Yun Wen

National Institute of Education (NIE), Nanyang Technological University, Singapore // yun.wen@nie.edu.sg

Yanjie Song

Education University of Hong Kong // ysong@eduhk.hk

ABSTRACT: 

Learning analytics (LA) has been increasingly using in teaching and learning. However, in the field of applied linguistics, the use of LA has only begun to touch the surface. There is a need for understanding how LA and learning design (LD) influence each other and provide useful information to language teachers in the context of specific courses or learning environments. In this study, a retrospective analysis was conducted to identify the factors influencing LA for collaborative language learning in classrooms, from a holistic perspective by integrating LA, LD, and teacher inquiry. The findings suggested that (1) LA focused on interactions can inform pedagogical refinement effectively when LD in language learning is premised on social constructivist theories; (2) LA supported teacher inquiry and LD on condition that the teacher holds innovation-oriented beliefs and the participatory culture between teachers and researchers. The study provided insights into the use of LA in collaborative language learning classrooms that focus on collecting and evaluating learners’ interaction process beyond gleaning linguistic or behavioral facts. Professional development implications and future research are also addressed. 

Keywords: 

Learning analytics, Learning design, Teacher inquiry, Second language learning, Collaborative language learning

Mehmet Kokoç

School of Applied Sciences, Trabzon University, Turkey // kokoc@trabzon.edu.tr 

Mehmet Kara

Vocational School of Design, Amasya University, Turkey // mehmet.kara@amasya.edu.tr

ABSTRACT: 

The purposes of the two studies reported in this research are to adapt and validate the instrument of the Evaluation Framework for Learning Analytics (EFLA) for learners into the Turkish context, and to examine how metacognitive and behavioral factors predict learner performance. Study 1 was conducted with 83 online learners enrolled in a 16-week course delivered through the Moodle learning management system. The findings from the confirmatory factor analysis indicated that a three-factor model of the EFLA for learners provided the best model fit for the collected data. The model is consistent with the factorial structure of the original instrument developed based on the data from the European learners. Study 2 aimed to reveal how the metacognitive and behavioral factors pertaining to the learning analytics dashboard predict learners’ academic performance. A total of 63 online learners enrolled in a 14-week online computing course participated in this study. The results from the logistic regression analysis indicated that online learners more frequently interacted with the learning analytics dashboard demonstrated greater academic performance. Furthermore, the dimensions of the EFLA, together with the interaction with the dashboard, significantly predicted learners’ academic performance. This multiple-study investigation contributes to the generalizability of the EFLA for learners and highlights the importance of metacognitive and behavioral factors for the impact of learning analytics dashboards on learner performance.

Keywords: 

Learning analytics, Learning analytics dashboards, Evaluation, Validation, Learning performance

Sarah D. Matthews

University of Central Florida, College of Engineering and Computer Science, Orlando, FL, USA // Sarah.Matthews@knights.ucf.edu

Michael D. Proctor

University of Central Florida, College of Engineering and Computer Science, Orlando, FL, USA // Michael.Proctor@ucf.edu

ABSTRACT: 

The COVID-19 pandemic hit the United States in 2020 resulting in a public health caseload surge precipitating deployment of military and federal medical units, states issuing emergency orders to engage retired medical professionals, and novice or inadequately trained healthcare workers thrust into service to meet the pressing need. The novelty and scope of the pandemic exposed a gap in the competency and the surge capacity of the public health workforce to address the societal needs during the pandemic. This research investigated the capability of an agent-based, online personalized (AOP) intelligent tutoring system (ITS) that adaptively uses aptitude treatment interaction (ATI) to deliver public health workforce training in a prescribed health regime and assure their competency. This research also considers the ability of such an AOP ITS to support rapidly surging capacity of the public workforce to scale to meet healthcare demands while remaining accessible and flexible enough to adapt to changing healthcare guidance. Findings indicate such a system increases participant performance while providing a high level of acceptance, ease of use by users, and competency assurance. However, discussion of our findings indicates limited potential for an AOP ITS using the current ATI paradigm to make a major contribution to adding public health workforce surge capacity unless workforce members are directed to utilize it and technology barriers in the current public health IT infrastructure are overcome.

Keywords: 

Agent, Intelligent tutoring system, Public health, Surge capacity, Competencies

Biyun Huang

Centre for Learning Sciences and Technologies, The Chinese University of Hong Kong, Hong Kong SAR, China // lucyhuang99@cuhk.edu.hk 

Khe Foon Hew

Faculty of Education, The University of Hong Kong, Hong Kong SAR, China // kfhew@hku.hk 

ABSTRACT: 

A design-based study was conducted in iterative cycles to test the effectiveness of the updated goal-access-feedback-challenge-collaboration (updated-GAFCC) gamification design model. The test-bed was a 10-week undergraduate introductory information management course. Students from three consecutive school years participated in the study, with the control group studying the conventional course without gamification (first year), treatment group_1 studying a gamified course following the original GAFCC model (second year), and treatment group_2 studying an optimized gamified course following the updated-GAFCC model (third year). The results of the design-based study indicated that (i) the updated-GAFCC model and the GAFCC model were effective in enhancing students’ learning achievements and task completion; (ii) the updated-GAFCC model was more successful in generating higher quality thinking artifacts than the GAFCC model; (iii) there were fewer lower-quality submissions in the updated-GAFCC condition than in the GAFCC condition; and (iv) 89% of the interviewed students in the updated-GAFCC condition were satisfied with the overall learning design, and felt that the gamified learning activities facilitated their learning. Overall, the findings contribute to our understanding of how pedagogical strategies can be incorporated into the theory-based design model to optimize learning experiences and academic outcomes.

Keywords: 

Gamification, Design model, Design-based research, Learning performance, Long-term

Yu-Ju Lan

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

Kao Chia-Ling Gupta

University of Hong Kong, China // kclgupta@gmail.com

Tai-Yi Huang

University of North Texas, USA // Tai-yi.huang@unt.edu

Shobhana Chelliah

University of North Texas, USA // Shobhana.Chelliah@unt.edu

J. Michael Spector

University of North Texas, USA // Mike.Spector@unt.edu

ABSTRACT: 

This paper aims at answering the “how” questions about organizing and hosting an online conference during the COVID-19 pandemic. The 3rd International Pan-Pacific Technology-Enhanced Language Learning (PPTELL) Conference and Critical Thinking Meeting (hereafter PPTELL 2020) hosted from June 29 to July 1, 2020, on Zoom is the example conference used in this paper to illustrate the challenges and approaches adopted before, during and after the conference. The mentioned conference was supposed to take place physically at the University of North Texas during the same period but was transformed into an online virtual conference due to the outbreak of COVID-19 in early 2020. It was an urgent decision, along with many unknown situations, such as the attendees’ different time zones and “Zoombombing.” A three-staged and target-action process guided the preparation and organization of the online conference, i.e., pre-, during, and post-conference. According to the live meeting results and the post-conference survey, PPTELL 2020 has earned a reputation from its quality and the satisfaction of the participants and attendees. Therefore, the experience shared in the current paper seems to be a good reference for the organizers and hosts of international conferences.

Keywords: 

Online conference, COVID-19, Zoombombing, PPTELL 2020 and Critical Thinking Meeting

Jiun-Yu Wu 

Institute of Education, National Chiao Tung University, Taiwan, R.O.C. // jiunyu.rms@gmail.com 

Chen-Hsuan Liao

Institute of Education, National Chiao Tung University, Taiwan, R.O.C. // abugu103@gmail.com

 Tzuying Cheng

Institute of Education, National Chiao Tung University, Taiwan, R.O.C. // anke801211@gmail.com 

Mei-Wen Nian

Institute of Education, National Chiao Tung University, Taiwan, R.O.C. // iammo0715@nctu.edu.tw

ABSTRACT: 

Amid the pandemic of coronavirus diseases, virtual conferences have become an alternative way to maintain the prosperity of the research community. This study investigated attendees’ participatory behavior in a virtual academic conference (TWELF2020, Taiwan) and studied the interrelationship among their mastery experience, competence, and engagement to shed light on the development of virtual conferences. Data were collected based on 602 unique IDs via their unstructured trace data and 106 respondents to the post-conference questionnaire. Ten indices were derived from participants’ unstructured log to describe the conference-based and session-based behaviors. Study results demonstrated that virtual conferences could facilitate the extended and deepened participation of the research community, nourish the participant-centered scholarship building, and create an engaging conference environment that reflects quality experiences regarding participants’ mastery experience, competence, and engagement. The implications of the study can inform future virtual conference organization to provide more engaging and rewarding conference experiences for participants of all gender and academic ranks.

Keywords: 

Pandemic, Virtual conference, Mastery experience, Competence, Engagement  

Margus Pedaste

Institute of Education, University of Tartu, Estonia // margus.pedaste@ut.ee

Meriliis Kasemets

Institute of Education, University of Tartu, Estonia // meriliis.kasemets@ut.ee

ABSTRACT: Travel restrictions regarding COVID-19 have created new challenges for organizing international scientific conferences. Most of the conferences have to be moved to a fully online format. In our study, we analyzed what challenges it created in organizing the International Conference on Advanced Learning Technologies and Technology-Enhanced Learning in July 2020 in Estonia at the University of Tartu. The conference had 131 attendees from all over the world – this resulted in significant challenges due to differences in time zones and difficulties in engaging and socializing people in an online event. In our study, we collected feedback on conference-organization related challenges from five local organizing team members with different roles in the team. Their interviews were analyzed using an abductive approach. The results showed that the challenges could be identified through eight main categories: value, management, timing, program, people, protection, scaffolds, and money. In each of them, several sub-categories were specified. It was concluded that there were both advantages and challenges in organizing an online conference compared to a regular one. For example, fewer challenges are related to travel, accommodation, food and drinks, but more attention needs to be paid to supporting the socialization of people, especially those living in different time zones. Another major challenge appeared to be uncertainty related to the conference budget. Significant advantages were that the carbon footprint of the conference was smaller, the conference was more accessible, and it was easier to solve all the technical issues of the participants.

Keywords: Online conference, COVID-19, ICALT, Zoom, Remo, Abductive analysis

Stephen J. H. Yang

National Central University, Taiwan // Stephen.Yang.Ac@gmail.com

ABSTRACT: As addressed by Stephen Yang in his ICCE 2019 keynote speech (Yang, 2019), precision education  is a new challenge when applying artificial intelligence (AI), machine learning, and learning analytics to improve teaching quality and learning performance. The goal of precision education is to identify at-risk students as early as possible and provide timely intervention on the basis of teaching and learning experiences (Lu et al., 2018). Drawing from this main theme of precision education, this special issue advocates an in-depth dialogue between cold technology and warm humanity, in turn offering greater understanding of precision education. For this special issue, thirteen research papers that specialize in precision education, AI, machine learning, and learning analytics to engage in an in-depth research experiences concerning various applications, methods, pedagogical models, and environments were exchanged to achieve better understanding of the application of AI in education. 

Keywords: Precision education, Artificial intelligence, Learning analytics, Human-centered AI 

Dirk Tempelaar

Maastricht University, School of Business and Economics, The Netherlands // d.tempelaar@maastrichtuniversity.nl 

Bart Rienties

The Open University, Institute of Educational Technology, UK // bart.rienties@open.ac.uk 

Quan Nguyen

School of Information, University of Michigan, Ann Arbor, MI, USA // quanngu@umich.edu

ABSTRACT: Precision education requires two equally important conditions: accurate predictions of academic performance based on early observations of the learning process and the availability of relevant educational intervention options. The field of learning analytics (LA) has made important contributions to the realisation of the first condition, especially in the context of blended learning and online learning. Prediction models that use data from institutional information systems and logs of learning management systems have gained a good reputation in predicting underperformance and dropout risk. However, less progress is made in resolving the second condition: applying LA generated feedback to design educational interventions. In our contribution, we make a plea for applying dispositional learning analytics (DLA) to make LA precise and actionable. DLA combines learning data, as in LA, with learners’ disposition data measured through self-report surveys. The advantage of DLA is twofold: first, it improves the accuracy of prediction, specifically early in the module, when limited LMS trace data are available. Second, the main benefit of DLA is in the design of effective interventions: interventions that focus on addressing individual learning dispositions that are less developed but important for being successful in the module. We provide an empirical analysis of DLA in an introductory mathematics module, demonstrating the important role that a broad range of learning dispositions can play in realising precision education.

Keywords: Blended learning, Dispositional learning analytics, Educational intervention, Flipped learning, Precision education

Hui-Chen Lin

Department of Neurology, Tri-Service General Hospital National Defense Medical Center, Taiwan // ceciliatsgh@gmail.com 

Yun-Fang Tu

Department of Library and Information Science, Research and Development Center for Physical Education, Health, and Information Technology, Fu Jen Catholic University, Taiwan // sandy0692@gmail.com 

Gwo-Jen Hwang 

Graduate Institute of Digital Learning and Education, National Taiwan University of Science and Technology, Taiwan // gjhwang.academic@gmail.com

Hsin Huang

Department of Nursing, Tri-Service General Hospital National Defense Medical Center, Taiwan // jhshing1029@gmail.com

ABSTRACT: Precision medicine has become an essential issue in the medical community as the quality of medical care is being emphasized nowadays. The technological data analysis and predictions made by Artificial Intelligence (AI) technologies have assisted medical staff in designing personalized medicine for patients, making AI technologies an important path to precision medicine. During the implementation of the new emerging technology, medical staff’s learning intentions will have a great influence on its effectiveness. With reference to the Technology Acceptance Model, this study explored medical staff’s attitudes, intentions, and relevant influencing factors in relation to AI application learning. A total of 285 valid questionnaires were collected. Five major factors, perceived usefulness (PU), perceived ease of use (PEU), subjective norms (SN), attitude towards AI use (ATU), and behavioral intention (BI), were used for analyzing the AI learning of medical staff in a hospital. Based on the SEM analytical results and the research model, the four endogenous constructs of PU, PEU, SN, and ATU explained 37.4% of the changes in BI. In this model, SN and PEU were the determining factors of BI. The total effects of SN and PEU were 0.448 and 0.408 respectively, followed by PU, with a total effect of 0.244. As a result, the intentions of medical staff to learn to use AI applications to support precision medicine can be predicted by SN, PEU, PU, and ATU. Among them, subjective norms considering the influences of both supervisors and peers, such as encouragement, communication, and sharing, may assist precision education in supporting the learning attitudes and behavior regarding precision medicine. The research results can provide recommendations for examining medical staff’s intention to use AI applications.

Keywords: Artificial intelligence, Subjective norms, Precision medicine, Precision education, Technology Acceptance Model 

Fuzheng Zhao

Kobe University, Japan // zhaofz635@gmail.com

Gwo-Jen Hwang

National Taiwan University of Science and Technology, Taiwan // gjhwang.academic@gmail.com

Chengjiu Yin

Kobe University, Japan // yin@lion.kobe-u.ac.jp

ABSTRACT: Educational data mining and learning analytics have become a very important topic in the field of education technology. Many frameworks have been proposed for learning analytics which make it possible to identify learning behavior patterns or strategies. However, it is difficult to understand the reason why behavior patterns occur and why certain strategies are used. In other words, all of the existing frameworks lack an important step, that is, result confirmation. In this paper, we propose a Result Confirmation-based Learning Behavior Analysis (ReCoLBA) framework, which adds a result confirmation step for exploring the hidden reasons underlying the learning patterns and strategies. Using this ReCoLBA framework, a case study was conducted which analyzed e-book reading data. In the case study, we found that the students had a tendency to delete markers after adding them. Through an investigation, we found that the students did this because they could not grasp the learning emphasis. To apply this finding, we proposed a learning strategy whereby the teacher highlights the learning emphasis before students read the learning materials. An experiment was conducted to examine the effectiveness of this strategy, and we found that it could indeed help students achieve better results, reduce repetitive behaviors and save time. The framework was therefore shown to be effective.

Keywords: Learning analytics, Learning behavior pattern, Learning Analysis framework, Result confirmation 

Christopher C. Y. Yang

Graduate School of Informatics, Kyoto University, Japan // yang.yuan.57e@st.kyoto-u.ac.jp

Irene Y. L. Chen

Department of Accounting, National Changhua University of Education, Taiwan //irene@cc.ncue.edu.tw

Hiroaki Ogata

Academic Center for Computing and Media Studies, Kyoto University, Japan // hiroaki.ogata@gmail.com

ABSTRACT:

 Precision education is now recognized as a new challenge of applying artificial intelligence, machine learning, and learning analytics to improve both learning performance and teaching quality. To promote precision education, digital learning platforms have been widely used to collect educational records of students’ behavior, performance, and other types of interaction. On the other hand, the increasing volume of students’ learning behavioral data in virtual learning environments provides opportunities for mining data on these students’ learning patterns. Accordingly, identifying students’ online learning patterns on various digital learning platforms has drawn the interest of the learning analytics and educational data mining research communities. In this study, the authors applied data analytics methods to examine the learning patterns of students using an ebook system for one semester in an undergraduate course. The authors used a clustering approach to identify subgroups of students with different learning patterns. Several subgroups were identified, and the students’ learning patterns in each subgroup were determined accordingly. In addition, the association between these students’ learning patterns and their learning outcomes from the course was investigated. The findings of this study provide educators opportunities to predict students’ learning outcomes by analyzing their online learning behaviors and providing timely intervention for improving their learning experience, which achieves one of the goals of learning analytics as part of precision education.

Keywords:

Precision education, Learning analytics, Educational data mining, Learning pattern, Ebook learning log

Albert C. M. Yang

Graduate School of Informatics, Kyoto University, Japan // yang.ming.35e@st.kyoto-u.ac.jp

Irene Y. L. Chen

Department of Accounting, National Changhua University of Education, Taiwan // irene@cc.ncue.edu.tw

Brendan Flanagan

Academic Center for Computing and Media Studies, Kyoto University, Japan // flanagan.brendanjohn.4n@kyoto-u.ac.jp

Hiroaki Ogata

Academic Center for Computing and Media Studies, Kyoto University, Japan // hiroaki.ogata@gmail.com

ABSTRACT: 

Precision education is a new challenge in leveraging artificial intelligence, machine learning, and learning analytics to enhance teaching quality and learning performance. To facilitate precision education, text marking skills can be used to determine students’ learning process. Text marking is an essential learning skill in reading. In this study, we proposed a model that leverages the state-of-the-art text summarization technique, Bidirectional Encoder Representations from Transformers (BERT), to calculate the marking score for 130 graduate students enrolled in an accounting course. Then, we applied learning analytics to analyze the correlation between their marking scores and learning performance. We measured students’ self-regulated learning (SRL) and clustered them into four groups based on their marking scores and marking frequencies to examine whether differences in reading skills and text marking influence students’ learning performance and awareness of self-regulation. Consistent with past research, our results did not indicate a strong relationship between marking scores and learning performance. However, high-skill readers who use more marking strategies perform better in learning performance, task strategies, and time management than high-skill readers who use fewer marking strategies. Furthermore, high-skill readers who actively employ marking strategies also achieve superior scores of environment structure, and task strategies in SRL than low-skill readers who are inactive in marking. The findings of this research provide evidence supporting the importance of monitoring and training students’ text marking skill and facilitating precision education.

Keywords: 

Text summarization, Marker grading, Self-regulated learning, Precision education, Text marking

Xuanqi Feng

Graduate School of Human-Environment Studies, Kyushu University, Japan // dukefeng@mark-lab.net 

Masanori Yamada

Faculty of Arts and Science, Kyushu University, Japan // mark@mark-lab.net 

ABSTRACT: 

It is challenging to utilize learning analytic technologies to examine gameplay log data for game-embedded assessment in the field of game-based learning. Analytical approaches based on a new perspective focusing on complicated contextual data are imperative in the current scenario. A relatively new concept called precision education, which focuses on individual learning and provides personalized and timely intervention to different learners, can be regarded as a new perspective for game-based learning. Additionally, the order of knowledge acquisition in the learning environment is a kind of learning path extracted from the contextual information of in-game behavior logs. Therefore, in this study, the authors propose a new analytical approach to identify learning path patterns and elucidate the features of these patterns for an educational game they developed. The statistical analysis shows that learners with diverse learning path patterns have different learning effects, suggesting that the learning path is an important factor in precision education. The practice of using the explanation method to examine the proposed approach can help us understand learners’ knowledge acquisition and provide evidence for enhancing the accuracy of precision education and improving the quality of the educational game. The findings are expected to contribute to both game-based learning and precision education.

Keywords: 

Learning analytics, Game-based learning, Learning path, Informal learning, Precision education

Feifei Han

Office of Pro-Vice-Chancellor (Arts, Education and Law), Griffith University, Australia // Griffith Institute for Educational Research, Griffith University // feifei.han@griffith.edu.au

Robert A. Ellis

Office of Pro-Vice-Chancellor (Arts, Education and Law), Griffith University, Australia // r.ellis@griffith.edu.au

ABSTRACT: 

One of the major objectives of precision education is to improve prediction of educational outcome. This study combined theory-driven and data-driven approaches to address the limitations of current practice of predicting learning outcomes only using a single approach. The study identified the online learning patterns by using students’ self-reported approaches and perceptions of online learning and by using the observational digital traces of the sequences of their online learning events in a blended course. The study examined predictions of the academic performance using the online learning patterns generated by the two approaches separately. It also investigated the extent to which the online learning patterns identified by the two approaches were associated with each other. The theory-driven approach adopted a hierarchical cluster analysis using the self-reported data and found a ‘deep’ and a ‘surface’ online learning patterns, which were related to differences in the academic performance. The data-driven approach used an agglomerative sequence clustering and detected four patterns of online learning, which not only differed by quantity (number of learning events), but also differed by quality (the proportions of types of learning events). A one-way ANOVA revealed that the online learning pattern which had the most learning events, and was characterized by high proportions of viewing course contents and of performing problem-solving exercises, had the highest academic performance. A cross-tabulation revealed significant association between the self-reported and observational online learning patterns, demonstrating consistency of the evidence by a theory-driven and a data-driven approach and triangulating the results of the two approaches.

Keywords:

Online learning patterns, Academic performance, Theory-driven approaches, Data-driven approaches, Blended course designs

Xieling Chen

Department of Mathematics and Information Technology, The Education University of Hong Kong, Hong Kong SAR // xielingchen0708@gmail.com

Di Zou

Department of English Language Education, The Education University of Hong Kong, Hong Kong SAR // dizoudaisy@gmail.com

Haoran Xie 

Department of Computing and Decision Sciences, Lingnan University, Hong Kong SAR // hrxie2@gmail.com

Gary Cheng

Department of Mathematics and Information Technology, The Education University of Hong Kong, Hong Kong SAR // chengks@eduhk.hk

ABSTRACT: 

Personalized language learning (PLL), a popular approach to precision language education, plays an increasingly essential role in effective language education to meet diverse learner needs and expectations. Research on PLL has become an active sub-field of research on technology-enhanced language learning and artificial intelligence applications in education. Based on the PLL literature from the Web of Science and Scopus databases, this study identified trends and prominent research issues within the field from 2000 to 2019 using structural topic modeling and bibliometrics. Trend analysis of articles demonstrated increasing interest in PLL research. Journals such as Educational Technology & Society and Computers & Education had contributed much to PLL research. PLL associated closely with mobile learning, game-based learning, and online/web-based learning. Moreover, personalized feedback and recommendations were important issues in PLL. Additionally, there was an increasing interest in adopting learning analytics and artificial intelligence in PLL research. Results obtained could help practitioners and scholars better understand the trends and status of PLL research and become aware of the hot topics and future directions.

Keywords: 

Personalized language learning, Topic modeling, Knowledge mapping, Bibliometrics, Precision education

Mehmet Kokoç

School of Applied Sciences, Trabzon University, Turkey // kokoc@trabzon.edu.tr

Gökhan Akçapınar

Faculty of Education, Hacettepe University, Turkey // gokhana@hacettepe.edu.tr 

Mohammad Nehal Hasnine

Research Center for Computing and Multimedia Studies, Hosei University, Japan // nehal.hasnine.79@hosei.ac.jp

ABSTRACT: 

This study analyzed students’ online assignment submission behaviors from the perspectives of temporal learning analytics. This study aimed to model the time-dependent changes in the assignment submission behavior of university students by employing various machine learning methods. Precisely, clustering, Markov Chains, and association rule mining analysis were used to analyze students’ assignment submission behaviors in an online learning environment. The results revealed that students displayed similar patterns in terms of assignment submission behavior. Moreover, it was observed that students’ assignment submission behavior did not change much across the semester. When these results are analyzed together with the students’ academic performance at the end of the semester, it was observed that students’ end-of-term academic performance can be predicted from their assignment submission behaviors at the beginning of the semester. Our results, within the scope of precision education, can be used to diagnose and predict students who are not going to submit the next assignments as the semester progresses as well as students who are going to fail at the end of the semester. Therefore, learning analytics interventions can be designed based on these results to prevent possible academic failures. Furthermore, the findings of the study are discussed considering the development of early-warning intervention systems for at-risk students and precision education.

Keywords: 

Precision education, Temporal learning analytics, Educational data mining, Assignment submission behavior, Learning performance

Alwyn Vwen Yen Lee

National Institute of Education, Nanyang Technological University, Singapore // alwyn.lee@nie.edu.sg 

ABSTRACT: 

The understanding of online classroom talk is a challenge even with current technological advancements. To determine the quality of ideas in classroom talk for individual and groups of students, a new approach such as precision education will be needed to integrate learning analytics and machine learning techniques to improve the quality of teaching and cater interventive practices for individuals based on best available evidence. This paper presents a study of 20 secondary school students engaged in asynchronous online discourse over a period of two weeks. The online discourse was recorded and classroom talk was coded before undergoing social network analysis and k-means clustering to identify three types of ideas (promising, potential, and trivial). The quality and distribution of ideas were then mapped to the different kinds of talk that were coded from the online discourse. Idea Progress Reports were designed and trialed to present collective and individual student’s idea trajectories during discourse. Findings show that the majority of ideas in exploratory talk are promising to the students, while ideas in cumulative and disputational talks are less promising or trivial. Feedback on the design of the Idea Progress Reports was collected with suggestions for it to be more informative and insightful for individual student. Overall, this research has shown that classroom talk can be associated with the quality of ideas using a quantitative approach and teachers can be adequately informed about collective and individual ideas in classroom talks to provide timely interventions.

Keywords: 

Precision education, Machine learning, Learning analytics, Idea Identification and Analysis (I2A), Idea Progress Reports (IPR)

Hui Luan

Institute for Research Excellence in Learning Sciences, National Taiwan Normal University, Taiwan // hluanv@gmail.com

Chin-Chung Tsai

Institute for Research Excellence in Learning Sciences, National Taiwan Normal University, Taiwan // Program of Learning Sciences, National Taiwan Normal University, Taiwan // tsaicc@ntnu.edu.tw

ABSTRACT: 

In recent years, in the field of education, there has been a clear progressive trend toward precision education. As a rapidly evolving AI technique, machine learning is viewed as an important means to realize it. In this paper, we systematically review 40 empirical studies regarding machine-learning-based precision education. The results showed that the majority of studies focused on the prediction of learning performance or dropouts, and were carried out in online or blended learning environments among university students majoring in computer science or STEM, whereas the data sources were divergent. The commonly used machine learning algorithms, evaluation methods, and validation approaches are presented. The emerging issues and future directions are discussed accordingly.

Keywords:

Precision education, Personalized learning, Individualized learning, Machine learning, Individual differences

Jiun-Yu Wu

Institute of Education, National Chiao Tung University, Taiwan, R.O.C. // jiunyu.rms@gmail.com

Christopher C. Y. Yang

Graduate School of Informatics, Kyoto University, Japan // yang.yuan.57e@st.kyoto-u.ac.jp

Chen-Hsuan Liao

Institute of Education, National Chiao Tung University, Taiwan, R.O.C. // abugu103@gmail.com

Mei-Wen Nian

Institute of Education, National Chiao Tung University, Taiwan, R.O.C. // anke801211@gmail.com

ABSTRACT: 

This methodological-theoretical synergy provides an integrative framework of learning analytics through the development of the human-and-machine symbiotic reinforcement learning. The framework intends to address the challenges of the current learning analytics model, including a lack of internal validity, generalizability, immediacy, transferability, and interpretability for precision education. The proposed framework consists of a master component (the brain) and its four subsuming components: social networking, the smart classroom, the intelligent agent, and the dashboard. The brain component takes in and analyzes multimodal streams of student data from the other components with the model-based reinforcement learning, which forms policies of adequate actions that maximize the long-term rewards for both the human and machine in the seamless learning environment. An example case plan in advanced statistics was demonstrated to illustrate the course description, data collected in each component, and how the components meet different features of the smart learning environment to deliver precision education. An empirical demonstration was provided using some selected mulitmodal data to inform the effectiveness of the proposed framework. The human-and-machine symbiotic reinforcement learning has theoretical and practical implications for the next-generation learning analytics models and research.

Keywords: 

Reinforcement learning, Learning analytics, Symbiotic learning, Smart learning environment, Precision education

Tzu-Chi Yang

Department of Mathematics and Information Education, National Taipei University of Education, Taiwan (R.O.C.)// Institute of Education/Center for Teacher Education, National Chiao Tung University, Taiwan (R.O.C.) // tcyang.academic@gmail.com 

Yih-Lan Liu

Institute of Education/Center for Teacher Education, National Chiao Tung University, Taiwan (R.O.C.) // ylliu@mail.nctu.edu.tw 

Li-Chun Wang

Department of Electrical Engineering, National Chiao Tung University, Taiwan (R.O.C.) // lichun@g2.nctu.edu.tw

ABSTRACT: 

The recently increased importance of practicing precision education has attracted much attention. To better understand students’ learning and the relationship between their individual differences and learning outcomes, the bird-eye view possible for educational policymakers and stakeholders from educational data mining and institutional research has gained importance and momentum. The deployment of specific predictive tasks based on institutional data is the most promising solution for dealing with a variety of issues on precision education. Most research in this field is focused on learning performance and related issues, such as at-risk students and drop-out tendencies. Seldom are the relationships between the learning performance and career decisions of students investigated. However, developing a deep understanding of students’ career decisions plays an important role in the practice of precision education. In this vein, this paper provides a comprehensive analysis and comparison of the state of the art of predictive techniques for providing a prediction for students’ career decisions. The results indicated that it is possible to perform early detection of students’ career decisions. The contributions of this study are discussed in terms of their implications for theory, methodology, and application.

Keywords:

Precision education, Institutional Research, Students’ life planning, Machine learning, Educational data mining