Learner Behavior Research Articles

Dropout is not always a failure! Exploration on the prior knowledge and learning behaviors of MOOC learners

Dropout is not always a failure! Exploration on the prior knowledge and learning behaviors of MOOC learners

The digital fingerprint of learner behavior: Empirical evidence for individuality in learning using deep learning

Personalized learning builds upon the fundamental assumption of uniqueness in learning behavior, often taken for granted. Quite surprisingly, however, the literature provides little to no empirical evidence backing the existence of individual learning behaviors. Driven by curiosity, we challenge this axiom. Our operationalization of a unique learning behavior draws an analogy to a fingerprint – a distinctive trait that sets individuals apart, which we correspondingly termed the ‘Digital Fingerprint of Learner Behavior’ (DFL). If such a thing as DFL truly exists, then given enough fine-grained behavioral data, we argue that it should be possible to model a DFL to a level of discriminability that enables training machine learning models to associate (map) between the (de-identified) digital traces of the same learner in diverse contexts. To test our hypothesis, we experimented with data from 24 MITx massive open online courses (MOOCs) offered via edX between 2014 and 2017. We focused our investigation on contexts where both the content and platform remain constant. A learner's DFL was computed from the learner's activity data within a specific course chapter, as stored in the system's logs. The results show that the mean level of accuracy (across courses) in identifying unseen DFLs is 0.582 (SD=0.173). Using Shapley Additive exPlanations (SHAP), we rank 686 features for their importance in differentiating between DFLs. To the best of our knowledge, this study is the first to provide empirical evidence that learners' behavior is unique to a degree that can distinguish between them on an individual level, similar to the level of identification provided by a fingerprint, and sets a benchmark for the task of DFL identification.

Examining beginner Spanish learners' interactions during text chat: Self‐monitoring and the impact of metacognitive instruction

AbstractText chat facilitates L2 use by providing learners with extended time to plan, monitor, and process production during interactional tasks. However, learners may not naturally take advantage of these affordances, especially for providing peer feedback. This study used video‐enhanced chat scripts to examine the behavior of beginner L2 Spanish learners (n = 46) completing decision‐consensus tasks via text chat, investigating how learners interacted with the medium to self‐monitor by deleting text before sending. Learners were also differentially assigned to receive metacognitive instruction (MI) to learn ways to provide corrective feedback to peers or engage in language‐related episodes (LREs). Results showed that all learners frequently used text chat's affordances to self‐correct and also integrated technology to resolve LREs (creating technology‐assisted LREs). However, MI did not have a significant impact on the production of corrective feedback or LREs, potentially due to attentional resources expended during self‐correction. These findings suggest that completing task‐based activities within text chat may be beneficial for promoting beginner learners' linguistic processing and integrating technology to facilitate discussion of L2 form but less impactful for promoting peer feedback.

Stepping Back: How Should Pass/Fail Scoring Influence Step 1 Timing?

Although most students complete Step 1 before clerkships, some institutions delay the exam until after clerkships. The change to pass/fail grading adds additional complexity that should be considered when deciding about exam timing. Both early and late administration may affect learning outcomes, learner behavior, student well-being, and residency match success. Step 1 completion before clerkships promotes learning outcomes (e.g., integration and mastery of foundational material), may encourage students to focus on the curriculum, and may better prepare students for clinical science exams (CSEs). However, delaying the exam ensures that students maintain foundational knowledge and may encourage clinical educators to demonstrate basic science illustrations. An early Step 1 may affect learner behavior by allowing clerkship students to focus on clinical learning. The associated National Board of Medical Examiners performance report may also be used for Step 2 and CSE preparation. However, delaying Step 1 allows greater scheduling flexibility based on developmental milestones. Administration of Step 1 before clerkships removes a significant stressor from the clinical year and decompresses the residency application period. However, a delayed Step 1 reduces the pressure on students to engage in numerous extracurricular and research activities to distinguish themselves due to the pass/fail change. An early Step 1 exam may also lead to improved CSE performance, which is often linked to clerkship honors criteria, an increasingly valuable distinction for residency match success after the change to pass/fail. In contrast, delaying Step 1 is associated with higher first-time pass rates, which may be especially important for students at risk for failure. Medical educators and students should collaboratively approach the question of Step 1 timing, considering these factors within the context of the medical school program, curricular constraints and priorities, and students' individual needs and goals.

Home Visitation and Its Effects on Learners Behavior at Mabini District, Bohol

Home Visitation and Its Effects on Learners Behavior at Mabini District, Bohol

Immersive human–computer interaction and digital entertainment new media application in English e-learning mode

With the rapid development of e-learning, more and more learners use electronic devices to learn English. The traditional way of learning is often lack of interest and interaction, therefore, this paper discusses how to improve the effect of English learning through the application of immersive human–computer interaction and digital entertainment new media. Research on the selection of appropriate virtual reality, augmented reality or game platforms, design interactive content and scenes according to learning objectives, and use elements such as virtual scenes, virtual characters and audio-visual effects to improve learners’ immersion. Through careful design and production, learners can be presented with a lively, interesting and challenging learning environment to stimulate their interest and enthusiasm in learning. Develop interactive learning tools and applications that enable learners to interact with virtual environments or games, make appropriate adjustments and feedback based on learner feedback and behavior, and provide personalized learning support and guidance. Through the collection of learner feedback and behavioral data, combined with the evaluation indicators of learning outcomes, the immersive learning environment and content can be evaluated and improved and optimized according to the evaluation results, so as to further enhance learners’ learning motivation and learning outcomes.

Learning Behavior Analysis and Prediction Based on Convolutional Neural Networks

With the development of information technology and the change of educational needs, personalized learning and intelligent education have become an important direction of current educational research. Traditional educational assessment and teaching methods are limited by subjectivity and limitations, and it is difficult to fully analyze and understand learners individual needs and behavior patterns. The analysis and prediction method of learning behavior based on convolutional neural networks provides new tools and perspectives for educational decision makers and teachers through automated and data-driven methods, which is expected to deepen the understanding of the laws behind learners behaviors and achieve the goal of education personalization and effect optimization. This paper will discuss the application status and future development direction of CNN in learning behavior analysis, and contribute new theoretical and practical experience to the innovation and progress of educational technology.

Advances in Video Analytics

AbstractLearners interact with content, assessments, peers, and instructors in digital learning environments. Videos, which are popular due to internet technologies, capture learners’ attention, boost motivation, and enhance learning. Learning analytics broadly optimize educational environments by analyzing data, with video analytics focusing specifically on video interactions to enhance learning outcomes. Video-player interactions (e.g., play, pause) and video content interactions (e.g., true-false questions) provide insights into learner behaviors. Lack of interaction is a major reason for high dropout rates in video platforms and MOOCs. Video analytics can help address this issue by analyzing and improving engagement with video content. This special issue has a specific focus on video analytics and impact of this field to the learning experience. Four articles were included in this special issue. The findings reveal that I) the type, length, and purpose of the video are important for student engagement, ii) important tips on video-based learning design are presented, iii) when interacting with the video player, pause, play, rewind and fast forward are the most commonly used interaction types., iv) providing more information about video interaction processes with dashboards would provide much more insight, and v) dividing the videos into more than one section both creates the perception of better structuring of the process and the segmentation of the videos contributes more to learning.

The influence of teacher emotional support on language learners’ basic psychological needs, emotions, and emotional engagement: Treatment-based evidence

This study assessed the influence of a 10-week teacher emotionally supportive quasi-experimental intervention on the perceived teacher emotional support (PTES), basic psychological needs (BPNs) satisfaction, emotions of anxiety and enjoyment, and emotional engagement of learners of English as a foreign language (EFL). Strategies targeting the three dimensions of TES, that is, positive climate, teacher sensitivity, and regard for students’ perspectives, were deployed exclusively in an experimental group (N = 63), which was compared to a control group (N = 58) in which a more typical teaching approach was followed. Classroom observations and questionnaires were used to capture three time points of changes in learner behaviors. The results of multivariate analyses revealed significant positive changes over time in students’ self-reported PTES behaviors, BPNs satisfaction, emotions, emotional engagement and observed behaviors solely in the experimental condition. Learner BPNs satisfaction showed the largest group differences by mid-term treatment. With the continuous deployment of the treatment, the between-group differences peaked toward the end of the experiment for all learner behaviors. The largest variance at this stage was in learner PTES. These experimentally driven findings provide compelling evidence for the advantages of TES pedagogical interventions for second language learners.

An analysis of turn-taking behaviors of Japanese learners of English in videoconferencing discussions

To successfully interact in a new language, learners must effectively take turns and manage the floor. Expectations and rules for turn-taking can differ between languages, making this more than a question of grammatical proficiency and vocabulary acquisition. In addition, the increased use of videoconferencing software in education and beyond makes learning to manage the floor in virtual spaces a useful skill for everyone, including language learners. To analyze the turn-taking behaviors of first language (L1) speakers of Japanese using English in videoconferencing interactions, three classes of first-year university students with intermediate level English skills were selected to have their in-class videoconferencing discussions recorded. Three of the recorded discussions were selected and analyzed to identify turn-taking behaviors and to analyze the techniques that influenced the speakers’ ability to succeed. Participants were observed to leave long gaps between speakers at turn transitions and rarely extended discussion topics to include multiple turns per speaker. There were three behaviors that were observed that helped some participants to reduce the gap between speakers. Clearly marking the end of a turn, energetic use of verbal backchannels, and comfort with unintentional overlap all seemed to be correlated with improved transition speed. Increased use of questions did lead to some topics being discussed beyond a single turn per speaker, but the difference was relatively small.

Monitoring student behavior in autonomous automatic speech recognition-based pronunciation practice

A growing number of studies are investigating the role of Automatic Speech Recognition (ASR) technology in second language (L2) pronunciation learning, highlighting its potential to promote autonomous learning and enhance the development of L2 pronunciation. However, how learners interact with ASR systems remains largely unknown. The current study examined the behavior of 46 English-as-a-foreign language learners over six sessions of autonomous ASR-based pronunciation practice. It also explored potential relationship between learners' behavior and individual differences in motivation and pronunciation attitude. Analyses of the learners' screen recordings shed light on their interaction with the ASR output, including the proportion of words successfully transcribed on the first attempt and unsuccessful cases requiring subsequent attempts—whether repaired, ignored, or remaining unrepaired despite several attempts. Results from motivation and attitude questionnaires revealed that learners with a higher pronunciation attitude at the study's onset were more likely to reattempt a word not successfully transcribed by the ASR system. Motivation, however, was not found to be linked to learner behavior. The findings contribute to the expanding research on ASR technology in L2 pronunciation learning by emphasizing the influence of learner behaviors and attitudes, and providing important insights for L2 practitioners and researchers.

Evaluating Citizen Science: Moving beyond Output Measures to Learner Behaviors, Interests, and Motivations

With the rising popularity of citizen science, there’s a growing focus on effective assessment methods for participant outcomes. Unlike many biodiversity programs, which typically emphasize programmatic outcomes, this study examined how Blue Thumb, an Oklahoma water quality citizen science program, influences volunteer behaviors, interests, and motivations using an explanatory survey based on Phillips et al.’s (2018) evaluation framework. Although there were no significant differences in behaviors and interests between new and experienced volunteers (N = 74), motivations varied significantly, distinguishing between intrinsic and extrinsic reasoning. Blue Thumb volunteers already displayed pro-environmentalism and strong scientific interests before joining, and these traits persisted rather than being cultivated through participation. The study’s insights informed Blue Thumb’s future volunteer recruiting and retention strategies. While contributing to research utilizing a common framework, future studies should incorporate qualitative methods to provide context for a deeper understanding of the quantitative findings.

ENVIRONMENTAL CARE: THE STRENGTHENING CHARACTER EDUCATION THROUGH ISLAMIC RELIGIOUS EDUCATION (IRE) DEVELOPMENT

Dogmatic Islamic Religious Education (IRE) learning in schools seems to lack of attention to environmental issues. Thus, IRE needs to incorporate environment issues. This study aims to analyze the concept and implementation of strengthening environmental care character education through IRE development as well as its implications in framework of eco-culture formation, using MTsN 1 Sumenep and SMPN 2 Larangan Pamekasan as case studies. Using a qualitative approach with a multi-site study design, This research used interviews, observations, and documentation in collected data. The research findings show that the concept of strengthening environmental care character education through IRE development describes efforts to strengthen environmental care character holistically-integrated through curricular, co-curricular, and self-development activities. While its implementation is carried out through 4 stages, namely design which includes identifying the types of activities, developing environment-based IRE study materials, preparing supporting facilities; implementation which includes integrated PAI learning, school management, and student development activities; monitoring and evaluation is carried out through supervision activities and monthly meetings; follow-up refers to school self-evaluation. While the implications can be seen from 3 aspects; the improvement of educational services, changes in learner behavior, a decrease in school operational costs and the existence of class funds.

Exploring Learner Behaviors and Perceptions on an SNA Dashboard: Interactivity and Data Visualization Literacy

Exploring Learner Behaviors and Perceptions on an SNA Dashboard: Interactivity and Data Visualization Literacy

Educational Big Data Analytics for Futuristic Smart Learning Using Deep Learning Techniques

The goal is to use the massive amounts of data created by digital education systems to develop intelligent and adaptable learning environments that are particularly suited to each student’s requirements. The rapid digitization of education systems has led to the proliferation of educational big data, presenting unprecedented opportunities to reshape learning environments into intelligent, responsive spaces that adapt to the needs of individual learners. This paper explores the integration of advanced deep learning techniques with educational big data analytics to forge the path towards futuristic smart learning ecosystems. By leveraging robust datasets derived from a myriad of educational interactions, ranging from student performance metrics to engagement patterns in digital learning platforms, we propose a multi-tiered analytical framework that harnesses the predictive power of deep learning. We commence by elucidating the scope and scale of educational big data, highlighting its potential to provide granular insights into student learning processes. The paper then delineates the architecture of a deep learning-based analytical model designed to process complex, multidimensional educational datasets. This model applies state-of-the-art algorithms to perform tasks such as predictive analytics for student performance, personalized content recommendation, and real-time engagement monitoring. Central to our discussion is the application of convolutional neural networks (CNNs), recurrent neural networks (RNNs), and deep belief networks (DBNs) in deciphering patterns and trends that escape traditional analytical methodologies. We emphasize the capacity of these techniques to capture the subtleties of learner behavior and to facilitate the development of adaptive learning pathways. Furthermore, we address the challenges of integrating deep learning with educational big data, including issues of data privacy, computational demands, and the need for robust model interpretation. The paper presents a series of case studies that demonstrate the successful application of our proposed framework in various educational settings, from K-12 to higher education and continuous professional development.

Comparative Analysis of the Switching Behavior of Synchronous and Asynchronous Online Learners in Higher Education

This study examines the phenomenon of learner behavior change in synchronous and asynchronous online learning environments. The article assesses the level, stage, reasons, and impact of change among online learners. To facilitate systematic analysis, a set of comparison parameters has been developed. These parameters include success and failure rates, mode and location of learning, level of interaction, group and individualized learning, courses with practical exercises, economic factors, self-regulated learning, indicators of change, technological tools, learner satisfaction, and the impact of a blended learning model. Overall, this article contributes to the understanding of online learners’ behavioral changes and provides valuable information to educators, administrators, and online education providers for designing effective online learning environments. The identified parameters can serve as a foundation for future research on how synchronous or asynchronous modes can either enhance or hinder the overall learning experience of online learners.

Student's Behavior Clustering based on Ubiquitous Learning Log Data using Unsupervised Machine Learning

Online learning is the source of data generation related to learner's learning behaviors, which is valuable for knowledge discovery. Existing research emphasized more on an understanding of student's performance and achievement from learning log data. In this study, we presented data-driven learning behavior clustering in authentic learning context to understand students' behavior while participating in the learning process. The objective of the study is to distinguish students according to their learning behavior characteristics and identify clusters of students at risk of unsuccessful learning achievement. Learning log data were collected from ubiquitous learning applications before conducting Exploratory Data Analysis (EDA) and cluster analysis. We used partitional clustering using K-means algorithm and hierarchical clustering based on the agglomerative method to improve clustering strategies. The result of this study revealed three different clusters of students supported by data visualization techniques. Cluster 1 comprised more students with active learning behavior based on the total logs, total problems posed, and the total attempts in fraction operation and simplification. Students in clusters 2 and 3 had a higher attempt at problem-solving instead of problem-posing. Both clusters also focused on fraction's conceptual understanding. Knowledge discovery of this study used real data generated from ubiquitous learning application namely U-Fraction. We combined two different types of clustering method for delivering more accurate portrait of a student's hidden learning behaviors. The outcome of this study can be a basis for educational stakeholders to provide preventive learning strategies tailored to a different cluster of students.

Study on Lying Behavior of Learners at the Integrated Islamic Elementary School (SDIT) Darul Ihsan Pontianak

Study on Lying Behavior of Learners at the Integrated Islamic Elementary School (SDIT) Darul Ihsan Pontianak

User experience of english online classroom E-learning based on routing algorithm and data visualization analysis

With the development of the Internet, E-learning has become an important way of education. However, in online English classes, user experience problems still exist, such as network delay and slow video loading speed, which affect the learning effect and user engagement. Therefore, this paper hopes to improve E-learning user experience by studying routing algorithm and data visualization analysis technology. The routing algorithm is used to optimize the network connection. By analyzing the network topology and data transmission path, the optimal route is selected to reduce the network delay and bandwidth consumption. Data visualization technology is used to analyze and display the behavior and learning data of learners in the classroom in real time, so as to help learners better understand their learning situation. The experimental results show that the application of routing algorithm and data visualization analysis technology can significantly improve the user experience of online English classes. The stability of learners’ network connection in the classroom is improved, the video loading speed is accelerated, and the learning effect is also significantly improved. By optimizing network connectivity and providing real-time data analytics, it is possible to meet the individual needs of learners and improve their learning effectiveness and engagement.

Relationship between learners' L2 Motivational Self System and parental investment behavior in learners' English learning.

Students' motivation and learning behavior are significantly impacted by parents' participation and investment. It has been demonstrated that parental investment behavior could exert a direct effect on students' L2 Motivational Self System (L2MSS) mediated by parental investment belief. Nevertheless, the relationship between components of parental investment behavior and students' language learning motivation remains a topic necessitating further scholarly investigation. In response to this gap, we conducted a quantitative study involving a survey of 900 high school students to explore the relationship between students' English learning motivation, as conceptualized by the L2 Motivational Self System and parental investment behavior based on a four-component model. The findings in this study indicated that high school students exhibited moderate levels of L2MSS and relatively low levels of parental investment behavior. Moreover, students' L2MSS was found to have a significant positive correlation with the global parental investment behavior, with parental emotional investment behavior emerging as a positive predictor of high school students' L2MSS. These results underscore the importance of parental provision of sufficient economic, relationship, knowledge, and emotional support in cultivating a nurturing and supportive familial context conducive to the development of students' positive future selves.