Use Of Learning Analytics Research Articles

Promoting higher education students’ self-regulated learning through learning analytics: A qualitative study

AbstractLearning analytics provides a novel means to support the development and growth of students into self-regulated learners, but little is known about student perspectives on its utilization. To address this gap, the present study proposed the following research question: what are the perceptions of higher education students on the utilization of a learning analytics dashboard to promote self-regulated learning? More specifically, this can be expressed via the following threefold sub-question: how do higher education students perceive the use of a learning analytics dashboard and its development as promoting the (1) forethought, (2) performance, and (3) reflection phase processes of self-regulated learning? Data for the study were collected from students (N = 16) through semi-structured interviews and analyzed using a qualitative content analysis. Results indicated that the students perceived the use of the learning analytics dashboard as an opportunity for versatile learning support, providing them with a means to control and observe their studies and learning, while facilitating various performance phase processes. Insights from the analytics data could also be used in targeting the students’ development areas as well as in reflecting on their studies and learning, both individually and jointly with their educators, thus contributing to the activities of forethought and reflection phases. However, in order for the learning analytics dashboard to serve students more profoundly across myriad studies, its further development was deemed necessary. The findings of this investigation emphasize the need to integrate the use and development of learning analytics into versatile learning processes and mechanisms of comprehensive support and guidance.

A Comprehensive Bibliometric Analysis of Learning Analytics in Education Research

The purpose of this study is to present data on the use of learning analytics in the field of education. For this purpose, data available in the Web of Science database were examined using bibliometric analysis. The publication distribution varied between 2011 and 2022, with the number of publications reaching its peak in 2020 and declining thereafter. Findings from co-authorship analysis indicate that Dragan Gašević is the most prolific author and the most cited researcher in the field of learning analytics. Results of co-authorship analysis by country suggest that researchers affiliated with institutions in Australia engage in more collaborative work and cooperation. Results of co-authorship analysis by organizations demonstrate that Monash University in Australia is the organization with the highest level of collaboration with other organizations. The most cited work is Ferguson's (2012) article titled "Learning analytics: drivers, developments and challenges." The most frequently used keywords in conjunction with learning analytics are higher education, educational data mining, online learning, learning design, and self-regulated learning. In recent years, there has been a focus on learning analytics related to learning management systems (LMS) rather than MOOCs in learning environments. The British Journal of Educational Technology, which was established in 1970, is the most cited journal that is commonly referenced in the articles. It is hoped that this study will serve as a valuable reference for researchers interested in the field and guide them in making accurate inferences based on a broad perspective.

Privacy versus pedagogy – students’ perceptions of using learning analytics in higher education

The rapid recent use of learning analytics (LA) in higher education, specifically during the COVID-19 pandemic, allows the monitoring of users' behavior while learning. Using LA may promote students' learning outcomes but also intrude into their privacy. This study aimed to explore students’ behaviour and perceptions towards privacy and data protection when using LA for pedagogical needs, examine the privacy trade-off of students' willingness to share personal information in exchange for pedagogical benefits and understand the predicting variables for this privacy trade-off. A model was developed containing five groups of influencing variables (demographic data, perceptions, feelings, behaviour and awareness) on the privacy trade-off. A total of 1,014 students completed an online questionnaire. The results found that students do care about their privacy but are not aware of privacy and data protection regulations. They are willing to trade off privacy for pedagogical benefits, and they trust their academic institutions, but they want transparency. Age, a sense of security in the academic institution, behaviour, data misuse concern and institution management of students' personal information are the significant predictors for a privacy trade-off. It is important to engage students in the process as they are the main beneficiaries of LA and build trust between them and the institution. Implications for practice or policy: Academic institution should actively provide information to raise privacy awareness. Academic institutions should engage students in the process of using LA to create a high degree of trust. Universities should endorse LA policy, fostering it for pedagogical purposes. Instructors may utilise LA to enrich student learning, respecting their willingness to share pedagogical data. Academic institutions should provide a high level of transparency in order to build students’ trust in their institutions.

Teachers’ Formative Assessment Practices in Their Mathematics Classroom Using Learning Analytics Visualizations

This study explores how the use of learning analytics (LA) visualizations to eliciting examples tasks (EET) on the topic of functions can shape middle school mathematics teachers’ formative assessment (FA) practices. Teachers’ practices were examined when provided with LA visualizations, offering them an interactive analysis of their students’ work. The findings showed that LA visualizations enhanced teachers’ evaluations of their students’ work, including challenges and strengths. Two key areas of FA were supported when teachers interacted with LA: designing and implementing classroom discussions and other learning activities and enhancing the quality of feedback. The patterns of use indicated that when teachers were not convinced of the data presented, they used critical thinking and accessed other reports and interactive tools to confirm their impressions.

Learning analytics in distance education: A systematic review study

AbstractThis study aims to reveal the main trends and findings of the studies examining the use of learning analytics in distance education. For this purpose, journal articles indexed in the SSCI index in the Web of Science database were reviewed, and a total of 400 journal articles were analysed within the scope of this study. The systematic review method was used as the research methodology. The first study on the subject was published in 2011, and the publications continued to increase over the years. It was found that the publications on the subject were primarily found in “Computers and Education” and “Education and Information Technologies” journals. It was observed that China, USA, and Spain were the leading countries where the related studies were conducted. The studies primarily used the quantitative method, and university students were included as the sample. In addition, within the scope of learning analytics, it was observed that the data were mainly analysed with “regression analysis”, “correlation analysis”, “special algorithms-models”, “ANOVA”, and “cluster analysis” methods. It was determined that the most preferred platforms in the studies were learning management systems and MOOCs, learning behaviours were mostly examined, and log data were mainly used in this process. It was observed that the variables tested in the studies mainly consisted of students’ behaviours on the platform, learning performances, communication processes, dropout behaviours and course designs. Furthermore, in the studies examined, the advantages of learning analytics in the context of distance education are mostly related to the possibilities of improving the teaching process, and as disadvantages, it is stated that learning analytics is not suitable for use in some situations, negatively affect students’ performances, have limited interaction with students and are an expensive investment.

Defining “data” in conversations with students about the ethical use of learning analytics

In any conversation about the development of ethical standards for practice, it is vital that all stakeholders have a shared understanding of the main concepts in order to reach agreement. In the context of higher education and learning analytics, while many conversations are underway, it is less clear that such a shared understanding exists around the concept of “data”. In order to understand this situation more fully we conducted a study to investigate students’ perceptions of the ethical and privacy considerations related to the data that universities collect and use about them for the purposes of learning analytics. In this paper, we focus specifically on the understandings students have of the types of data that can be collected about them within the educational environment. The outcomes showed that there was a diversity of understandings, but that five main data types emerged. In developing a better understanding of the ways students understand data, it can assist institutions to have more effective conversations with students about the ethical use of learning analytics.

Review of Literature on the Use of Learning Analytics and Learning Analytical Dashboard (LAD) in Improving Student Performance in Higher Education Institutions in Kenya

Purpose – The purpose of the paper is mainly to review the various literature done on Learning Analytics in higher education institutions in Kenya. From the review it aims at identifying the current status of Learning Analytics in Kenya and proposing recommendations for improvement of the same. Methodology – The paper uses desktop approach to breakdown the various studies done in Kenya on Learning Analytics in higher education institutions in Kenya and identifies any research gaps/ areas for improvement from the review of literature. Value – The analysis is valuable to Higher Education institutions in Kenya in coming up with a model that will be used to improve student performance in blended learning through use of learning analytics and learning analytics dashboard. Findings – the findings of the study revealed that there was limited research on learning analytics in Kenya. Moreover, the current study had not used raw data such as the behavioral pattern of students. Therefore, future study can be done using the raw data from Moodle to develop a model for improving student performance.

Innovation of Teaching Methods of Civic and Political Education in Colleges and Universities with the Assistance of Artificial Intelligence

Abstract Artificial intelligence empowers education and promotes its integration and innovation, enabling the use of learning analytics to accurately diagnose the learning situation and implement precise teaching. In this paper, we have leveraged artificial intelligence to innovate traditional teaching methods, developed a wisdom teaching model specifically for college civic education, and concentrated on the precise application of wisdom teaching in this context. First, it creates student profiles based on academic data and then provides personalized recommendations for learning resources by combining a collaborative filtering algorithm and an improved K-means algorithm. Finally, we analyze the actual impact of the teaching intervention by comparing the student portraits before and after the intervention. When the number of neighbors K takes the same value, the improved cosine similarity calculation method outperforms other calculation methods, and the RMSE, MAE, accuracy, recall, and F1 values are 0.14, 0.24, 0.99, 0.74, and 0.85 when K = 60, respectively, which indicates that the algorithm improvement in this paper is practical. This paper categorizes the sample students into four groups based on their learning behavior and outcomes: potential, excellent, marginal, and hardworking. All these student types have demonstrated some progress following the teaching intervention, confirming the efficacy of this paper’s artificial intelligence-based teaching approach for college civic education. Simultaneously, the created profiles of these student types establish the groundwork for future personalized and accurate instruction.

Transforming the Culture and Innovating the Practices of Digital Assessment: How MOOCs Can Scale Up the Impact

The emerging new learning scenarios made possible by the rapid dissemination of adaptive digital technologies are deeply changing higher education processes. They provide educators and learners with new opportunities for transforming their educational practices. encouraging learning innovation and challenging traditional teaching methods. This transformation is particularly significant in how learning achievement is measured, assessed, and certified. The use of learning analytics and adaptive digital technologies not only contribute to consolidate an assessment for learning approach as call for it. In the face of this challenge, educators must speedily adapt their practices to ensure that they are effectively using digital tools to support this new assessment culture. In the framework of the Erasmus+ project PROF-XXI, a Massive Open Online Course (MOOC) entitled "The 21st Century Teacher: Learning to Teach with Digital Technologies" was designed and delivered to facilitate teachers’ access to good practices in this field. The MOOC included four modules covering various aspects of digital education and was designed taking as a reference the competences defined in the DigCompEdu framework. It was delivered on an Open edX instance. A total of 5,660 participants, mainly from Latin America, took this course in two cohorts, with 681 participants completing it. In this paper we describe the experience of designing a learning module specifically dedicated to digital assessment innovation for this MOOC, The aim was to inspire educational change and promote a more sustainable approach to assessment for learning based on the PrACT framework. The main results of this experience are presented and discussed, as well as lessons learnt.

Learning Analytics: A View on the Design and Assessment of Asynchronous Online Discussions for Better Teaching Performance

In recent years, the impact of learning analytics has been investigated and explored in higher education contexts. This article aims to show how their application in online educational contexts is providing great support for teaching performance, especially in relation to the methodology applied, the monitoring of students’ interactions and participation, and the evaluation of activities, and how it can favor improvements in student performance and satisfaction. We therefore present the results obtained from use of the DIANA (DIAlog ANAlysis) tool designed for the research project “Use of learning analytics in digital environments: impact on the improvement of university teaching practice” (LAxDigTeach-21085GE). This study followed a mixed methodology (qualitative and quantitative) to better complement the data provided by learning analytics, given that the numerical data must be meaningful in the specific context in which they are collected. The results obtained are in line with previous research and show that the use of learning analytics have helped to improve teaching performance in relation to monitoring student interactions, participation, and evaluation, but are limited in terms of improving their performance and satisfaction. No generalized conclusions can be drawn as yet in light of the fact that the research project of which this study is a part has recently completed the pilot stage and we have only analyzed the information obtained in one of the participating subjects.

Learning analytics in formative assessment: A systematic literature review

This systematic review examines the use of learning analytics (LA) in formative assessment (FA). LA is a powerful tool that can support FA by providing real-time feedback to students and teachers. The review analyzes studies published on Web of Science and Scopus databases between 2011 and 2022 that provide an overview of the current state of published research on the use of LA for FA in diverse learning environments and through different delivery modes. This review also explores the significant potential of LA in FA practices in digital learning. A total of 63 studies met all selection criteria and were fully reviewed by conducting multiple analyses including selected bibliometrics, a categorical meta-trends analysis and inductive content analysis. The results indicate that the number of LA in FA studies has experienced a significant surge over the past decade. The results also show the current state of research on LA in FA, through a range of disciplines, journals, research methods, learning environments and delivery modes. This review can help inform the implementation of LA in educational contexts to support effective FA practices. However, the review also highlights the need for further research.

Learning analytics to explore the motivational profiles of non-traditional practical nurse students: a mixed-methods approach

Learning analytics provides a novel means to examine various aspects of students’ learning and to support them in their individual endeavors. The purpose of this study was to explore the potential of learning analytics to provide insights into non-traditional, vocational practical nurse students’ (N = 132) motivational profiles for choosing their studies, using a mixed-methods approach. Non-traditional students were somewhat older learners than those following a more straightforward educational pathway and had diverse educational or professional backgrounds. Institutional admission data and analytics were used to identify their specific study motives and distinct motivational profiles, and to illustrate the connections between the motives emerging in the motivational profiles. Furthermore, the association between the motivational profiles and study performance was examined. The results of qualitative content analysis indicated that non-traditional practical nurse students pursued such specialized training for various reasons, and that pragmatic, professional rationales were emphasized over prosocial, altruistic factors. Through the adoption of person-centered latent class analysis, three motivational profiles were identified: self-aware goal-achievers, qualification attainers, and widely oriented humanitarians. Additionally, the analyses of epistemic networks for the profiles showed the complex interplay between the motives, confirming that some motive connections appear to be more prominent than others. Moreover, the findings indicated that study motives reported at admission did not seem to dictate students’ later study performance, as no statistically significant associations were found between the motivational profile and the students’ final grade point average or study dropout. This investigation paves the way for more-targeted motivational support and the use of learning analytics in the context of vocational education and training.

Learning Analytics for Data-Driven Decision Making

This study examines the use of learning analytics to enhance instructional personalization and student engagement in online higher education. The research focuses on the engagement levels of students based on different access methods (mobile and non-mobile), the relationships among engagement indicators, and the strategies for instructional personalization. Quantitative research methodology is employed to analyse and measure students' engagement levels. The findings indicate that students using non-mobile devices exhibit higher engagement in terms of average minutes, item accesses, and content accesses, while mobile access shows higher engagement in terms of course accesses, course interactions, and average interactions. Significant correlations are observed among engagement indicators, highlighting the importance of course interactions, content accesses, and assessment accesses in promoting student engagement. Accordingly, a critical model for effective student engagement in online learning courses is proposed.

Transforming Learning Support in ODFL: The LESS model

The purpose of this paper is to share insights gained from the discovery, design, and delivery phases of creating a three-tiered model of non-academic learning support in ODFL, “Learner Engagement & Success Services (LESS),” at Open Polytechnic | Te Pūkenga, New Zealand. Presented as a case-study, this paper discusses the early vision of the model, examines theoretical understandings related to learner support and engagement, and describes the successes and challenges faced in bringing this vision to reality. It outlines challenges related to the emergent use of learning analytics (LA) in identifying “exception” learners, the contestability of ethical use and choices of data, and the rapid evolution and devolution of commercially available solution tools. The tiered model is described as a scalable blend of human and technology enabled interventions and services, underpinned by the values of agency and equity. The component parts of this model are replicable, and its success is measurable.

Learning analytics for online game-Based learning: a systematic literature review

ABSTRACT Game-based learning researchers have been investigating various means to maximise learning in educational games. One promising venue in recent years has been the use of learning analytics in online game-based learning environments. However, little is known about how different elements of learning analytics (e.g. data types, techniques methods, and stakeholders) contribute to game-based learning practices within online learning environments. There is a need for a comprehensive review to bridge this gap. In this systematic review, we examined the related literature in five major international databases including Web of Science, Scopus, ERIC, IEEE, and compiled Proceedings of the International Conference on Learning Analytics and Knowledge. Twenty relevant publications were identified and analysed. The analysis was conducted using four core elements of learning analytics, namely the types of data that the system collects (what), the methods used for performing analytics (how), the reasons the system captures, analyzes, and reports data (why), and the recipients of the analytics (who). This study synthesises the existing literature, provides a conceptual framework as to how learning analytics can enhance online game-based learning practices in higher education, and sets the agenda for future research.

TEACHERS' DATA LITERACY SKILLS FOR PEDAGOGICAL DECISION MAKING: NEEDS ANALYSIS IN LITHUANIA AND GERMANY

The purpose of the article is to analyse the needs of general education schoolteachers’ data literacy skills that are important for the effective use of learning analytics in the teaching-learning process. The theoretical part of the article presents the idea of big data in education, highlights the aspects of pedagogical value of learning analytics technologies, provides the overview of learning analytic tools. Some overview and comparison of spread of learning analytics tools in general education schools in Lithuania and Germany is presented in the context of data-driven education. The empirical part of the article presents some results from a big qualitative study of teachers’ experiences applying learning analytics tools in teaching - learning process. The main question of the current research is what data literacy skills teachers need in order to use learning analytics tools and make data based pedagogical decisions. Semi-structured interviews were conducted with 10 Lithuanian and 9 German teachers from general education schools, who already have had experience in working with learning experience platforms (digital learning platforms that integrate learning analytics tools). Interview data were analysed by means of content analysis. The results of the qualitative study showed that in order to use learning analytics tools it is important for teachers to have such skills as: digital literacy, data collection, data analysis and interpretation, etc. Comparative analysis of informants’ answers showed that teachers in Lithuania and Germany expressed similar needs for data literacy skills.

Learning Analytics on Student Engagement to Enhance Students’ Learning Performance: A Systematic Review

The study of learning analytics provides statistical analysis and extract insights from data, particularly in education. Various studies regarding student engagement in online learning have been conducted at tertiary institutions to verify its effects on students’ learning performance. However, there exists a knowledge gap whereby the types of student-engagement issues derived from learning analytics have not been collectively studied thus far. In order to bridge the knowledge gap, this paper engages a new systematic literature review (SLR) that analysed 42 articles using Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). The existing research on student engagement in online learning does not extensively integrate the five types of online engagement proposed by Redmond et al., and the use of learning analytics on the subject matter is also limited. Thus, this review sheds light on the types of student engagement indicated by using learning analytics, hoping to enhance students’ learning performance in online learning. As revealed in the findings, some studies measured multifaceted engagement to enhance students’ learning performance, but they are limited in number. Thus, it is recommended that future research incorporate multifaceted engagement such as social, cognitive, collaborative, behavioural, and emotional engagement in online learning and utilise learning analytics to improve students’ learning performance. This review could serve as the basis for future research in online higher education.

Amplifying Student and Administrator Perspectives on Equity and Bias in Learning Analytics

When higher education institutions (HEIs) have the potential to collect large amounts of learner data, it is important to consider the spectrum of stakeholders involved with and impacted by the use of learning analytics. This qualitative research study aims to understand the degree of concern with issues of bias and equity in the uses of learner data as perceived by students, diversity and inclusion leaders, and senior administrative leaders in HEIs. An interview study was designed to investigate stakeholder voices that generate, collect, and utilize learning analytics from eight HEIs in the United States. A phased inductive coding analysis revealed similarities and differences in the three stakeholder groups regarding concerns about bias and equity in the uses of learner data. The study findings suggest that stakeholders have varying degrees of data literacy, thus creating conditions of inequality and bias in learning data. By centring the values of these critical stakeholder groups and acknowledging that intersections and hierarchies of power are critical to authentic inclusion, this study provides additional insight into proactive measures that institutions could take to improve equity, transparency, and accountability in their responsible learning analytics efforts.

Transparency and Trustworthiness in User Intentions to Follow Career Recommendations from a Learning Analytics Tool

Transparency and trustworthiness are among the key requirements for the ethical use of learning analytics (LA) and artificial intelligence (AI) in the context of social inclusion and equity. However, research on these issues pertaining to users is lacking, leaving it unclear as to how transparent and trustworthy current LA tools are for their users and how perceptions of these variables relate to user behaviour. In this study, we investigate user experiences of an LA tool in the context of career guidance, which plays a crucial role in supporting nonlinear career pathways for individuals. We review the ethical challenges of big data, AI, and LA in connection to career guidance and analyze the user experiences (N = 106) of the LA career guidance tool, which recommends study programs and institutions to users. Results indicate that the LA career guidance tool was evaluated as trustworthy but not transparent. Accuracy was found to be a stronger predictor for the intention to follow on the recommendations of the LA guidance tool than was understanding the origins of the recommendation. The user’s age emerged as an important factor in their assessment of transparency. We discuss the implications of these findings and suggest emphasizing accuracy in the development of LA tools for career guidance.

Transparency and Trustworthiness in User Intentions to Follow Career Recommendations from a Learning Analytics Tool

Transparency and trustworthiness are among the key requirements for the ethical use of learning analytics (LA) and artificial intelligence (AI) in the context of social inclusion and equity. However, research on these issues pertaining to users is lacking, leaving it unclear as to how transparent and trustworthy current LA tools are for their users and how perceptions of these variables relate to user behaviour. In this study, we investigate user experiences of an LA tool in the context of career guidance, which plays a crucial role in supporting nonlinear career pathways for individuals. We review the ethical challenges of big data, AI, and LA in connection to career guidance and analyze the user experiences (N = 106) of the LA career guidance tool, which recommends study programs and institutions to users. Results indicate that the LA career guidance tool was evaluated as trustworthy but not transparent. Accuracy was found to be a stronger predictor for the intention to follow on the recommendations of the LA guidance tool than was understanding the origins of the recommendation. The user’s age emerged as an important factor in their assessment of transparency. We discuss the implications of these findings and suggest emphasizing accuracy in the development of LA tools for career guidance.