2020, respectively). Our data includes the most well-known potential AA drivers across four dimensions: students, parents, schools, and teachers. Our substantive findings highlight that student age and legal guardian education were key AA drivers, while Internet access and gender gained importance during the pandemic. Additional drivers, including school size, family nationality, and socioeconomic factors (such as the rate of students receiving school support), also emerged as relevant, particularly under pandemic conditions. This study quantitatively assesses these AA determinants across two distinct academic years, providing nuanced insights into the impact of COVID-19 on education. These results offer valuable guidance for policymakers to implement interventions addressing evolving needs and disparities exacerbated by remote learning. This study contributes to AA literature by utilizing extensive data and machine learning models to reveal enduring and emerging factors affecting educational outcomes during challenging times.

Keywords: 

Education, Academic achievement, Data science, COVID-19

Zahra Banitalebi

Shahid Beheshti University, Tehran, Iran // z_banitalebi@sbu.ac.ir

Masoomeh Estaji

Allameh Tabataba’i University, Tehran, Iran // mestaji74@gmail.com

Gavin T. L. Brown

University of Auckland, New Zealand // gt.brown@auckland.ac.nz

ABSTRACT: 

The significance of teacher’s assessment literacy (AL) was originally captured by the 1990 standards for teacher’s competence in educational assessment. Competence in assessment has changed with the widespread use of recent technology advancements in educational assessment. Consequently, new measures are needed to measure Teacher Assessment Literacy in Digital Environments (TALiDE). A new multidimensional, bifactor, scenario-based self-report inventory was developed with expert review (i.e., face and content validation) and confirmatory factor analysis based on responses from teachers who use digital assessments. The TALiDE questionnaire identifies five roles that summarize teacher competence domains (i.e., Judge, Communicator, Moral Agent, Knower, and Technologist) and the five aspects of each role (i.e., attitude, agency, resources, training needs, and actual practices). Means across the five roles were similar (range 3.46-3.63) but attitudes were notably more positive than actual practice (3.84 vs. 3.17, respectively). The TALiDE scale can help reveal teacher professional development requirements when being asked to use digital assessments educationally. It also makes clear the complex roles and facets of teacher assessment literacy that need to be addressed in teacher education and professional development.

Keywords: 

Bifactor analysis, Digital assessment, Scale validation, Scenario-based questionnaire, Teacher assessment literacy

Yu-Ju Lan

Department of Chinese as a Second Language, National Taiwan Normal University, Taiwan // yujulan@gmail.com

Kai-Yu Tang

Graduate Institute of Library & Information Science, National Chung Hsing University, Taiwan // Innovation and Development Center of Sustainable Agriculture, National Chung Hsing University, Taiwan // kytang@dragon.nchu.edu.tw

ABSTRACT: 

Research into the use of virtual reality (VR) for educational purposes is increasing, but most of the existing literature focuses on the positive effects of VR on users, with its possible negative effects remaining underexplored. An inadequate understanding of the negative effects can put users at risk. To address this research gap, the main purpose of this study is to characterize the negative effects of VR use by conducting a systematic review of the literature published in the last decade (2010-2023). Web of Science and Scopus were searched using two sets of keywords. Following PRISMA, 20 studies were analyzed and identified five main categories of negative effects of VR in the research taxonomy of the coding scheme: physical, psychological, cognitive, behavioral symptoms, and cybercrime. The number of publications reporting the negative effects of VR was largest in 2020, followed by 2023, 2015, and 2016. Data were collected mostly using subjective methods, including questionnaire surveys and interviews. Based on the results, a series of frameworks are proposed for future research: VR addiction, reality–virtuality harmony, and VR navigation frameworks. Implications for educators, parents, and future researchers are provided.

Keywords: 

Virtual reality, Negative symptoms, VR addiction, Reality–virtuality harmony, VR navigation

Chin-Yuan Lai

Center for General Education, National Taichung University of Science and Technology, Taiwan // yuanlai@nutc.edu.tw

Li-Ju Chen

The Sub-groups of Technology Domain, Kaohsiung Regional Advisory Group of Compulsory Education, Taiwan // ljchen.liz@gmail.com

ABSTRACT: 

Web-based multimedia annotation is a valuable tool for engaging learners with diverse materials. This study aimed to assess the effects of multimedia annotation on student performance, self-regulation, and cognitive load in an online learning environment. We developed and implemented a multimedia annotation system in an online biology class using a quasi-experimental design. The study included 64 twelfth-grade students from two classes in northern Taiwan. One class, consisting of 31 students, used the web-based multimedia annotation system (the experimental group), while the other class, with 33 students, used traditional paper-and-pencil annotations (the control group). The results showed that web-based annotation tools did not have a significant impact on students’ performance or cognitive load. However, they significantly improved students’ self-regulation skills, including self-observation, self-judgment, and self-response. Based on these findings, we recommend that future research ensure students have ample time to become familiar with the system, which can help reduce external cognitive load and enhance learning effectiveness. Additionally, these tools enabled teachers to better assess students’ understanding and provide targeted feedback. Web-based annotation can be a valuable resource for teachers to identify and address students’ learning difficulties in online environments. The study suggests several recommendations for integrating web-based multimedia annotation tools into instructional practices.

Keywords: 

Online annotation, Self-regulation, Cognitive load

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

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 

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 

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)

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 

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)

Juliana Fosua Gyasi

Faculty of Education, Beijing Normal University, Beijing, China // jufosua.econnect@gmail.com

Lanqin Zheng

Faculty of Education, Beijing Normal University, Beijing, China // bnuzhenglq@bnu.edu.cn

Stephen Frank Love

Faculty of Education, Beijing Normal University, Beijing, China // stephenfranklove1379@gmail.com

Francis Ohene Boateng

Department of Mathematics Education, Akenten Appiah-Menka University of Skills Training and Entrepreneurial Development, Ghana // foboateng@aamusted.edu.gh

ABSTRACT: 

Online collaborative learning has the potential to help learners of all cultures and languages in the artificial intelligence (AI) age. However, studies on the use of human–AI collaboration to promote online collaborative learning are lacking. This study attempts to fill this gap by examining the effects of three approaches to human‒AI collaboration on online collaborative learning performance, namely, the AI feedforward and feedback (AIFF) approach, the AI partner (AIP) approach, and a combined AI feedforward and feedback and AI partner (AIFFP) approach. The study was conducted at a public university in Ghana. A total of 100 college students participated in the study for two months. Both quantitative and qualitative methods were employed to analyze the data. The findings imply that the AIFFP approach had more substantial impacts on collaborative knowledge building, group performance, cognitive engagement, and socially shared regulation (SSR) behaviors than the AIFF approach, the AIP approach, and the conventional online collaborative learning (COCL) approach. The findings and their practical and technical implications are discussed in depth. This research suggests that educators and institutions should use human‒AI collaboration to increase performance and efficiency.

Keywords: 

Online collaborative learning, Human–AI collaboration, Generative artificial intelligence, Knowledge building, Group performance

Dilek Doğan

Ankara University, Ankara, Türkiye // dilek.dogan@ankara.edu.tr

Ömer Demir

Hakkari University, Hakkari, Türkiye // omerdemir@hakkari.edu.tr

Murat Çınar

Türkiye Ministry of National Education, Adana, Türkiye // murat_cinar@rocketmail.com

Hakan Tüzün

Hacettepe University, Ankara, Türkiye // htuzun@hacettepe.edu.tr

Michael K. Thomas

University of Illinois Chicago, Chicago, IL, USA // micthom@uic.edu

ABSTRACT: 

3D Multi-User Virtual Environments (MUVEs) have increasingly become more practical with faster Internet connections, the high processing capacity of ICT devices, and the readiness of learners. However, the educational potential of these immersive worlds in academic settings is closely linked to the combination of appropriate pedagogical and technical design elements. 3D MUVEs that only offer a more immersive experience for their users run the risk of becoming purely performative platforms if they fail to offer appropriately enhanced pedagogical approaches. We, therefore, invite the academic community to focus more on the pedagogical approaches of 3D MUVEs. In this context, this collection of papers aims to outline pedagogical approaches and implementations based on immersive user experiences in 3D MUVEs. It contributes to this field of education with three distinguished papers ranging from theoretical frameworks of user acceptance to training practices for skill development. The findings and suggestions in these papers will provide valuable insights for the academic community and for practitioners willing to benefit from the affordances of 3D MUVEs for learning purposes.

Keywords: 

Virtual reality, MUVEs, 3D environments, Blended learning, Metaverse

Feng Zhang

Faculty of Artificial Intelligence in Education, Central China Normal University, Wuhan, China // zhangfeng77@mails.ccnu.edu.cn

Gege Li

Faculty of Artificial Intelligence in Education, Central China Normal University, Wuhan, China // ligg323@mails.ccnu.edu.cn 

Heng Luo

Faculty of Artificial Intelligence in Education, Central China Normal University, Wuhan, China // luoheng@mail.ccnu.edu.cn

ABSTRACT: 

With the development of virtual reality technology, 3D multi-user virtual environments (MUVEs) have attracted increasing research attention and are thought to bring many learning benefits in higher education. However, the widespread and sustained application of MUVEs in higher education lies in learners’ intention to use them, but the mechanism that affects this intention remains unclear. Based on this research gap, this study established and tested a structural equation model (SEM) to explore the sophisticated interwoven relationships between students’ technology acceptance, learning engagement, learning satisfaction, and their intention to use MUVEs. During COVID-19 pandemic, data collection was conducted at a university in central China and a sample of 237 participants were obtained. The core data of SEM includes path coefficients (e.g., standardized regression coefficients), which represent the direct or indirect effects between variables, and fit indices (e.g., CFI, TLI, RMSEA, and SRMR), which assess the degree of alignment between the model and the observed data. The results showed that: (1) Perceived usefulness has a stronger influence on predicting learning engagement than perceived ease of use. (2) Emotional engagement plays a key mediating role between technology use and satisfaction. (3) The direct predictive ability of satisfaction on intention to use is stronger than that of perceived usefulness. (4) The positive effect of the path cognitive engagement→emotional engagement→satisfaction→intention to use is the largest. Based on these findings, this study provides implications for optimizing the design and implementation of MUVEs in higher education from the perspectives of teachers, students, and platform developers.

Keywords: 

Multi-user virtual environments, Technology acceptance, Learning engagement, Satisfaction, Intention to use

Shih-Yeh Chen

Department of Engineering Science, National Cheng Kung University, Taiwan // sychen‑ncku@gs.ncku.edu.tw

Ying-Hsun Lai

Department of Computer Science and Information Engineering, National Taitung University, Taiwan // yhlai@nttu.edu.tw

Pei-Hsuan Lin

Department of Information Management, National Taichung University of Science and Technology, Taiwan // peihsuan@nutc.edu.tw

Yu-Sheng Su

Department of Computer Science and Information Engineering, National Chung Cheng University, Taiwan// Advanced Institute of Manufacturing with High-tech Innovations, National Chung Cheng University, Taiwan // Department of Computer Science and Engineering, National Taiwan Ocean University, Taiwan // ccucssu@gmail.com

ABSTRACT: 

There is an increasing international focus on multi-user virtual reality (VR) training for all emergency medical technicians within the framework of their respective disaster medical preparedness initiatives. While multi-user VR is often deployed in a variety of digital educational settings, it has not been meaningfully explored in emergency medicine. This study was conducted to fill this research gap by developing a multi-user VR training system for the donning and doffing procedures associated with Level C protective gear. Furthermore, the current investigation also considered the determinants influencing whether emergency medical technicians are willing to adopt and accept multi-user VR training courses via conceptualizing the extended unified theory of acceptance and use of technology. An empirical survey was conducted involving 80 emergency medical technicians in the eastern region of Taiwan to evaluate their experiences with the multi-user VR system. The data were analyzed using Partial Least Squares Structural Equation Modeling, and the results showed that performance expectancy, social impact, facilitating conditions, price value, and habit significantly influenced emergency medical technicians’ readiness to adopt multi-user VR training courses. Moreover, performance expectancy, facilitating conditions, and behavioral intention emerged as pivotal determinants of course adoption. However, effort expectancy and hedonic motivation did not exert a significant influence on adoption behavior. The findings of the study have important theoretical and practical implications.

Keywords: 

Multi-user virtual reality, Virtual environment for emergency medical technicians, Extended unified theory of acceptance and use of technology (UTAUT 2), Behavioral intention, Emergency medical preparedness

Yu-Ju Lan

National Taiwan Normal University, Taiwan // yujulan@gmail.com

Scott Grant

Monash University, Australia // Scott.Grant@monash.edu

Hui-Chin Yeh

National Yunlin University of Science & Technology, Taiwan // hyeh@yuntech.edu.tw

ABSTRACT: 

This study investigated the use of virtual chatbots in a 3D multi-user virtual environment (3D MUVE) to enhance the communication skills of Chinese as a foreign language (CFL) learners. Several virtual chat agents, developed using pattern matching techniques and embedded in Second Life, created a blended learning environment in which CFL learners completed authentic language tasks through scaffolded self-directed learning (SDL). The chatbots were able to interact with the learners, answer their questions, and provide guidance as needed. A mixed methods approach was used to collect quantitative and qualitative data from 49 Chinese Studies students at an Australian university. Quantitative data was collected using a one-group post-test only design where learning outcomes were assessed using a Moodle post-test. Qualitative data included log data of students’ activities in Second Life. After a whole-class lesson, students logged into Second Life to complete assigned tasks supported by progress indicators and scaffolding for different activities. Results indicated that CFL learners improved their communication skills and reduced errors caused by their native language. The combination of virtual chat agents and a scaffolded SDL context seemed to effectively support CFL students’ interpersonal communication skills while reducing vocabulary and sentence structure errors.

Keywords: 

Virtual reality, 3D multi-user virtual environments, Virtual chatbots, Chinese as a foreign language, Scaffolded self-directed learning