Computers & Education最新文献

Co-viewing video lectures, where students watch lectures simultaneously with one or more remote peers and engage in interpersonal communication on learning tasks online, is becoming increasingly popular. However, there is a dearth of studies examining students' behavioral and cognitive patterns during the co-viewing process. This study employed eye-tracking and screen recording methods to examine undergraduate students' learning processes in co-viewing video lectures. It specifically focused on how students used video lectures and engaged in discussions to complete a design task. Lag sequential analysis of learning behaviors revealed that, compared to low-performers, high-performers were more engaged in behavioral sequences involving clarifying learning tasks. Furthermore, epistemic network analysis of conversations revealed that high-performers demonstrated stronger connections between deep-level knowledge and a high level of uptake in design skills. In contrast, low-performers exhibited more connections between superficial-level knowledge and a low level of uptake in these skills. The findings suggest that in co-viewing video lectures, it's crucial to engage students in clarifying behavioral sequences. Furthermore, the co-occurrence of deep-level knowledge, design skills, and a high level of uptake in shared ideas significantly enhances learning. Our research on learning processes suggests important strategies for students during co-viewing of video lectures: teachers are encouraged to design effective interactive activities and create suitable support interventions to enhance student learning performance in video lectures.

Adaptive teaching is regarded to address students' heterogeneity in schools and to individually support their learning. Technology may help to teach adaptively. However, it is still unclear whether realizing adaptive teaching with technology is a feasible teaching practice in real-world classrooms. More importantly, it is an open question which boundary conditions constrain the feasibility of adaptive teaching with technology. We realized a four-year co-design-project in which researchers and teachers co-constructively developed adaptive teaching units (duration: 3–4 weeks) by deliberately integrating technology across secondary education. To assess the feasibility of technology-enhanced adaptive teaching, we followed a sequential-explanatory mixed-methods approach during the co-design-process. We combined a quantitative study (N = 183), investigating students' learning gains and potential moderators, and a qualitative study by means of semi-structured teacher interviews (N = 3) on implementation conditions of the adaptive teaching units. We observed that the teaching units were particularly beneficial for students with low prior knowledge and when they were highly adaptive to students’ needs. Fine-grained qualitative findings indicated that formative assessments and consolidation phases were crucial for the feasibility of technology-enhanced adaptive teaching, as well as a parsimonious use of technologies. These findings indicate that the feasibility of adaptive teaching with technology depends on boundary conditions, highlighting the need for further support to unfold the potential of adaptive teaching.

In the Massive Online Open Course (MOOC) forum, learning engagement encompasses three fundamental dimensions—cognitive, emotional, and behavioral engagement—that intricately interact to jointly influence students' learning achievements. However, the interplay between multiple engagement dimensions and their correlations with learning achievement remain understudied, particularly across different academic disciplines. This study adopts an automated configurational approach that integrates bidirectional encoder representation from transformers (BERT) and fuzzy set qualitative comparative analysis (fsQCA) to explore the configurations of learning engagement, their connections with learning achievement, and variations across disciplines. Our analysis reveals a nuanced profile of learners' learning engagement, indicating the high-achieving individuals demonstrated more frequent posting and commenting behaviors and the high-level cognitive engagement than low-achieving individuals. Second, our analysis revealed multiple configurations where the coexistence or absence of factors at different levels of the cognitive, behavioral, and emotional dimensions significantly impacted learning achievement. Learners who conducted posting and replying behaviors, expressed positive emotions, and engaged in deep cognitive engagement tended to achieve superior learning outcomes. Third, there were significant differences in behavioral and emotional engagement among learners across different academic disciplines. Specifically, pure discipline learners were more inclined to engage in posting behaviors than the applied discipline learners. Across academic disciplines, positive emotions correlated strongly with higher achievement. These findings deepen our understanding of the multifaceted characteristics of learning engagement in MOOCs and highlight the importance of disciplinary distinctions, providing a foundation for educators and designers to optimize learners’ MOOC effects and tailor learning experiences in diverse disciplinary contexts.

With the increasing prevalence of digital devices such as smartphones, tablets and e-readers, more and more reading is happening in digital formats – also in classrooms across the world. The present study focuses on lower secondary school students and their reading comprehension and attitudes toward reading on paper and screens. The study uses an innovative methodological approach were eye-tracking technology is used to observe ten carefully sampled eight grade students with different reading levels. The students read a selection of texts and answered questions from the Norwegian national reading assessment on comparable versions on paper and on screen. By analyzing eye movement data (reading transitions), including more than 25000 fixations, in combination with text reading comprehension outcomes, students' cued retrospective reporting from reading, and interview data we have obtained detailed and comprehensive data on students' reading comprehension, reading behavior, and reading experiences across different media. A key result is that reading on screen leads to more shallow processing and can hinder reading comprehension. Importantly, our results from the students' cued retrospective reporting of their eye tracking, showed that they were unaware of their reading behavior and didn't reflect much on reading in different media. These findings have implications for the increasing shift to digital learning environments in the educational context. It is important to recognize the difference between reading processes, and policymakers and practitioners cannot assume that these processes are the same across individuals and different delivery modes.

The recent Covid-19 pandemic made universities rethink their traditional educational models, shifting, in some cases, to pure online or hybrid models. Hybrid settings usually involve onsite (i.e., in the classroom) and online (e.g., in a different classroom, at home) students simultaneously under the instruction of the same teacher. However, while these models provide more flexibility to students, hybridity poses additional challenges for the specific case of collaborative learning, likely increasing the teachers' orchestration load and potentially hampering fruitful interactions among learners. In order to gather empirical evidence on the impact of hybridity in collaborative learning, this paper reports a study conducted in a hybrid classroom where a Jigsaw collaborative pattern was implemented with the Engageli software. The study involved 2 teachers and 67 students enrolled in a computer science undergraduate course. Teachers' post-interviews, questionnaires and an epistemic network analysis (ENA) were used to produce study findings. Results show that teachers reported a medium-to-high orchestration load for implementing and setting up the collaborative activities in the hybrid classroom. Among the factors that contributed most to such load, teachers highlighted the creation and live management of groups and collaborative documents. Additionally, the ENA showed that teachers put much effort on monitoring group interactions and solving technical issues. Finally, we observed relevant differences on students' perceptions (e.g., satisfaction with the attention received by the teachers) based on the cohort sizes and on the students’ attendance modality (onsite vs. online).

Engaging children in describing real-life scenes provides an effective approach to fostering language production and developing their language skills, enabling them to establish meaningful connections between their language proficiency and authentic contexts. However, for such learning tasks, there has been a lack of research focusing on promoting self-directed language learning by using artificial intelligence techniques, primarily due to the challenges of handling multimodal information involved in such tasks. To address this gap, this study introduced a two-stage automated evaluation method that employed emerging cross-modal matching AI techniques. Firstly, an automated scoring model was developed to evaluate the quality of students' responses to scene description tasks. Compared with manually assigned human scores, our model scored students' descriptions accurately, as evidenced by a small testing mean absolute error of 0.3969 for a total score of 10 points. Based on the scoring results, immediate feedback was then provided to students by generating targeted comments and suggestions. The goal of this feedback was to assist students in progressively improving their descriptions of daily-life scenes, thereby enabling them to practice their language skills independently. To assess the effectiveness of the feedback, a comprehensive investigation was conducted involving 157 students from middle schools in China, and both qualitative and quantitative experimental data were collected from the students. It is found that the quality of students' descriptions was improved significantly with the assistance of immediate feedback. On average, students achieved an increase of 1.48 points in their scores after making revisions based on the feedback. In addition, students reported positive learning experiences and expressed favorable opinions regarding the language learning tasks with the automated evaluation. The findings of this study have significant implications for future research and educational practice. They not only highlighted the potential of emerging cross-modal matching AI techniques in automatically evaluating learning tasks involving multimodal data but also suggested that providing immediate targeted feedback based on automated scoring results can effectively promote students’ self-directed language learning.

Prior research has suggested that participation in peer feedback activities, specifically the giving and receiving of comments, and the implementation of suggested ideas helps to improve writing quality. However, it is not always the case that received feedback is implemented. Currently, there is uncertainty regarding how the total number of comments received during a peer feedback activity interacts with the number of implemented comments and subsequent text quality. Therefore, the present study examined the online peer feedback sessions of 186 graduate students enrolled in a scientific writing course at a Korean university. It aimed to investigate the possible moderating effect of the total number of comments on the relationship between implementation of comments and student writing performance. The results showed that there is no relationship between the number of received comments and student writing performance. Furthermore, no significant correlation was observed between the number of implemented comments and student writing performance. However, importantly, a moderating effect of the total number of comments on the relationship between implementation of comments and student writing performance was found. In other words, the more total comments students receive, the less their implementation will be associated with significant improvement of writing quality. This result is significant because it suggests that in online collaborative learning contexts, there is an upper limit of peer comments, above which the original author can no longer effectively implement them to improve writing performance.

ChatGPT, a state-of-the-art artificial intelligence (AI) chatbot, has gained considerable attention as a transformative yet controversial tool for enhancing teaching and learning experiences. Several reviews and numerous articles have been written about harnessing ChatGPT in education since its release on November 30, 2022. Besides summarising its strengths, weaknesses, opportunities, and threats (SWOT) as identified in previous systematic reviews of ChatGPT research, this systematic review aims to develop a new understanding of its influence on student engagement by synthesising the existing related research using a three-dimensional framework comprising behavioural, emotional, and cognitive aspects. We searched relevant databases and included 72 empirical studies published within one year of ChatGPT's initial release. The findings reveal robust but narrowly focused evidence related to behavioural engagement (i.e., work with ChatGPT) and disengagement (i.e., academic dishonesty). The evidence related to the emotional aspect is mixed, with instances of both engagement (e.g., satisfaction and interest/fun) and disengagement (e.g., disappointment and worry/anxiety). There is broad but weak evidence regarding cognitive engagement (e.g., increased understanding and positive self-perception) and disengagement (e.g., reduced critical thinking and overreliance). Our review uncovers several under-explored indicators of student engagement, pointing to the need for further research. Specifically, future studies could focus on students' study habits and attendance (behavioural engagement), social interaction (emotional engagement), and self-regulation and critical thinking (cognitive engagement) in ChatGPT-supported learning environments.

Blended learning offers a learner-centred approach that employs both in-class learning and digital technology to facilitate online learning. Such an approach is especially advantageous to adult-learners in higher education as it meets their educational needs. However, adult-learners’ participation in blended learning programmes remains challenging due to a general lack of online interaction, and no clear teaching strategies that address this concern. Literature relating to adult-learners’ educational needs and online interaction was consulted in order to design teaching strategies that foster adult-learners’ online interaction. The aim of this study is to further validate these teaching strategies, hence a multiple case study was carried out using a mixed method approach. As such, eight teachers and sixteen students from four courses across three universities in Belgium and the Netherlands were interviewed. Additionally, a questionnaire testing a pre-defined set of variables was distributed to 84 students. The results lead to a set of validated teaching strategies that help teachers to further develop their professional skills and expertise. The teaching strategies can be grouped into three categories, namely 1) the teacher's online presence, 2) collaborative learning activities and preparatory learning activities, and 3) the distribution of learning content and learning activities across online and in-class learning. An elaborate set of validated teaching strategies is included. This study aids towards teacher professional development and adds evidence-based knowledge to teaching strategies and instructional frameworks for adult-learners in higher education.

Teaching/learning strategies, adult-learning, cooperative/collaborative learning, teacher professional development, online learning.

Predicting students' problem-solving success in computer-based simulations at an early stage allows adaptive educational systems to provide learners with personalized support. In this paper, we predict students' problem-solving success by applying a machine-learning model, the random forest, to produce a binary classification (more vs. less successful students). During a business-related problem scenario that lasted 55 min, early behavioral data (during the first 5, 10, and 20 min) such as mouse clicks and keyboard strokes (approx. 29,800 early-window clickstreams and keystrokes during the first 20 min) of 234 trainees were recorded, mirroring the students' problem-solving behavior. We used the n-gram sequence mining technique, which was originally introduced within the emerging disciplines of natural language processing, text mining, and machine learning and has proven to be effective, particularly in the examination of online behavior. We trained the random forest model with training datasets that included all features (bigrams), as well as selected features (the most predictable bigrams explaining inter-group differences). Our results show that early predictions based on the first 10 and 20 min contained sufficient information to accurately predict problem-solving success, while predictions that are too early (based on the first 5 min) do not. As the size of the initial time window expanded, the classification performance improved. Moreover, the selection of the most predictable features improved the models' performance for all three time intervals. The model that was trained with only selected robust features that occurred in the first 20 min achieved the highest ROC AUC score of almost 0.70. This result falls within the range of accuracy scores observed in similar studies. From the instructor's perspective, predictions help in the early identification of weak students and can provide them with personalized learning prompts. For more successful students, tasks can be enriched adaptively.