Computers and Education Open最新文献_第4页

Adaptive instructional designs in blended learning to enhance student engagement and self-regulation 混合式学习中的适应性教学设计，提高学生的参与度和自我调节能力

IF 5.7 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers and Education Open

Pub Date : 2025-10-25 DOI: 10.1016/j.caeo.2025.100299

Jinsong Zou , Songyu Jiang

Blended-learning courses in Chinese higher education often suffer from student classroom-avoidance behaviours and weak self-regulation. To address this problem, the present study integrates the Technology Acceptance Model (TAM), the Community of Inquiry (CoI) framework, and Self-Regulated Learning (SRL) theory into a single structural model that explains how technology acceptance and teaching practice shape learning engagement. Beyond theory testing, the study also aims to curb avoidance behaviours by creating a closed-loop mechanism that links formative-assessment reform, the cultivation of integrated regulation competences, and subsequent behavioural transformation.

Survey data from 714 undergraduates across 25 Chongqing institutions were analysed with structural equation modelling. Perceived ease of use (PEU) showed a strong, significant relation with teaching presence (TP) (β = 0.466), whereas perceived usefulness (PU) did not. TP, operationalised via assessment for/as/of learning practices, directly enhanced learning presence (LP) (β = 0.333) and boosted blended-learning motivation (BLM) (β = 0.466). BLM exerted the largest direct effect on LP (β = 0.414) and fully mediated the PEU → LP and TP → LP pathways. These findings confirm motivation as the pivotal conduit through which user-friendly technology and formative teaching presence foster deeper learning engagement and reduce avoidance behaviours.

By validating a unified TAM–CoI–SRL model on a large, socio-economically diverse sample, the study advances blended-learning theory and offers practitioners evidence that enhancing usability and assessment-driven teaching presence is essential for improving learning presence and mitigating disengagement in digitally mediated classrooms.

我国高等教育中混合式学习课程存在学生回避课堂行为和自我调节能力弱的问题。为了解决这一问题，本研究将技术接受模型（TAM）、探究共同体（CoI）框架和自我调节学习（SRL）理论整合为一个单一的结构模型，解释技术接受和教学实践如何影响学习参与。除了理论测试之外，该研究还旨在通过创建一个将形成性评估改革、综合监管能力培养和随后的行为转变联系起来的闭环机制来遏制回避行为。采用结构方程模型对重庆市25所高校714名大学生的调查数据进行分析。感知易用性（PEU）与教学存在感（TP）表现出强烈的显著关系（β = 0.466），而感知有用性（PU）则没有显著关系。通过对学习实践的评估来实施TP，直接提高了学习存在感（LP）（β = 0.333）和混合式学习动机（BLM）（β = 0.466）。BLM对LP的直接影响最大（β = 0.414），完全介导PEU→LP和TP→LP通路。这些发现证实了动机是关键的渠道，通过它，用户友好的技术和形成性的教学存在促进了更深层次的学习参与，减少了回避行为。通过在一个大的、社会经济多样化的样本上验证统一的TAM-CoI-SRL模型，该研究推进了混合学习理论，并为实践者提供了证据，证明增强可用性和评估驱动的教学存在对于改善学习存在和减轻数字媒介课堂的脱离感至关重要。

{"title":"Adaptive instructional designs in blended learning to enhance student engagement and self-regulation","authors":"Jinsong Zou , Songyu Jiang","doi":"10.1016/j.caeo.2025.100299","DOIUrl":"10.1016/j.caeo.2025.100299","url":null,"abstract":"<div><div>Blended-learning courses in Chinese higher education often suffer from student classroom-avoidance behaviours and weak self-regulation. To address this problem, the present study integrates the Technology Acceptance Model (TAM), the Community of Inquiry (CoI) framework, and Self-Regulated Learning (SRL) theory into a single structural model that explains how technology acceptance and teaching practice shape learning engagement. Beyond theory testing, the study also aims to curb avoidance behaviours by creating a closed-loop mechanism that links formative-assessment reform, the cultivation of integrated regulation competences, and subsequent behavioural transformation.</div><div>Survey data from 714 undergraduates across 25 Chongqing institutions were analysed with structural equation modelling. Perceived ease of use (PEU) showed a strong, significant relation with teaching presence (TP) (<em>β</em> = 0.466), whereas perceived usefulness (PU) did not. TP, operationalised via assessment for/as/of learning practices, directly enhanced learning presence (LP) (<em>β</em> = 0.333) and boosted blended-learning motivation (BLM) (<em>β</em> = 0.466). BLM exerted the largest direct effect on LP (<em>β</em> = 0.414) and fully mediated the PEU → LP and TP → LP pathways. These findings confirm motivation as the pivotal conduit through which user-friendly technology and formative teaching presence foster deeper learning engagement and reduce avoidance behaviours.</div><div>By validating a unified TAM–CoI–SRL model on a large, socio-economically diverse sample, the study advances blended-learning theory and offers practitioners evidence that enhancing usability and assessment-driven teaching presence is essential for improving learning presence and mitigating disengagement in digitally mediated classrooms.</div></div>","PeriodicalId":100322,"journal":{"name":"Computers and Education Open","volume":"9 ","pages":"Article 100299"},"PeriodicalIF":5.7,"publicationDate":"2025-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145415733","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Competencies for teaching with and about artificial intelligence in the natural sciences — DiKoLAN AI 自然科学中的人工智能教学能力- DiKoLAN AI

IF 5.7 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers and Education Open

Pub Date : 2025-10-24 DOI: 10.1016/j.caeo.2025.100303

Johannes Huwer , Christoph Thyssen , Sebastian Becker-Genschow , Lena von Kotzebue , Alexander Finger , Erik Kremser , Sandra Berber , Mathea Brückner , Nikolai Maurer , Till Bruckermann , Monique Meier , Lars-Jochen Thoms

The rapid advancement and widespread adoption of digital technologies have transformed the education sector. Among these developments, the emergence of generative artificial intelligence (AI) tools such as ChatGPT has had a considerable impact on teaching and learning practices. While the integration of AI into educational settings is becoming increasingly common, subject-specific analyses, especially in STEM education, are still lacking. This paper examines the specific challenges and potential of AI in the context of STEM education. It does so by exploring how AI has transformed scientific disciplines and how these changes impact teaching and learning. It highlights the necessity for educators to acquire specific competencies to effectively incorporate AI into their instructional practices. Building on existing frameworks such as DigCompEdu and the subject-specific DiKoLAN, the paper proposes an AI-focused framework: DiKoLAN AI. This framework aligns AI-related teacher competencies with instructional practice in science education. It also provides a structure for categorizing existing teacher training programs. The paper outlines the development of the DiKoLAN AI framework and its content consensus validation by a total of 64 experts through three iterative cycles. Its practical application is demonstrated through 20 case studies from different authors, which offer a practical approach for supporting teacher training and curriculum design in AI-integrated STEM education. The paper concludes with a discussion of opportunities, challenges and future research needs for teacher professionalization.

数字技术的迅速发展和广泛采用改变了教育部门。在这些发展中，ChatGPT等生成式人工智能（AI）工具的出现对教学和学习实践产生了相当大的影响。虽然人工智能与教育环境的整合变得越来越普遍，但特定学科的分析，特别是在STEM教育中，仍然缺乏。本文探讨了人工智能在STEM教育背景下的具体挑战和潜力。它通过探索人工智能如何改变科学学科以及这些变化如何影响教学来实现这一目标。它强调了教育工作者获得特定能力以有效地将人工智能纳入其教学实践的必要性。在现有框架（如DigCompEdu和特定学科的DiKoLAN）的基础上，本文提出了一个以人工智能为重点的框架：DiKoLAN AI。该框架将与人工智能相关的教师能力与科学教育的教学实践相结合。它还提供了一个对现有教师培训项目进行分类的结构。本文概述了DiKoLAN AI框架的开发及其内容共识验证，共有64位专家通过三个迭代周期进行验证。本文通过来自不同作者的20个案例研究展示了其实际应用，为支持人工智能集成STEM教育中的教师培训和课程设计提供了实用方法。文章最后对教师专业化的机遇、挑战和未来的研究需求进行了讨论。

{"title":"Competencies for teaching with and about artificial intelligence in the natural sciences — DiKoLAN AI","authors":"Johannes Huwer , Christoph Thyssen , Sebastian Becker-Genschow , Lena von Kotzebue , Alexander Finger , Erik Kremser , Sandra Berber , Mathea Brückner , Nikolai Maurer , Till Bruckermann , Monique Meier , Lars-Jochen Thoms","doi":"10.1016/j.caeo.2025.100303","DOIUrl":"10.1016/j.caeo.2025.100303","url":null,"abstract":"<div><div>The rapid advancement and widespread adoption of digital technologies have transformed the education sector. Among these developments, the emergence of generative artificial intelligence (AI) tools such as ChatGPT has had a considerable impact on teaching and learning practices. While the integration of AI into educational settings is becoming increasingly common, subject-specific analyses, especially in STEM education, are still lacking. This paper examines the specific challenges and potential of AI in the context of STEM education. It does so by exploring how AI has transformed scientific disciplines and how these changes impact teaching and learning. It highlights the necessity for educators to acquire specific competencies to effectively incorporate AI into their instructional practices. Building on existing frameworks such as DigCompEdu and the subject-specific DiKoLAN, the paper proposes an AI-focused framework: DiKoLAN AI. This framework aligns AI-related teacher competencies with instructional practice in science education. It also provides a structure for categorizing existing teacher training programs. The paper outlines the development of the DiKoLAN AI framework and its content consensus validation by a total of 64 experts through three iterative cycles. Its practical application is demonstrated through 20 case studies from different authors, which offer a practical approach for supporting teacher training and curriculum design in AI-integrated STEM education. The paper concludes with a discussion of opportunities, challenges and future research needs for teacher professionalization.</div></div>","PeriodicalId":100322,"journal":{"name":"Computers and Education Open","volume":"9 ","pages":"Article 100303"},"PeriodicalIF":5.7,"publicationDate":"2025-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145415734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Exploring STEM educators’ perspectives on the integration of AI-enabled technologies in teaching and learning 探讨STEM教育工作者对人工智能技术在教学和学习中的整合的观点

IF 5.7 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers and Education Open

Pub Date : 2025-10-24 DOI: 10.1016/j.caeo.2025.100304

Hulya Avci , Stephanie J. Lunn , Zahra Hazari

The innovation and proliferation of artificial intelligence (AI) technologies are transforming Science, Technology, Engineering, and Mathematics (STEM) disciplines and pedagogical practices, making it essential to prepare educators for this evolving landscape. This qualitative study explores STEM educators’ perceptions of how they developed proficiency with AI-enabled technologies, what motivated their adoption, and the challenges they faced in classroom integration. Using a cross-sectional survey design, we collected open-ended responses from 27 in-service secondary STEM educators in the United States and analyzed the data through the reflexive thematic analysis. Our findings revealed that STEM educators recognized AI as a cognitive scaffold that can support student understanding of complex content and a socio-affective tool that can help enhance teacher–student relationships. The participants highlighted that by streamlining routine tasks, AI allowed them to reclaim time for meaningful student interactions and responsive instruction. The participants also reported developing their AI proficiency through self-directed learning, peer support, and professional networks, driven by motivations to enhance instructional efficiency, support personalized learning, and prepare students for AI-driven futures. However, the participants emphasized several significant challenges, including lack of institutional support, the overwhelming abundance of AI tools, concerns about affordability, and students’ growing overreliance on AI systems. These insights highlight the need for structured, discipline-specific professional development for in-service educators to promote effective and human-centered integration of AI in STEM education. This study contributes to a growing body of research by centering STEM educators’ perspectives and offering novel insights into how to support their effective integration of AI technologies into STEM teaching and learning.

人工智能（AI）技术的创新和扩散正在改变科学、技术、工程和数学（STEM）学科和教学实践，因此教育工作者必须为这一不断变化的环境做好准备。这项定性研究探讨了STEM教育工作者对他们如何熟练使用人工智能技术的看法，他们采用人工智能技术的动机，以及他们在课堂整合中面临的挑战。采用横断面调查设计，我们收集了美国27名在职中学STEM教育工作者的开放式回答，并通过反身性主题分析对数据进行了分析。我们的研究结果表明，STEM教育工作者认为人工智能是一种认知支架，可以帮助学生理解复杂的内容，也是一种社会情感工具，可以帮助加强师生关系。参与者强调，通过简化日常任务，人工智能使他们能够腾出时间进行有意义的学生互动和响应式教学。参与者还报告说，他们通过自主学习、同伴支持和专业网络来提高人工智能水平，其动机是提高教学效率，支持个性化学习，并为学生为人工智能驱动的未来做好准备。然而，与会者强调了几个重大挑战，包括缺乏机构支持、人工智能工具的压倒性丰富、对负担能力的担忧以及学生对人工智能系统的日益过度依赖。这些见解强调了在职教育工作者需要结构化的、特定学科的专业发展，以促进人工智能在STEM教育中的有效和以人为本的整合。本研究通过集中STEM教育工作者的观点，并就如何支持他们有效地将人工智能技术整合到STEM教学和学习中提供新颖的见解，为越来越多的研究做出了贡献。

{"title":"Exploring STEM educators’ perspectives on the integration of AI-enabled technologies in teaching and learning","authors":"Hulya Avci , Stephanie J. Lunn , Zahra Hazari","doi":"10.1016/j.caeo.2025.100304","DOIUrl":"10.1016/j.caeo.2025.100304","url":null,"abstract":"<div><div>The innovation and proliferation of artificial intelligence (AI) technologies are transforming Science, Technology, Engineering, and Mathematics (STEM) disciplines and pedagogical practices, making it essential to prepare educators for this evolving landscape. This qualitative study explores STEM educators’ perceptions of how they developed proficiency with AI-enabled technologies, what motivated their adoption, and the challenges they faced in classroom integration. Using a cross-sectional survey design, we collected open-ended responses from 27 in-service secondary STEM educators in the United States and analyzed the data through the reflexive thematic analysis. Our findings revealed that STEM educators recognized AI as a cognitive scaffold that can support student understanding of complex content and a socio-affective tool that can help enhance teacher–student relationships. The participants highlighted that by streamlining routine tasks, AI allowed them to reclaim time for meaningful student interactions and responsive instruction. The participants also reported developing their AI proficiency through self-directed learning, peer support, and professional networks, driven by motivations to enhance instructional efficiency, support personalized learning, and prepare students for AI-driven futures. However, the participants emphasized several significant challenges, including lack of institutional support, the overwhelming abundance of AI tools, concerns about affordability, and students’ growing overreliance on AI systems. These insights highlight the need for structured, discipline-specific professional development for in-service educators to promote effective and human-centered integration of AI in STEM education. This study contributes to a growing body of research by centering STEM educators’ perspectives and offering novel insights into how to support their effective integration of AI technologies into STEM teaching and learning.</div></div>","PeriodicalId":100322,"journal":{"name":"Computers and Education Open","volume":"9 ","pages":"Article 100304"},"PeriodicalIF":5.7,"publicationDate":"2025-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145415735","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Grammar and engagement in focus: Evaluating Gemini AI’s impact on an educational environment 语法和专注：评估双子座人工智能对教育环境的影响

IF 5.7 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers and Education Open

Pub Date : 2025-10-18 DOI: 10.1016/j.caeo.2025.100302

Miew Luan Ng , Behnam Behforouz , Ali Al Ghaithi

The present paper aims to investigate the role of Gemini AI in improving learners' grammatical knowledge and measuring their level of engagement within the learning process using this AI tool. So, 75 Omani pre-intermediate EFL learners were equally divided into three groups, with 25 students in each group, including a control group (traditional in-class training), an experimental A group (traditional in-class and Gemini AI within the classroom), and an experimental B group (traditional in-class training and Gemini AI outside of the classroom). To collect the data, researcher-made grammar tests were designed, piloted, validated, and their reliability was established. Additionally, to measure the engagement level of students, a questionnaire was adapted from [1]. The study's findings on grammar tests showed the progress of all groups from pretest to posttest. Initially, experimental group A performed significantly better than the other two groups, and experimental group B outperformed the control group. The analysis of the delayed posttest showed consistent results in which the experimental group A significantly performed better than the other two groups. Although the other groups did not show significant enhancements, experimental group B outperformed the control group. The thorough analysis of the engagement questionnaire also revealed similar results in which the experimental group A's engagement level was significantly higher than that of the other groups, followed by better performance of experimental group B, and then the control group. The study's findings are helpful for teachers, students, and institutions.

本文旨在研究双子座人工智能在提高学习者语法知识方面的作用，并使用该人工智能工具测量他们在学习过程中的参与程度。因此，75名阿曼中级前英语学习者被平均分为三组，每组25名学生，包括对照组（传统的课堂培训），实验a组（传统的课堂培训和课堂内的Gemini AI）和实验B组（传统的课堂培训和课堂外的Gemini AI）。为了收集数据，研究者设计了自制的语法测试，进行了试点、验证，并建立了它们的可靠性。此外，为了衡量学生的参与水平，我们采用了一份来自[1]的调查问卷。该研究的语法测试结果显示了所有小组从测试前到测试后的进步。最初，实验组A的表现明显优于其他两组，实验组B的表现优于对照组。延迟后测分析结果一致，实验组A明显优于其他两组。虽然其他组没有明显的改善，但实验B组的表现优于对照组。对敬业度问卷的深入分析也发现了类似的结果，实验组A的敬业度水平显著高于其他组，其次是实验组B，最后是对照组。这项研究的发现对教师、学生和机构都有帮助。

{"title":"Grammar and engagement in focus: Evaluating Gemini AI’s impact on an educational environment","authors":"Miew Luan Ng , Behnam Behforouz , Ali Al Ghaithi","doi":"10.1016/j.caeo.2025.100302","DOIUrl":"10.1016/j.caeo.2025.100302","url":null,"abstract":"<div><div>The present paper aims to investigate the role of Gemini AI in improving learners' grammatical knowledge and measuring their level of engagement within the learning process using this AI tool. So, 75 Omani pre-intermediate EFL learners were equally divided into three groups, with 25 students in each group, including a control group (traditional in-class training), an experimental A group (traditional in-class and Gemini AI within the classroom), and an experimental B group (traditional in-class training and Gemini AI outside of the classroom). To collect the data, researcher-made grammar tests were designed, piloted, validated, and their reliability was established. Additionally, to measure the engagement level of students, a questionnaire was adapted from [<span><span>1</span></span>]. The study's findings on grammar tests showed the progress of all groups from pretest to posttest. Initially, experimental group A performed significantly better than the other two groups, and experimental group B outperformed the control group. The analysis of the delayed posttest showed consistent results in which the experimental group A significantly performed better than the other two groups. Although the other groups did not show significant enhancements, experimental group B outperformed the control group. The thorough analysis of the engagement questionnaire also revealed similar results in which the experimental group A's engagement level was significantly higher than that of the other groups, followed by better performance of experimental group B, and then the control group. The study's findings are helpful for teachers, students, and institutions.</div></div>","PeriodicalId":100322,"journal":{"name":"Computers and Education Open","volume":"9 ","pages":"Article 100302"},"PeriodicalIF":5.7,"publicationDate":"2025-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145361067","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Patterns of mathematics problem solving and synthetic facial expressions in a personal instructing agent 解决数学问题的模式和合成面部表情的个人指导代理

IF 5.7 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers and Education Open

Pub Date : 2025-10-17 DOI: 10.1016/j.caeo.2025.100301

John Lorenz Dela Cruz , Paulyn Joy Dela Cruz , Joyce Antonette Guadalupe , Jiabianca Macaraeg , Piolo Jose Montesa , Mark Paul Ramos , Rex P. Bringula , Kaoru Sumi

This study explored the patterns of mathematics problem-solving and synthetic facial expressions (SFEs) exhibited by a personal instructing agent named PIA. Toward this goal, 81 Grade 8 students participated in a three-day experiment where they were randomly assigned either to the facial (FG) or non-facial (NFG) group. The students’ interactions within the PIA were collected and stored as log files. The attributes extracted from the log files included types of mathematics problems solved (i.e., schema), status of the mathematics problems solved, difficulty levels of mathematics problems solved, and SFEs exhibited by the PIA. Lag sequential analysis (LSA) disclosed that there were similarities and differences in the sequence of math problem-solving behaviors among students. The Apriori algorithm revealed that struggling students tend to solve problems successfully, irrespective of their sex; however, struggling female students tend to solve more problems successfully than their male counterparts. Nonetheless, regardless of their levels of math competency and the version of software used, all students solved problems they were comfortable with and always started with easier problems, gradually progressing. Limitations and future research were also discussed.

本研究探讨了一个名为PIA的个人指导代理在数学问题解决和合成面部表情（sfe）方面的表现模式。为了实现这一目标，81名八年级学生参加了为期三天的实验，他们被随机分配到面部组（FG）和非面部组（NFG）。学生在PIA中的交互被收集并存储为日志文件。从日志文件中提取的属性包括解决的数学问题的类型（即模式）、解决的数学问题的状态、解决的数学问题的难度级别以及PIA显示的sfe。滞后序列分析（LSA）揭示了学生在数学问题解决行为的顺序上存在相似性和差异性。Apriori算法显示，无论性别如何，努力学习的学生都倾向于成功解决问题；然而，苦苦挣扎的女学生往往比男同学更能成功地解决问题。尽管如此，无论他们的数学能力水平和使用的软件版本如何，所有的学生都解决了他们熟悉的问题，并且总是从容易的问题开始，逐渐进步。讨论了研究的局限性和未来的研究方向。

{"title":"Patterns of mathematics problem solving and synthetic facial expressions in a personal instructing agent","authors":"John Lorenz Dela Cruz , Paulyn Joy Dela Cruz , Joyce Antonette Guadalupe , Jiabianca Macaraeg , Piolo Jose Montesa , Mark Paul Ramos , Rex P. Bringula , Kaoru Sumi","doi":"10.1016/j.caeo.2025.100301","DOIUrl":"10.1016/j.caeo.2025.100301","url":null,"abstract":"<div><div>This study explored the patterns of mathematics problem-solving and synthetic facial expressions (SFEs) exhibited by a personal instructing agent named PIA. Toward this goal, 81 Grade 8 students participated in a three-day experiment where they were randomly assigned either to the facial (FG) or non-facial (NFG) group. The students’ interactions within the PIA were collected and stored as log files. The attributes extracted from the log files included types of mathematics problems solved (i.e., schema), status of the mathematics problems solved, difficulty levels of mathematics problems solved, and SFEs exhibited by the PIA. Lag sequential analysis (LSA) disclosed that there were similarities and differences in the sequence of math problem-solving behaviors among students. The Apriori algorithm revealed that struggling students tend to solve problems successfully, irrespective of their sex; however, struggling female students tend to solve more problems successfully than their male counterparts. Nonetheless, regardless of their levels of math competency and the version of software used, all students solved problems they were comfortable with and always started with easier problems, gradually progressing. Limitations and future research were also discussed.</div></div>","PeriodicalId":100322,"journal":{"name":"Computers and Education Open","volume":"9 ","pages":"Article 100301"},"PeriodicalIF":5.7,"publicationDate":"2025-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145361066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Teacher, peer, or AI? Comparing effects of feedback sources in higher education 老师、同伴还是人工智能？高等教育中反馈来源的效果比较

IF 5.7 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers and Education Open

Pub Date : 2025-10-17 DOI: 10.1016/j.caeo.2025.100300

Joshua Weidlich , Flurin Gotsch , Kai Schudel , Claudia Marusic-Würscher , Jennifer Mazzarella , Hannah Bolten , Dario Bütler , Simon Luger , Bettina Wohlfender , Katharina Maag Merki

With the emergence of Large Language Models (LLMs), AI-generated feedback is gaining traction as a scalable feedback source for higher education. To qualify as viable alternatives for higher education classrooms, its relative effectiveness compared to teacher feedback and peer feedback needs to be better understood. To this end, a randomized field experiment (N = 90) compared the effects of three feedback sources—teacher, peer, and LLM—on students’ feedback perceptions and their achievement. To eliminate potential confounds, we (a) controlled for learning gains that may result from students giving feedback to their peers and (b) blinded feedback sources from feedback recipients. Results showed that students rated teacher feedback as less fair and harder to accept than peer and LLM feedback. Students receiving teacher feedback also indicated less willingness to revise their work based on the feedback. Conversely, teacher feedback produced the strongest improvements in scientific argumentation and formal quality of students’ work. Here, LLM feedback yielded the smallest improvement overall. Lastly, feedback literacy and intrinsic motivation partly moderated feedback effects on perceptions and achievement outcomes. For example, students with more productive attitudes toward feedback achieved higher argumentation quality after receiving teacher feedback than their peers. Findings indicate that the impact of feedback on both student perceptions and performance depends on the interplay between the feedback source and learner dispositions; deliberately aligning these factors could therefore amplify the benefits of feedback interventions. Future research should explore hybrid and adaptive feedback models that integrate human and AI input.

随着大型语言模型（llm）的出现，人工智能生成的反馈作为高等教育的可扩展反馈来源正在获得关注。为了使其成为高等教育课堂的可行替代方案，需要更好地了解其与教师反馈和同伴反馈相比的相对有效性。为此，一项随机现场实验（N = 90）比较了三种反馈来源——教师、同伴和法学硕士——对学生反馈感知和他们的成就的影响。为了消除潜在的混淆，我们(a)控制了学生向同龄人提供反馈可能带来的学习收益，(b)对反馈接受者的反馈来源进行了盲化。结果显示，学生认为教师的反馈比同伴和法学硕士的反馈更不公平，更难以接受。收到老师反馈的学生也表示不太愿意根据反馈修改他们的作业。相反，教师反馈在科学论证和学生作业的正式质量方面产生了最大的改善。在这里，LLM反馈产生了最小的整体改进。最后，反馈素养和内在动机在一定程度上调节了反馈对认知和成就结果的影响。例如，对反馈持更积极态度的学生在接受教师反馈后的论证质量高于同龄人。研究结果表明，反馈对学生认知和成绩的影响取决于反馈来源和学习者倾向之间的相互作用；因此，有意地调整这些因素可以放大反馈干预的好处。未来的研究应该探索混合和自适应反馈模型，将人类和人工智能的输入结合起来。

{"title":"Teacher, peer, or AI? Comparing effects of feedback sources in higher education","authors":"Joshua Weidlich , Flurin Gotsch , Kai Schudel , Claudia Marusic-Würscher , Jennifer Mazzarella , Hannah Bolten , Dario Bütler , Simon Luger , Bettina Wohlfender , Katharina Maag Merki","doi":"10.1016/j.caeo.2025.100300","DOIUrl":"10.1016/j.caeo.2025.100300","url":null,"abstract":"<div><div>With the emergence of Large Language Models (LLMs), AI-generated feedback is gaining traction as a scalable feedback source for higher education. To qualify as viable alternatives for higher education classrooms, its relative effectiveness compared to teacher feedback and peer feedback needs to be better understood. To this end, a randomized field experiment (<em>N</em> = 90) compared the effects of three feedback sources—teacher, peer, and LLM—on students’ feedback perceptions and their achievement. To eliminate potential confounds, we (a) controlled for learning gains that may result from students giving feedback to their peers and (b) blinded feedback sources from feedback recipients. Results showed that students rated teacher feedback as less fair and harder to accept than peer and LLM feedback. Students receiving teacher feedback also indicated less willingness to revise their work based on the feedback. Conversely, teacher feedback produced the strongest improvements in scientific argumentation and formal quality of students’ work. Here, LLM feedback yielded the smallest improvement overall. Lastly, feedback literacy and intrinsic motivation partly moderated feedback effects on perceptions and achievement outcomes. For example, students with more productive attitudes toward feedback achieved higher argumentation quality after receiving teacher feedback than their peers. Findings indicate that the impact of feedback on both student perceptions and performance depends on the interplay between the feedback source and learner dispositions; deliberately aligning these factors could therefore amplify the benefits of feedback interventions. Future research should explore hybrid and adaptive feedback models that integrate human and AI input.</div></div>","PeriodicalId":100322,"journal":{"name":"Computers and Education Open","volume":"9 ","pages":"Article 100300"},"PeriodicalIF":5.7,"publicationDate":"2025-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145415732","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Impact of virtual reality learning environments on skills development in students with ASD 虚拟现实学习环境对ASD学生技能发展的影响

IF 5.7 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers and Education Open

Pub Date : 2025-10-02 DOI: 10.1016/j.caeo.2025.100298

Rui Manuel Silva , Paulo Martins , Tânia Rocha

Background

Students with Autism Spectrum Disorder (ASD) often face significant challenges in traditional educational environments, including difficulties in social interaction, engagement, and adapting to standard learning methods. These barriers can hinder their academic and personal development, highlighting the need for more inclusive and adaptive educational solutions.

Objective

This study investigated whether immersive VR-based STEM learning environments can support the cognitive, social and behavioural development of pupils with ASD. We evaluated usability and accessibility needs, validated the artefact through expert consensus, and measured pre–post changes using established standardised instruments.

Methodology

The research followed the Design Science Research (DSR) approach within STEM (Science, Technology, Engineering, and Mathematics) to develop VR-based learning experiences adapted to the needs of students with ASD. The Delphi method involved experts in defining best practices and educational strategies, helping to ensure that the proposed solutions were appropriate and aligned with student characteristics. The study included a control and an experimental group, both composed of students with ASD and typically developing students, assessing the impact of VR on learning and socialisation.

Results

The findings suggest that VR-based learning environments may support improvements in cognitive, behavioural and social skills, although causal inference is limited by the small sample size and absence of randomisation.

Conclusions

This study provides preliminary evidence that VR-based learning environments may help address educational barriers for students with ASD by offering structured, engaging and adaptable environments that could support inclusion and development.

自闭症谱系障碍（ASD）学生在传统的教育环境中经常面临重大挑战，包括社会互动、参与和适应标准学习方法方面的困难。这些障碍会阻碍他们的学业和个人发展，因此需要更具包容性和适应性的教育解决方案。目的探讨沉浸式vr STEM学习环境对ASD患儿认知、社交和行为发展的促进作用。我们评估可用性和可访问性需求，通过专家共识验证工件，并使用已建立的标准化工具测量前后变化。该研究遵循STEM（科学、技术、工程和数学）中的设计科学研究（DSR）方法，开发基于vr的学习体验，以适应自闭症学生的需求。德尔菲法让专家来定义最佳实践和教育策略，帮助确保提出的解决方案是适当的，并与学生的特点保持一致。该研究包括一个对照组和一个实验组，都由自闭症学生和正常发展的学生组成，评估虚拟现实对学习和社交的影响。研究结果表明，基于虚拟现实的学习环境可能支持认知、行为和社交技能的提高，尽管因果推理受到小样本量和缺乏随机化的限制。本研究提供了初步证据，表明基于虚拟现实的学习环境可以通过提供结构化、引人入胜和适应性强的环境来支持包容和发展，从而有助于解决自闭症学生的教育障碍。

{"title":"Impact of virtual reality learning environments on skills development in students with ASD","authors":"Rui Manuel Silva , Paulo Martins , Tânia Rocha","doi":"10.1016/j.caeo.2025.100298","DOIUrl":"10.1016/j.caeo.2025.100298","url":null,"abstract":"<div><h3>Background</h3><div>Students with Autism Spectrum Disorder (ASD) often face significant challenges in traditional educational environments, including difficulties in social interaction, engagement, and adapting to standard learning methods. These barriers can hinder their academic and personal development, highlighting the need for more inclusive and adaptive educational solutions.</div></div><div><h3>Objective</h3><div>This study investigated whether immersive VR-based STEM learning environments can support the cognitive, social and behavioural development of pupils with ASD. We evaluated usability and accessibility needs, validated the artefact through expert consensus, and measured pre–post changes using established standardised instruments.</div></div><div><h3>Methodology</h3><div>The research followed the Design Science Research (DSR) approach within STEM (Science, Technology, Engineering, and Mathematics) to develop VR-based learning experiences adapted to the needs of students with ASD. The Delphi method involved experts in defining best practices and educational strategies, helping to ensure that the proposed solutions were appropriate and aligned with student characteristics. The study included a control and an experimental group, both composed of students with ASD and typically developing students, assessing the impact of VR on learning and socialisation.</div></div><div><h3>Results</h3><div>The findings suggest that VR-based learning environments may support improvements in cognitive, behavioural and social skills, although causal inference is limited by the small sample size and absence of randomisation.</div></div><div><h3>Conclusions</h3><div>This study provides preliminary evidence that VR-based learning environments may help address educational barriers for students with ASD by offering structured, engaging and adaptable environments that could support inclusion and development.</div></div>","PeriodicalId":100322,"journal":{"name":"Computers and Education Open","volume":"9 ","pages":"Article 100298"},"PeriodicalIF":5.7,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145264959","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

User perception and anatomical understanding from use of 3D anatomy technology by exercise science students: A pilot study 运动科学学生使用3D解剖技术的用户感知和解剖理解：一项试点研究

IF 5.7 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers and Education Open

Pub Date : 2025-10-01 DOI: 10.1016/j.caeo.2025.100295

Leesa Anne Grier , Anthony Leicht

Understanding anatomy poses a significant challenge for non-medical university students, particularly those without prior science knowledge, such as exercise science students. Technology-Enhanced Learning (TEL) resources, including 3D anatomy platforms, have been explored to support student engagement and improve academic outcomes with variable results.

Objectives: This pilot study investigated the potential of the Complete Anatomy (CA) 3DTEL resource, combined with traditional blended learning methods, to enhance anatomy knowledge of first-year, undergraduate exercise science students.

Design: Cohort observational study.

Methods: The study followed 36 participants (46 % female, 54 % male) across two 13-week units. Guided implementation of CA was introduced during unit 1, while students engaged with the resource independently during unit 2. User satisfaction was assessed via surveys, and academic performance evaluated by comparing final unit grades with a 2022 cohort that did not use the resource.

Results: Results indicated stable student satisfaction and a significantly different academic performance for the 2023 cohort, with median grades increasing from a Pass (50.0–64.9 %) to a Credit (65.0–74.9 %).

Conclusion: These findings suggest that integrating 3DTEL resources with blended learning can positively support anatomy learning for science-naïve students.

理解解剖学对非医科大学的学生来说是一个巨大的挑战，特别是那些没有科学知识的学生，比如运动科学的学生。包括3D解剖平台在内的技术增强学习（TEL）资源已被探索，以支持学生的参与，并以不同的结果改善学术成果。目的：本初步研究探讨了完整解剖学（CA） 3DTEL资源与传统混合式学习方法相结合的潜力，以提高运动科学本科一年级学生的解剖学知识。设计：队列观察研究。方法：该研究对36名参与者（46%女性，54%男性）进行了为期13周的随访。第一单元引入了CA的指导实施，而学生在第二单元独立使用资源。通过调查来评估用户满意度，通过将最终单元成绩与未使用该资源的2022年队列进行比较来评估学习成绩。结果：结果表明，2023年的学生满意度稳定，学习成绩显著不同，成绩中位数从及格（50.0% - 64.9%）增加到学分（65.0% - 74.9%）。结论：将3DTEL资源与混合学习相结合，对science-naïve学生的解剖学学习具有积极的支持作用。

{"title":"User perception and anatomical understanding from use of 3D anatomy technology by exercise science students: A pilot study","authors":"Leesa Anne Grier , Anthony Leicht","doi":"10.1016/j.caeo.2025.100295","DOIUrl":"10.1016/j.caeo.2025.100295","url":null,"abstract":"<div><div>Understanding anatomy poses a significant challenge for non-medical university students, particularly those without prior science knowledge, such as exercise science students. Technology-Enhanced Learning (TEL) resources, including 3D anatomy platforms, have been explored to support student engagement and improve academic outcomes with variable results.</div><div>Objectives: This pilot study investigated the potential of the Complete Anatomy (CA) 3DTEL resource, combined with traditional blended learning methods, to enhance anatomy knowledge of first-year, undergraduate exercise science students.</div><div>Design: Cohort observational study.</div><div>Methods: The study followed 36 participants (46 % female, 54 % male) across two 13-week units. Guided implementation of CA was introduced during unit 1, while students engaged with the resource independently during unit 2. User satisfaction was assessed via surveys, and academic performance evaluated by comparing final unit grades with a 2022 cohort that did not use the resource.</div><div>Results: Results indicated stable student satisfaction and a significantly different academic performance for the 2023 cohort, with median grades increasing from a Pass (50.0–64.9 %) to a Credit (65.0–74.9 %).</div><div>Conclusion: These findings suggest that integrating 3DTEL resources with blended learning can positively support anatomy learning for science-naïve students.</div></div>","PeriodicalId":100322,"journal":{"name":"Computers and Education Open","volume":"9 ","pages":"Article 100295"},"PeriodicalIF":5.7,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145320141","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Direct vs. video observation of skill performance: effects on peer feedback dynamics in motor learning 技能表现的直接观察与视频观察：同伴反馈动态对运动学习的影响

IF 5.7 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers and Education Open

Pub Date : 2025-10-01 DOI: 10.1016/j.caeo.2025.100296

Omar Trabelsi , Mohamed Yaakoubi , Ahmed Ghorbel , Amir Romdhani , Mustapha Bouchiba , Mohamed Abdelkader Souissi , Okba Selmi , Katja Weiss , Thomas Rosemann , Adnene Gharbi , Beat Knechtle

Video technology facilitates feedback delivery in motor learning, benefiting teacher/coach and peer feedback. Most studies focus on attributing motor learning improvements to video-based feedback but never examine how peer feedback dynamics change to mediate the effect of video observation (VO) on learning outcomes. Therefore, this study compares the frequency, type, and accuracy of peer feedback based on direct observation (DO) and VO in a long jump learning context. Forty-one sports science students (M^age: 20.13±0.71) participated in a four-session long jump learning unit. Students were then randomly paired for experimental procedures: one performed a jump while the other observed. Observers first provided 30 seconds of verbal feedback based on DO and then after viewing a video recording of the jump (VO). Roles were then switched. Audio recordings were transcribed and analyzed for overall feedback frequency. Feedback instances were then classified as implicit or explicit, with the latter assessed for accuracy. The main results showed that VO significantly increased the median frequency of overall feedback and explicit feedback compared to DO, with no significant difference in implicit feedback. The median accuracy of explicit feedback was also significantly higher based on VO compared to DO. These findings help explain the previously documented beneficial effects of video-based peer feedback in motor learning. They demonstrate that VO enables peers to provide more feedback, particularly explicit, more than implicit, with higher accuracy.

视频技术有助于在运动学习中提供反馈，有利于教师/教练和同伴反馈。大多数研究将运动学习的改善归因于基于视频的反馈，但从未研究同伴反馈动态变化如何调节视频观察（VO）对学习结果的影响。因此，本研究比较了跳远学习情境下基于直接观察的同伴反馈（DO）和基于直接观察的同伴反馈（VO）的频率、类型和准确性。41名体育理科生（年龄：20.13±0.71）参加了四节跳远学习单元。然后，学生们被随机配对，进行实验程序：一个人跳，另一个人观察。观察者首先根据DO提供30秒的口头反馈，然后观看跳跃的视频记录（VO）。然后角色互换了。录音被转录并分析总体反馈频率。然后将反馈实例分类为隐式或显式，并评估后者的准确性。主要结果表明，VO显著提高了整体反馈和显式反馈的中位数频率，而内隐反馈的中位数频率差异不显著。基于VO的显式反馈的中位数准确性也显著高于基于DO的中位数准确性。这些发现有助于解释先前记录的基于视频的同伴反馈对运动学习的有益影响。他们证明，VO使同行能够提供更多的反馈，特别是明确的，而不是隐含的，具有更高的准确性。

{"title":"Direct vs. video observation of skill performance: effects on peer feedback dynamics in motor learning","authors":"Omar Trabelsi , Mohamed Yaakoubi , Ahmed Ghorbel , Amir Romdhani , Mustapha Bouchiba , Mohamed Abdelkader Souissi , Okba Selmi , Katja Weiss , Thomas Rosemann , Adnene Gharbi , Beat Knechtle","doi":"10.1016/j.caeo.2025.100296","DOIUrl":"10.1016/j.caeo.2025.100296","url":null,"abstract":"<div><div>Video technology facilitates feedback delivery in motor learning, benefiting teacher/coach and peer feedback. Most studies focus on attributing motor learning improvements to video-based feedback but never examine how peer feedback dynamics change to mediate the effect of video observation (VO) on learning outcomes. Therefore, this study compares the frequency, type, and accuracy of peer feedback based on direct observation (DO) and VO in a long jump learning context. Forty-one sports science students (M<sup>age</sup>: 20.13±0.71) participated in a four-session long jump learning unit. Students were then randomly paired for experimental procedures: one performed a jump while the other observed. Observers first provided 30 seconds of verbal feedback based on DO and then after viewing a video recording of the jump (VO). Roles were then switched. Audio recordings were transcribed and analyzed for overall feedback frequency. Feedback instances were then classified as implicit or explicit, with the latter assessed for accuracy. The main results showed that VO significantly increased the median frequency of overall feedback and explicit feedback compared to DO, with no significant difference in implicit feedback. The median accuracy of explicit feedback was also significantly higher based on VO compared to DO. These findings help explain the previously documented beneficial effects of video-based peer feedback in motor learning. They demonstrate that VO enables peers to provide more feedback, particularly explicit, more than implicit, with higher accuracy.</div></div>","PeriodicalId":100322,"journal":{"name":"Computers and Education Open","volume":"9 ","pages":"Article 100296"},"PeriodicalIF":5.7,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145264961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Social media discussions on educators: Selecting and appraisal of recent research using TF-IDF 社会媒体对教育者的讨论：使用TF-IDF选择和评价最近的研究

IF 5.7 Q2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers and Education Open

Pub Date : 2025-09-30 DOI: 10.1016/j.caeo.2025.100293

Mateo R. Borbon Jr. , Ryan A. Ebardo

This systematic literature review, analyzing 36 peer-reviewed publications from 2019 to February of 2025, addresses a critical gap by examining the use of social media analytics (SMA) for faculty evaluation. Employing a novel methodological approach that combines machine learning-assisted screening (ASReview) with TF-IDF, the study finds that platforms like Twitter and Facebook are increasingly analyzed using sentiment analysis, machine learning, and text mining. These techniques provide real-time, unfiltered student feedback on teaching effectiveness, complementing traditional evaluation instruments and helping to monitor institutional reputation. While SMA offers valuable insights, the review highlights significant challenges, including data quality and credibility, algorithmic bias, ethical concerns, and generalizability. Effectively leveraging SMA's potential requires addressing these issues through robust theoretical frameworks, balanced institutional policies, and enhanced digital literacy to improve teaching practices while safeguarding academic integrity.

这篇系统的文献综述分析了2019年至2025年2月的36篇同行评审的出版物，通过研究社交媒体分析（SMA）在教师评估中的使用，解决了一个关键的差距。该研究采用了一种将机器学习辅助筛选（ASReview）与TF-IDF相结合的新方法，发现Twitter和Facebook等平台越来越多地使用情感分析、机器学习和文本挖掘进行分析。这些技术提供了实时的、未经过滤的学生对教学效果的反馈，补充了传统的评估工具，并有助于监测机构的声誉。虽然SMA提供了有价值的见解，但该综述强调了重大挑战，包括数据质量和可信度、算法偏见、伦理问题和可泛化性。有效利用SMA的潜力需要通过健全的理论框架、平衡的制度政策和增强数字素养来解决这些问题，以改善教学实践，同时维护学术诚信。

引用次数: 0