A survey on big data-enabled innovative online education systems during the COVID-19 pandemic

Translation and psychometric validation of the Peer Evaluation Scale for Team-based Learning (PES-TBL) for Chinese medical students

AIM

To translate, culturally adapt and evaluate the psychometric properties of the Peer Evaluation Scale for Team-based Learning (PES-TBL) for students in nursing and medical disciplines.

BACKGROUND

Effective peer evaluation tools provide a more scientific and objective assessment of collaborative learning. However, there is a lack of peer evaluation instruments designed for group learning in China.

DESIGN

A cross-sectional, methodological study.

METHODS

The PES-TBL was first translated and adapted into Chinese. A panel of 10 experts in nursing and clinical medicine education evaluated the content validity. The psychometric properties of scale were assessed in a sample of undergraduate and postgraduate students.The reliability of the PES-TBL was assessed. Exploratory and confirmatory factor analyses were performed to explore and verify its dimensionality and construct validity.

RESULTS

A total of 564 students were included, the overall content validity index was 0.975. The exploratory factor analysis confirmed a three-factor structure, including responsibility, initiative and collaboration. The reliability of the scale was adequate, with Cronbach's α (0.983), Spearman-Brown split-half index (0.916) and re-test reliability index (0.955). Confirmatory factor analysis showed a three-factor structure explaining 84.336 % of the total variance. All model fit indices fell within acceptable ranges, indicating good structural validity for the Chinese version of the PES-TBL.

CONCLUSION

The Chinese version of PES-TBL proved to be a reliable tool to test group learning performance in both nursing and medical students and it could be a broadly useful instrument in nursing and medical education.

Spatiotemporal distribution and dynamics evolution of artificial intelligence development in China

The quantified measurement and comprehensive analysis of artificial intelligence development (AIDEV) are vital for countries to form AI industrial ecology and promote the long-term development of regional AI technology. Based on the innovation ecosystems (IE) theory, this paper constructs an evaluation system to measure and analyze the spatiotemporal distribution and dynamic evolution of the AIDEV in China from 2011 to 2020. The results show that the AIDEV of China presents an overall upward trend and an obvious unbalance in the spatial distribution which is "eastern > central > western". Meanwhile, the provinces of low-level AIDEV are catching up with the high-level provinces, which leads to the regional difference of AIDEV narrowing. Moreover, the concentration and polarization phenomenon of AIDEV in China has been weakening and the AIDEV will continue to increase in the next three years. Further, there is a significantly positive spatial autocorrelation of AIDEV. Finally, high AIDEV provinces will increase the probability of surrounding provinces' AIDEV to develop. This paper expands the research stream in the field of AI research, extends the application scenarios of IE theory, and puts forward some relevant policy recommendations.

Knowledge relation rank enhanced heterogeneous learning interaction modeling for neural graph forgetting knowledge tracing

Knowledge tracing models have gained prominence in educational data mining, with applications like the Self-Attention Knowledge Tracing model, which captures the exercise-knowledge relationship. However, conventional knowledge tracing models focus solely on static question-knowledge and knowledge-knowledge relationships, treating them with equal significance. This simplistic approach often succumbs to subjective labeling bias and lacks the depth to capture nuanced exercise-knowledge connections. In this study, we propose a novel knowledge tracing model called Knowledge Relation Rank Enhanced Heterogeneous Learning Interaction Modeling for Neural Graph Forgetting Knowledge Tracing. Our model mitigates the impact of subjective labeling by fine-tuning the skill relation matrix and Q-matrix. Additionally, we employ Graph Convolutional Networks (GCNs) to capture intricate interactions between students, exercises, and skills. Specifically, the Knowledge Relation Importance Rank Calibration method is employed to generate the skill relation matrix and Q-matrix. These calibrated matrices, alongside heterogeneous interactions, serve as input for the GCN to compute exercise and skill embeddings. Subsequently, exercise embeddings, skill embeddings, item difficulty, and contingency tables collectively contribute to an exercise relation matrix, which is then fed into an attention mechanism for predictions. Experimental evaluations on two publicly available educational datasets demonstrate the superiority of our proposed model over baseline models, evidenced by enhanced performance across three evaluation metrics.

Lessons for Medical and Health Education Learned from the COVID-19 Pandemic

Our paper analyzes lessons for medical education and health education stemming from the experience gained in the course of the COVID-19 pandemic. Moreover, it tackles the issue of the social health and psychological wellbeing of medical students involved in online education during the COVID-19 pandemic. The paper systematizes up-to-date data on how medical schools and universities have adapted to the conditions of the COVID-19 pandemic and implemented novel effective solutions for the learning process, such as transitioning from traditional in-person classes to online learning, incorporating virtual simulations and telemedicine experiences for clinical training, and collaborating with health authorities to provide support in testing and contact tracing efforts. The paper contains an analysis of various aspects of medical education, such as the changes in practical classes, the impact of the pandemic on the formation of communication skills, methods for assessing students' knowledge and skills, and many others. It also considers case studies related to the implementation of educational programs, methodologies, and novel digital technologies in a pandemic. Additionally, the paper features an empirical study that is based on the results of our own survey that was carried out with the help of a snowball convenient sampling that involved 710 medical students between 19 and 25 years of age (56% females and 44% males) from 4 Russian regions (Moscow, Krasnodar, Kazan, and Saint Petersburg). We applied the correlation between stress scores, anxiety scores, factors of stress, and strategies for coping with stress and various economic and demographic variables (age, environment, and gender) that were analyzed using the chi-square test. Our results demonstrate that over 85% of the students in our sample yielded an above-average vulnerability to stress due to the COVID-19 restrictions. At the same time, around 61% of the students experienced severe anxiety during online education in the COVID-19 pandemic. The important factors leading to stress and anxiety were the fear of getting infected and social distancing, and the best strategy to deal with stress and increase wellbeing was self-control. Through a comprehensive review of the literature and empirical estimations, our paper identifies key areas of improvement, including curriculum adaptation, technology integration, faculty development, student support, and interprofessional collaboration. The proposed recommendations aim at strengthening medical education systems and preparing healthcare professionals to effectively navigate future pandemics.

Prediction of Cognitive Load from Electroencephalography Signals Using Long Short-Term Memory Network

In recent years, the development of adaptive models to tailor instructional content to learners by measuring their cognitive load has become a topic of active research. Brain fog, also known as confusion, is a common cause of poor performance, and real-time detection of confusion is a challenging and important task for applications in online education and driver fatigue detection. In this study, we propose a deep learning method for cognitive load recognition based on electroencephalography (EEG) signals using a long short-term memory network (LSTM) with an attention mechanism. We obtained EEG signal data from a database of brainwave information and associated data on mental load. We evaluated the performance of the proposed LSTM technique in comparison with random forest, Adaptive Boosting (AdaBoost), support vector machine, eXtreme Gradient Boosting (XGBoost), and artificial neural network models. The experimental results demonstrated that the proposed approach had the highest accuracy of 87.1% compared to those of other algorithms, including random forest (64%), AdaBoost (64.31%), support vector machine (60.9%), XGBoost (67.3%), and artificial neural network models (71.4%). The results of this study support the development of a personalized adaptive learning system designed to measure and actively respond to learners’ cognitive load in real time using wireless portable EEG systems.