Computers and Education Artificial Intelligence最新文献_第5页

RICE AlgebraBot: Lessons learned from designing and developing responsible conversational AI using induction, concretization, and exemplification to support algebra learning

Q1 Social Sciences

Computers and Education Artificial Intelligence

Pub Date : 2024-12-06 DOI: 10.1016/j.caeai.2024.100338

Chenglu Li , Wanli Xing , Yukyeong Song , Bailing Lyu

The importance and challenge of Algebra learning is widely recognized, with students across the U.S. facing difficulties due to the subject's complexity. While extensive research has focused on enhancing Algebra learning in K-12 education, the reusability, scalability, and effectiveness of the strategies employed (e.g., manual interventions and digital tutoring platforms) remain limited. Conversational AI (ConvAI), enabled by the advancement of large language models (LLMs), emerges as a potential tool for automatic, personalized, and effective student support. However, ethical concerns surrounding diversity, safety, sentiment, and stereotype associated with ConvAI are prominent, and empirical studies examining its application in education are scarce. The purpose of this study is to develop a ConvAI system that mitigates the potential ethical concerns and empirically evaluate the effect of such a system for math learning. Specifically, we first examined computational strategies to mitigate the ethical concerns of ConvAI in educational setting with educational big data (n_pretraining = 2,097,139) and found that researchers could effectively enhance ConvAI responsibility through the investigated algorithmic strategies. Then, a ConvAI system was constructed using these strategies, guided by learning sciences principles. Lastly, we examined students' eye-tracking patterns, acceptance, and learning processes when using this ConvAI system to learn Algebra through a random experiment (n_participant = 151). Participants using the developed ConvAI demonstrated generally increased visual attention levels as compared to the control group. Moreover, participants expressed a positive acceptance towards the ConvAI technology. Finally, participants' interaction patterns with the ConvAI technology influenced their Algebra learning. These results provide insights for both educational researchers and practitioners to integrate ConvAI in learning environments.

{"title":"RICE AlgebraBot: Lessons learned from designing and developing responsible conversational AI using induction, concretization, and exemplification to support algebra learning","authors":"Chenglu Li , Wanli Xing , Yukyeong Song , Bailing Lyu","doi":"10.1016/j.caeai.2024.100338","DOIUrl":"10.1016/j.caeai.2024.100338","url":null,"abstract":"<div><div>The importance and challenge of Algebra learning is widely recognized, with students across the U.S. facing difficulties due to the subject's complexity. While extensive research has focused on enhancing Algebra learning in K-12 education, the reusability, scalability, and effectiveness of the strategies employed (e.g., manual interventions and digital tutoring platforms) remain limited. Conversational <span>AI</span> (ConvAI), enabled by the advancement of large language models (LLMs), emerges as a potential tool for automatic, personalized, and effective student support. However, ethical concerns surrounding diversity, safety, sentiment, and stereotype associated with ConvAI are prominent, and empirical studies examining its application in education are scarce. The purpose of this study is to develop a ConvAI system that mitigates the potential ethical concerns and empirically evaluate the effect of such a system for math learning. Specifically, we first examined computational strategies to mitigate the ethical concerns of ConvAI in educational setting with educational big data (n<sub>pretraining</sub> = 2,097,139) and found that researchers could effectively enhance ConvAI responsibility through the investigated algorithmic strategies. Then, a ConvAI system was constructed using these strategies, guided by learning sciences principles. Lastly, we examined students' eye-tracking patterns, acceptance, and learning processes when using this ConvAI system to learn Algebra through a random experiment (n<sub>participant</sub> = 151). Participants using the developed ConvAI demonstrated generally increased visual attention levels as compared to the control group. Moreover, participants expressed a positive acceptance towards the ConvAI technology. Finally, participants' interaction patterns with the ConvAI technology influenced their Algebra learning. These results provide insights for both educational researchers and practitioners to integrate ConvAI in learning environments.</div></div>","PeriodicalId":34469,"journal":{"name":"Computers and Education Artificial Intelligence","volume":"8 ","pages":"Article 100338"},"PeriodicalIF":0.0,"publicationDate":"2024-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143145839","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

AI advocates and cautious critics: How AI attitudes, AI interest, use of AI, and AI literacy build university students' AI self-efficacy

Q1 Social Sciences

Computers and Education Artificial Intelligence

Pub Date : 2024-12-06 DOI: 10.1016/j.caeai.2024.100340

Arne Bewersdorff , Marie Hornberger , Claudia Nerdel , Daniel S. Schiff

This study investigates how cognitive, affective, and behavioral variables related to artificial intelligence (AI) build AI self-efficacy among university students. Based on these variables, we identify three meaningful student groups, which can guide educational initiatives. We recruited 1465 undergraduate and graduate students from the United States, the United Kingdom, and Germany and measured their AI self-efficacy, AI literacy, interest in AI, attitudes towards AI, and AI use. Using a path model, we examine the correlations and paths among these variables. Results reveal that AI usage and positive AI attitudes significantly predict interest in AI, which in turn and together with AI literacy, enhance AI self-efficacy. Moreover, using Gaussian Mixture Models, we identify three groups of students: 'AI Advocates,' 'Cautious Critics,' and 'Pragmatic Observers,' each exhibiting unique patterns of AI-related cognitive, affective, and behavioral traits. Our findings demonstrate the necessity of educational strategies that not only focus on AI literacy but also aim to foster students' AI attitudes, usage, and interest to effectively promote AI self-efficacy. Furthermore, we argue that educators who aim to design inclusive AI educational programs should take into account the distinct needs of different student groups identified in this study.

{"title":"AI advocates and cautious critics: How AI attitudes, AI interest, use of AI, and AI literacy build university students' AI self-efficacy","authors":"Arne Bewersdorff , Marie Hornberger , Claudia Nerdel , Daniel S. Schiff","doi":"10.1016/j.caeai.2024.100340","DOIUrl":"10.1016/j.caeai.2024.100340","url":null,"abstract":"<div><div>This study investigates how cognitive, affective, and behavioral variables related to artificial intelligence (AI) build AI self-efficacy among university students. Based on these variables, we identify three meaningful student groups, which can guide educational initiatives. We recruited 1465 undergraduate and graduate students from the United States, the United Kingdom, and Germany and measured their AI self-efficacy, AI literacy, interest in AI, attitudes towards AI, and AI use. Using a path model, we examine the correlations and paths among these variables. Results reveal that AI usage and positive AI attitudes significantly predict interest in AI, which in turn and together with AI literacy, enhance AI self-efficacy. Moreover, using Gaussian Mixture Models, we identify three groups of students: 'AI Advocates,' 'Cautious Critics,' and 'Pragmatic Observers,' each exhibiting unique patterns of AI-related cognitive, affective, and behavioral traits. Our findings demonstrate the necessity of educational strategies that not only focus on AI literacy but also aim to foster students' AI attitudes, usage, and interest to effectively promote AI self-efficacy. Furthermore, we argue that educators who aim to design inclusive AI educational programs should take into account the distinct needs of different student groups identified in this study.</div></div>","PeriodicalId":34469,"journal":{"name":"Computers and Education Artificial Intelligence","volume":"8 ","pages":"Article 100340"},"PeriodicalIF":0.0,"publicationDate":"2024-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143145837","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Pre-service teachers preparedness for AI-integrated education: An investigation from perceptions, capabilities, and teachers’ identity changes

Q1 Social Sciences

Computers and Education Artificial Intelligence

Pub Date : 2024-12-06 DOI: 10.1016/j.caeai.2024.100341

Lihang Guan , Yue Zhang , Mingyue Michelle Gu

Artificial intelligence (AI) has provided numerous learning benefits due to and beyond its personalization capabilities. AI significantly alters the interaction dynamics among students, teachers, and technology, necessitating the collaborative integration of AI in education instead of mere mechanical use. This study investigates pre-service teachers' perceptions and capabilities within the framework of social cognitive theory, focusing on their attitudes, intentions, self-efficacy, and AI literacy for effective AI integration. We discuss how these perceptions and capabilities help pre-service teachers prepare for the changing identities of being a teacher in AI-integrated education. Thematic analysis of qualitative interviews with a sample of 24 pre-service teachers suggests that (1) these pre-service teachers only use AI when warranted, (2) they need more understanding regarding AI fundamentals and ethics for AI's integration into education, and (3) they mechanically view AI as a tool to be utilized, rather than a dynamic collaborator in education. The findings indicate a need for more awareness of the possible changes in teachers' functions and roles in collaborative AI-integrated education, leading to specific teacher-training demands that aid them in success in AI-integrated education.

{"title":"Pre-service teachers preparedness for AI-integrated education: An investigation from perceptions, capabilities, and teachers’ identity changes","authors":"Lihang Guan , Yue Zhang , Mingyue Michelle Gu","doi":"10.1016/j.caeai.2024.100341","DOIUrl":"10.1016/j.caeai.2024.100341","url":null,"abstract":"<div><div>Artificial intelligence (AI) has provided numerous learning benefits due to and beyond its personalization capabilities. AI significantly alters the interaction dynamics among students, teachers, and technology, necessitating the collaborative integration of AI in education instead of mere mechanical use. This study investigates pre-service teachers' perceptions and capabilities within the framework of social cognitive theory, focusing on their attitudes, intentions, self-efficacy, and AI literacy for effective AI integration. We discuss how these perceptions and capabilities help pre-service teachers prepare for the changing identities of being a teacher in AI-integrated education. Thematic analysis of qualitative interviews with a sample of 24 pre-service teachers suggests that (1) these pre-service teachers only use AI when warranted, (2) they need more understanding regarding AI fundamentals and ethics for AI's integration into education, and (3) they mechanically view AI as a tool to be utilized, rather than a dynamic collaborator in education. The findings indicate a need for more awareness of the possible changes in teachers' functions and roles in collaborative AI-integrated education, leading to specific teacher-training demands that aid them in success in AI-integrated education.</div></div>","PeriodicalId":34469,"journal":{"name":"Computers and Education Artificial Intelligence","volume":"8 ","pages":"Article 100341"},"PeriodicalIF":0.0,"publicationDate":"2024-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143145482","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Harnessing the power of AI-instructor collaborative grading approach: Topic-based effective grading for semi open-ended multipart questions

Q1 Social Sciences

Computers and Education Artificial Intelligence

Pub Date : 2024-12-01 DOI: 10.1016/j.caeai.2024.100339

Phyo Yi Win Myint, Siaw Ling Lo, Yuhao Zhang

Semi open-ended multipart questions consist of multiple sub questions within a single question, requiring students to provide certain factual information while allowing them to express their opinion within a defined context. Human grading of such questions can be tedious, constrained by the marking scheme and susceptible to the subjective judgement of instructors. The emergence of large language models (LLMs) such as ChatGPT has significantly advanced the prospect of automatic grading in educational settings. This paper introduces a topic-based grading approach that harnesses LLM capabilities alongside a refined marking scheme to ensure fair and explainable assessment processes. The proposed approach involves segmenting student responses according to sub questions, extracting topics utilizing LLM, and refining the marking scheme in consultation with instructors. The refined marking scheme is derived from LLM-extracted topics, validated by instructors to augment the original grading criteria. Leveraging LLM, we match student responses with refined marking scheme topics and employ a Python program to assign marks based on the matches. Various prompt versions are compared using relevant metrics to determine the most effective prompts. We evaluate LLM's grading proficiency through three approaches: zero-shot prompting, few-shot prompting, and our proposed method. Results indicate that while zero-shot and few-shot prompting methods fall short compared to human grading, the proposed approach achieves the best performance (highest percentage of exact match marks, lowest mean absolute error, highest Spearman correlation, highest Cohen's weighted kappa) and closely mirrors the distribution observed in human grading. Specifically, the collaborative approach enhances the grading process by refining the marking scheme to student responses, improving transparency and explainability through topic-based matching, and significantly increasing the effectiveness of LLMs when combined with instructor input, rather than as standalone automated grading systems.

{"title":"Harnessing the power of AI-instructor collaborative grading approach: Topic-based effective grading for semi open-ended multipart questions","authors":"Phyo Yi Win Myint, Siaw Ling Lo, Yuhao Zhang","doi":"10.1016/j.caeai.2024.100339","DOIUrl":"10.1016/j.caeai.2024.100339","url":null,"abstract":"<div><div>Semi open-ended multipart questions consist of multiple sub questions within a single question, requiring students to provide certain factual information while allowing them to express their opinion within a defined context. Human grading of such questions can be tedious, constrained by the marking scheme and susceptible to the subjective judgement of instructors. The emergence of large language models (LLMs) such as ChatGPT has significantly advanced the prospect of automatic grading in educational settings. This paper introduces a topic-based grading approach that harnesses LLM capabilities alongside a refined marking scheme to ensure fair and explainable assessment processes. The proposed approach involves segmenting student responses according to sub questions, extracting topics utilizing LLM, and refining the marking scheme in consultation with instructors. The refined marking scheme is derived from LLM-extracted topics, validated by instructors to augment the original grading criteria. Leveraging LLM, we match student responses with refined marking scheme topics and employ a Python program to assign marks based on the matches. Various prompt versions are compared using relevant metrics to determine the most effective prompts. We evaluate LLM's grading proficiency through three approaches: zero-shot prompting, few-shot prompting, and our proposed method. Results indicate that while zero-shot and few-shot prompting methods fall short compared to human grading, the proposed approach achieves the best performance (highest percentage of exact match marks, lowest mean absolute error, highest Spearman correlation, highest Cohen's weighted kappa) and closely mirrors the distribution observed in human grading. Specifically, the collaborative approach enhances the grading process by refining the marking scheme to student responses, improving transparency and explainability through topic-based matching, and significantly increasing the effectiveness of LLMs when combined with instructor input, rather than as standalone automated grading systems.</div></div>","PeriodicalId":34469,"journal":{"name":"Computers and Education Artificial Intelligence","volume":"7 ","pages":"Article 100339"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143154121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Enhancing Feedback Quality at Scale: Leveraging Machine Learning for Learner-Centered Feedback

Q1 Social Sciences

Computers and Education Artificial Intelligence

Pub Date : 2024-12-01 DOI: 10.1016/j.caeai.2024.100332

Ahmad Ari Aldino , Yi-Shan Tsai , Rafael Ferreira Mello , Dragan Gašević , Guanliang Chen

In higher education, delivering effective feedback is pivotal for enhancing student learning but remains challenging due to the scale and diversity of student populations. Learner-centered feedback, a robust approach to effective feedback that tailors to individual student needs, encompasses three key dimensions—Future Impact, Sensemaking, and Agency, which collectively include eight specific components, thereby enhancing its relevance and impact in the learning process. However, providing consistent and effective learner-centered feedback at scale is challenging for educators. This study addresses this challenge by automating the analysis of feedback content to promote effective learner-centered feedback principles. We gathered a large-scale dataset of 16,531 feedback entries from 95 courses from one Master's and one Bachelor's program within a large faculty at an Australian university, with each entry labeled in accordance with the learner-centered feedback framework. Employing a range of machine learning and deep learning techniques, including Random Forest, XGBoost, BERT, and GPT-3.5, we systematically investigated the effectiveness of different approaches for constructing classifiers to accurately categorize feedback into various learner-centered components. The results demonstrated that the BERT-based classifiers outperformed other models in most feedback categories (achieving Cohen's κ up to 0.956 and F1 score up to 0.998), but showed relatively low performance in categories with less training data. This automated analysis aids in scrutinizing feedback quality, thereby supporting educators in enhancing their feedback practices to be more aligned with learner-centered principles.

{"title":"Enhancing Feedback Quality at Scale: Leveraging Machine Learning for Learner-Centered Feedback","authors":"Ahmad Ari Aldino , Yi-Shan Tsai , Rafael Ferreira Mello , Dragan Gašević , Guanliang Chen","doi":"10.1016/j.caeai.2024.100332","DOIUrl":"10.1016/j.caeai.2024.100332","url":null,"abstract":"<div><div>In higher education, delivering effective feedback is pivotal for enhancing student learning but remains challenging due to the scale and diversity of student populations. Learner-centered feedback, a robust approach to effective feedback that tailors to individual student needs, encompasses three key dimensions—<span>Future Impact</span>, <span>Sensemaking</span>, and <span>Agency</span>, which collectively include eight specific components, thereby enhancing its relevance and impact in the learning process. However, providing consistent and effective learner-centered feedback at scale is challenging for educators. This study addresses this challenge by automating the analysis of feedback content to promote effective learner-centered feedback principles. We gathered a large-scale dataset of 16,531 feedback entries from 95 courses from one Master's and one Bachelor's program within a large faculty at an Australian university, with each entry labeled in accordance with the learner-centered feedback framework. Employing a range of machine learning and deep learning techniques, including Random Forest, XGBoost, BERT, and GPT-3.5, we systematically investigated the effectiveness of different approaches for constructing classifiers to accurately categorize feedback into various learner-centered components. The results demonstrated that the BERT-based classifiers outperformed other models in most feedback categories (achieving Cohen's <em>κ</em> up to 0.956 and F1 score up to 0.998), but showed relatively low performance in categories with less training data. This automated analysis aids in scrutinizing feedback quality, thereby supporting educators in enhancing their feedback practices to be more aligned with learner-centered principles.</div></div>","PeriodicalId":34469,"journal":{"name":"Computers and Education Artificial Intelligence","volume":"7 ","pages":"Article 100332"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143153742","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Teachers’ generative AI self-efficacy, valuing, and integration at work: Examining job resources and demands 教师在工作中的生成性人工智能自我效能感、重视与整合：考察工作资源与需求

Q1 Social Sciences

Computers and Education Artificial Intelligence

Pub Date : 2024-12-01 DOI: 10.1016/j.caeai.2024.100333

Rebecca J. Collie , Andrew J. Martin , Dragan Gasevic

Generative AI (genAI) tools have involved rapid and broad uptake since their wide release in late 2022, including among teachers. We investigated several factors that play a role in teachers’ motivation and engagement to harness genAI in teaching and learning. We examined contextual factors (in-school support to apply genAI, time pressure, disruptive student behavior) as predictors of motivation (genAI self-efficacy and genAI valuing) and, in turn, engagement (i.e., genAI integration in teaching-related work and student learning activities) over the course of one school term. Among 368 Australian primary and secondary school teachers, our findings revealed that genAI support was associated with greater genAI self-efficacy and genAI valuing. Time pressure was also linked with greater genAI valuing, whereas disruptive student behavior was not linked with the genAI motivation or engagement variables. In turn, genAI self-efficacy was linked with greater levels of both types of genAI integration. GenAI valuing was associated with greater genAI integration in teaching-related work only. Our results provide knowledge about factors relevant for supporting genAI and its application among teachers in Australia—and also hold relevance to teachers in other countries.

自2022年底广泛发布以来，生成式人工智能（genAI）工具得到了迅速和广泛的接受，包括在教师中。我们调查了在教师的动机和参与中发挥作用的几个因素，以在教学和学习中利用基因人工智能。我们考察了在一个学期的课程中，情境因素（校内对应用基因ai的支持、时间压力、破坏性学生行为）作为动机（基因ai自我效能和基因ai价值）的预测因子，以及参与度（即基因ai融入教学相关工作和学生学习活动）的预测因子。在368名澳大利亚中小学教师中，我们的研究结果显示，基因支持与更高的基因自我效能感和基因价值相关。时间压力也与更大的基因价值有关，而破坏性学生行为与基因动机或参与变量无关。反过来，基因自我效能与两种类型的基因整合的更高水平有关。基因ai评价仅与教学相关工作中基因ai整合程度较高有关。我们的研究结果提供了支持genAI及其在澳大利亚教师中的应用的相关因素的知识，也与其他国家的教师相关。

{"title":"Teachers’ generative AI self-efficacy, valuing, and integration at work: Examining job resources and demands","authors":"Rebecca J. Collie , Andrew J. Martin , Dragan Gasevic","doi":"10.1016/j.caeai.2024.100333","DOIUrl":"10.1016/j.caeai.2024.100333","url":null,"abstract":"<div><div>Generative AI (genAI) tools have involved rapid and broad uptake since their wide release in late 2022, including among teachers. We investigated several factors that play a role in teachers’ motivation and engagement to harness genAI in teaching and learning. We examined contextual factors (in-school support to apply genAI, time pressure, disruptive student behavior) as predictors of motivation (genAI self-efficacy and genAI valuing) and, in turn, engagement (i.e., genAI integration in teaching-related work and student learning activities) over the course of one school term. Among 368 Australian primary and secondary school teachers, our findings revealed that genAI support was associated with greater genAI self-efficacy and genAI valuing. Time pressure was also linked with greater genAI valuing, whereas disruptive student behavior was not linked with the genAI motivation or engagement variables. In turn, genAI self-efficacy was linked with greater levels of both types of genAI integration. GenAI valuing was associated with greater genAI integration in teaching-related work only. Our results provide knowledge about factors relevant for supporting genAI and its application among teachers in Australia—and also hold relevance to teachers in other countries.</div></div>","PeriodicalId":34469,"journal":{"name":"Computers and Education Artificial Intelligence","volume":"7 ","pages":"Article 100333"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142746867","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

AI-powered EFL pedagogy: Integrating generative AI into university teaching preparation through UTAUT and activity theory

Q1 Social Sciences

Computers and Education Artificial Intelligence

Pub Date : 2024-12-01 DOI: 10.1016/j.caeai.2024.100335

Muhammad Zaim , Safnil Arsyad , Budi Waluyo , Havid Ardi , Muhd. Al Hafizh , Muflihatuz Zakiyah , Widya Syafitri , Ahmad Nusi , Mei Hardiah

This study explores the integration of generative AI into English as a Foreign Language (EFL) teaching preparation within Indonesian higher education, addressing the growing need to understand how emerging technologies can enhance pedagogical practices in a rapidly evolving educational landscape. By employing the Unified Theory of Acceptance and Use of Technology (UTAUT) and Activity Theory, the research provides a robust analytical framework to examine the factors influencing lecturers' adoption of generative AI. The study is particularly relevant as generative AI offers significant potential to improve teaching efficiency and content personalization, yet its adoption presents challenges in aligning outputs with educational standards and maintaining meaningful teacher-student interaction. Using a mixed-methods approach, the research combined quantitative data from structured questionnaires with qualitative insights from reflective compositions, where lecturers critically evaluated their experiences with generative AI. Structural Equation Modeling (SEM) revealed that performance expectancy and social influence significantly and positively influenced behavioral intention, while effort expectancy had no significant effect. Facilitating conditions, unexpectedly, negatively impacted behavioral intention, likely due to satisfaction with existing resources reducing the perceived necessity for new tools. A strong positive correlation between behavioral intention and actual use behavior demonstrated the critical role of intention in driving adoption. Thematic analysis provided further depth by emphasizing both the benefits and challenges of generative AI, accentuating the importance of balancing its use with human instruction to ensure quality teaching and interaction. The study stresses the need for the strategic integration of generative AI, offering practical and theoretical insights into its adoption and implications for advancing EFL teaching in higher education.

{"title":"AI-powered EFL pedagogy: Integrating generative AI into university teaching preparation through UTAUT and activity theory","authors":"Muhammad Zaim , Safnil Arsyad , Budi Waluyo , Havid Ardi , Muhd. Al Hafizh , Muflihatuz Zakiyah , Widya Syafitri , Ahmad Nusi , Mei Hardiah","doi":"10.1016/j.caeai.2024.100335","DOIUrl":"10.1016/j.caeai.2024.100335","url":null,"abstract":"<div><div>This study explores the integration of generative AI into English as a Foreign Language (EFL) teaching preparation within Indonesian higher education, addressing the growing need to understand how emerging technologies can enhance pedagogical practices in a rapidly evolving educational landscape. By employing the Unified Theory of Acceptance and Use of Technology (UTAUT) and Activity Theory, the research provides a robust analytical framework to examine the factors influencing lecturers' adoption of generative AI. The study is particularly relevant as generative AI offers significant potential to improve teaching efficiency and content personalization, yet its adoption presents challenges in aligning outputs with educational standards and maintaining meaningful teacher-student interaction. Using a mixed-methods approach, the research combined quantitative data from structured questionnaires with qualitative insights from reflective compositions, where lecturers critically evaluated their experiences with generative AI. Structural Equation Modeling (SEM) revealed that performance expectancy and social influence significantly and positively influenced behavioral intention, while effort expectancy had no significant effect. Facilitating conditions, unexpectedly, negatively impacted behavioral intention, likely due to satisfaction with existing resources reducing the perceived necessity for new tools. A strong positive correlation between behavioral intention and actual use behavior demonstrated the critical role of intention in driving adoption. Thematic analysis provided further depth by emphasizing both the benefits and challenges of generative AI, accentuating the importance of balancing its use with human instruction to ensure quality teaching and interaction. The study stresses the need for the strategic integration of generative AI, offering practical and theoretical insights into its adoption and implications for advancing EFL teaching in higher education.</div></div>","PeriodicalId":34469,"journal":{"name":"Computers and Education Artificial Intelligence","volume":"7 ","pages":"Article 100335"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143154122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Strengthening students’ research efficacy in higher institutions. A joint mediating effect of the impact of Artificial intelligence using Partial Least Squares Structural Equation Modelling (PLS-SEM)

Q1 Social Sciences

Computers and Education Artificial Intelligence

Pub Date : 2024-12-01 DOI: 10.1016/j.caeai.2024.100337

Usani Joseph Ofem , Ene I. Ene , Eunice Ngozi Ajuluchukwu , Hope Amba Neji , Imelda Barong Edam-Agbor , Faith Sylvester Orim , Chidirim Esther Nworgwugwu , Sylvia Victor Ovat , James Omaji Ukatu , Patience Ekpang , Faith Igu Ogochukwu , Hycenth Edet Ntah , Ememadukwu David Ameh

Research efficacy is a vital contributor to enhancing research productivity within higher education institutions. Despite its importance, there has been limited focus on understanding this construct and its enablers. This study addresses this gap by exploring research efficacy through the lens of artificial intelligence (AI) utilization among students. The findings reveal that AI utilization has a significant positive impact on various dimensions of research efficacy, including background, review, methodological, analytical, and reporting efficacy. Additionally, review efficacy is shown to significantly influence background, methodological, and reporting efficacy, while it does not impact analytical efficacy. Furthermore, both background and review efficacy play crucial mediating roles between AI utilization and methodological and analytical efficacy. This study highlights the necessity of a solid foundational understanding of research practices to enable informed decision-making and enhance overall research outcomes. The implications of these findings for improving research efficacy in higher education are thoroughly discussed.

{"title":"Strengthening students’ research efficacy in higher institutions. A joint mediating effect of the impact of Artificial intelligence using Partial Least Squares Structural Equation Modelling (PLS-SEM)","authors":"Usani Joseph Ofem , Ene I. Ene , Eunice Ngozi Ajuluchukwu , Hope Amba Neji , Imelda Barong Edam-Agbor , Faith Sylvester Orim , Chidirim Esther Nworgwugwu , Sylvia Victor Ovat , James Omaji Ukatu , Patience Ekpang , Faith Igu Ogochukwu , Hycenth Edet Ntah , Ememadukwu David Ameh","doi":"10.1016/j.caeai.2024.100337","DOIUrl":"10.1016/j.caeai.2024.100337","url":null,"abstract":"<div><div>Research efficacy is a vital contributor to enhancing research productivity within higher education institutions. Despite its importance, there has been limited focus on understanding this construct and its enablers. This study addresses this gap by exploring research efficacy through the lens of artificial intelligence (AI) utilization among students. The findings reveal that AI utilization has a significant positive impact on various dimensions of research efficacy, including background, review, methodological, analytical, and reporting efficacy. Additionally, review efficacy is shown to significantly influence background, methodological, and reporting efficacy, while it does not impact analytical efficacy. Furthermore, both background and review efficacy play crucial mediating roles between AI utilization and methodological and analytical efficacy. This study highlights the necessity of a solid foundational understanding of research practices to enable informed decision-making and enhance overall research outcomes. The implications of these findings for improving research efficacy in higher education are thoroughly discussed.</div></div>","PeriodicalId":34469,"journal":{"name":"Computers and Education Artificial Intelligence","volume":"7 ","pages":"Article 100337"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143153806","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The behavioural intentions for integrating artificial intelligence in science teaching among pre-service science teachers in South Africa and Thailand

Q1 Social Sciences

Computers and Education Artificial Intelligence

Pub Date : 2024-12-01 DOI: 10.1016/j.caeai.2024.100334

Lindelani Mnguni , Prasart Nuangchalerm , R. Ahmad Zaky El Islami , Doras Sibanda , Indah Juwita Sari , Moleboheng Ramulumo

Developing countries in the Global South exhibit diverse trends in the integration of digital technologies, such as Artificial Intelligence in teaching. Complex context-specific factors, including teacher preparedness, influence these trends. Using the Theory of Planned behavior as a theoretical framework, this study explores the behavioral intentions of pre-service science teachers in South Africa and Thailand towards integrating AI into their teaching to inform teacher training, support, and resource allocation policies. The main research question is: "What are the behavioral intentions of pre-service science teachers in South Africa and Thailand towards integrating AI into their teaching practices?" The study followed a non-experimental comparative descriptive survey involving 97 South African and 95 Thai final-year BEd students. Data were collected using a structured online questionnaire and analyzed using several statistical tools to compare the TPB constructs between the two samples. South African and Thai pre-service teachers exhibited favorable attitudes and behavioral intentions toward AI integration in teaching. However, Thai students showed significantly higher control and normative beliefs, indicating greater confidence and perceived social support for AI integration than South African students. The findings suggest that targeted teacher training programs and supportive educational policies are essential for enhancing AI readiness, particularly in resource-constrained settings.

{"title":"The behavioural intentions for integrating artificial intelligence in science teaching among pre-service science teachers in South Africa and Thailand","authors":"Lindelani Mnguni , Prasart Nuangchalerm , R. Ahmad Zaky El Islami , Doras Sibanda , Indah Juwita Sari , Moleboheng Ramulumo","doi":"10.1016/j.caeai.2024.100334","DOIUrl":"10.1016/j.caeai.2024.100334","url":null,"abstract":"<div><div>Developing countries in the Global South exhibit diverse trends in the integration of digital technologies, such as Artificial Intelligence in teaching. Complex context-specific factors, including teacher preparedness, influence these trends. Using the Theory of Planned behavior as a theoretical framework, this study explores the behavioral intentions of pre-service science teachers in South Africa and Thailand towards integrating AI into their teaching to inform teacher training, support, and resource allocation policies. The main research question is: \"What are the behavioral intentions of pre-service science teachers in South Africa and Thailand towards integrating AI into their teaching practices?\" The study followed a non-experimental comparative descriptive survey involving 97 South African and 95 Thai final-year BEd students. Data were collected using a structured online questionnaire and analyzed using several statistical tools to compare the TPB constructs between the two samples. South African and Thai pre-service teachers exhibited favorable attitudes and behavioral intentions toward AI integration in teaching. However, Thai students showed significantly higher control and normative beliefs, indicating greater confidence and perceived social support for AI integration than South African students. The findings suggest that targeted teacher training programs and supportive educational policies are essential for enhancing AI readiness, particularly in resource-constrained settings.</div></div>","PeriodicalId":34469,"journal":{"name":"Computers and Education Artificial Intelligence","volume":"7 ","pages":"Article 100334"},"PeriodicalIF":0.0,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143153170","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Explainable artificial intelligence-machine learning models to estimate overall scores in tertiary preparatory general science course 用可解释的人工智能-机器学习模型估算大学预科普通科学课程的总分

Q1 Social Sciences

Computers and Education Artificial Intelligence

Pub Date : 2024-11-20 DOI: 10.1016/j.caeai.2024.100331

Sujan Ghimire , Shahab Abdulla , Lionel P. Joseph , Salvin Prasad , Angela Murphy , Aruna Devi , Prabal Datta Barua , Ravinesh C. Deo , Rajendra Acharya , Zaher Mundher Yaseen

Educational data mining is valuable for uncovering latent relationships in educational settings, particularly for predicting students' academic performance. This study introduces an interpretable hybrid model, optimised through Tree-structured Parzen Estimation (TPE) and Support Vector Regression (SVR), to predict overall scores (OT) utilising five assignments and one examination mark as predictors. Neural Network-based, Tree-Based, Ensemble-Based, and Boosting-based methods are evaluated against the hybrid TPE-optimised SVR model for forecasting final examination grades among 492 students enrolled in the TPP7155 (General Science) course at the University of Southern Queensland, Australia, during the 2020-2021 academic year. Additionally, Local Interpretable Model-agnostic Explanations (LIME) and SHapley Additive explanations (SHAP) techniques are employed to elucidate the inner workings of these prediction models. The findings highlight the superior performance of the proposed model, exhibiting the lowest Root Mean Squared Error (

R M S E

) and Relative Root Mean Squared Error (

R R M S E

), as well as the highest Willmott's index (WI), Legates–McCabe index (LM), and Nash–Sutcliffe Efficiency (NS). With assignment and examination marks identified as pivotal predictors of OT. SHAP and LIME analyses reveal the examination score (ET) as the most influential feature, impacting predicted OT by an average of ±4.93. Conversely, Assignment 1 emerges as the least informative feature, contributing merely ±0.64 to OT predictions. This research underscores the efficacy of the proposed interpretable hybrid TPE-optimised SVR model in discerning relationships among continuous learning variables, thereby empowering educators with early intervention capabilities and enhancing their ability to anticipate student performance prior to course completion.

教育数据挖掘对于揭示教育环境中的潜在关系，尤其是预测学生的学业成绩非常有价值。本研究引入了一个可解释的混合模型，通过树状结构帕尔森估计（TPE）和支持向量回归（SVR）进行优化，利用五门功课和一门考试的分数作为预测因子，预测总分（OT）。在预测澳大利亚南昆士兰大学 2020-2021 学年 TPP7155（普通科学）课程 492 名学生的期末考试成绩时，对基于神经网络、基于树、基于集合和基于提升的方法与混合 TPE 优化 SVR 模型进行了评估。此外，研究还采用了本地可解释模型解释（LIME）和SHapley加性解释（SHAP）技术来阐明这些预测模型的内部运作。研究结果凸显了所提模型的卓越性能，显示出最低的均方根误差（RMSE）和相对均方根误差（RRMSE），以及最高的威尔莫特指数（WI）、莱格茨-麦凯比指数（LM）和纳什-苏特克利夫效率（NS）。作业和考试分数是预测加时赛的关键因素。SHAP 和 LIME 分析显示，考试分数（ET）是影响最大的特征，对预测加时赛的平均影响为 ±4.93。相反，作业 1 是信息量最小的特征，对 OT 预测的影响仅为 ±0.64。这项研究强调了所提出的可解释混合 TPE 优化 SVR 模型在辨别连续学习变量之间关系方面的功效，从而赋予教育工作者早期干预能力，并提高他们在课程完成前预测学生成绩的能力。

{"title":"Explainable artificial intelligence-machine learning models to estimate overall scores in tertiary preparatory general science course","authors":"Sujan Ghimire , Shahab Abdulla , Lionel P. Joseph , Salvin Prasad , Angela Murphy , Aruna Devi , Prabal Datta Barua , Ravinesh C. Deo , Rajendra Acharya , Zaher Mundher Yaseen","doi":"10.1016/j.caeai.2024.100331","DOIUrl":"10.1016/j.caeai.2024.100331","url":null,"abstract":"<div><div>Educational data mining is valuable for uncovering latent relationships in educational settings, particularly for predicting students' academic performance. This study introduces an interpretable hybrid model, optimised through Tree-structured Parzen Estimation (TPE) and Support Vector Regression (SVR), to predict overall scores (OT) utilising five assignments and one examination mark as predictors. Neural Network-based, Tree-Based, Ensemble-Based, and Boosting-based methods are evaluated against the hybrid TPE-optimised SVR model for forecasting final examination grades among 492 students enrolled in the TPP7155 (General Science) course at the University of Southern Queensland, Australia, during the 2020-2021 academic year. Additionally, Local Interpretable Model-agnostic Explanations (LIME) and SHapley Additive explanations (SHAP) techniques are employed to elucidate the inner workings of these prediction models. The findings highlight the superior performance of the proposed model, exhibiting the lowest Root Mean Squared Error (<span><math><mi>R</mi><mi>M</mi><mi>S</mi><mi>E</mi></math></span>) and Relative Root Mean Squared Error (<span><math><mi>R</mi><mi>R</mi><mi>M</mi><mi>S</mi><mi>E</mi></math></span>), as well as the highest Willmott's index (<em>WI</em>), Legates–McCabe index (<em>LM</em>), and Nash–Sutcliffe Efficiency (<em>NS</em>). With assignment and examination marks identified as pivotal predictors of OT. SHAP and LIME analyses reveal the examination score (ET) as the most influential feature, impacting predicted OT by an average of ±4.93. Conversely, Assignment 1 emerges as the least informative feature, contributing merely ±0.64 to OT predictions. This research underscores the efficacy of the proposed interpretable hybrid TPE-optimised SVR model in discerning relationships among continuous learning variables, thereby empowering educators with early intervention capabilities and enhancing their ability to anticipate student performance prior to course completion.</div></div>","PeriodicalId":34469,"journal":{"name":"Computers and Education Artificial Intelligence","volume":"7 ","pages":"Article 100331"},"PeriodicalIF":0.0,"publicationDate":"2024-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142707426","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0