More Like this

1. Characterising Individual-Level Collaborative Learning Behaviours Using Ordered Network Analysis and Wearable Sensors

   Yan, L; Swiecki, Z; Gašević, D; Zhao; L; Li, X; Martinez-Maldonado; R; Tan; Y; et al (October 2023, Springer, Cham)

   Irgens, G; Knight, S (Ed.)

   Wearable positioning sensors are enabling unprecedented opportunities to model students’ procedural and social behaviours during collaborative learning tasks in physical learning spaces. Emerging work in this area has mainly focused on modelling group-level interactions from low-level x-y positioning data. Yet, little work has utilised such data to automatically identify individual-level differences among students working in co-located groups in terms of procedural and social aspects such as task prioritisation and collaboration dynamics, respectively. To address this gap, this study characterised key differences among 124 students’ procedural and social behaviours according to their perceived stress, collaboration, and task satisfaction during a complex group task using wearable positioning sensors and ordered networked analysis. The results revealed that students who demonstrated more collaborative behaviours were associated with lower stress and higher collaboration satisfaction. Interestingly, students who worked individually on the primary and secondary learning tasks reported lower and higher task satisfaction, respectively. These findings can deepen our understanding of students’ individual-level behaviours and experiences while learning in groups. 

   more » « less
2. Differentiable Programming for Hyperspectral Unmixing using a Physics-based Dispersion Model

   Janiczek, John; Thaker, Parth; Dasarathy, Gautam; Edwards, Christopher; Christensen, Philip; Jayasuriya, Suren (January 2020, European Conference on Computer Vision (ECCV))

   Hyperspectral unmixing is an important remote sensing task with applications including material identification and analysis. Characteristic spectral features make many pure materials identifiable from their visible-to-infrared spectra, but quantifying their presence within a mixture is a challenging task due to nonlinearities and factors of variation. In this paper, spectral variation is considered from a physics-based approach and incorporated into an end-to-end spectral unmixing algorithm via differentiable programming. The dispersion model is introduced to simulate realistic spectral variation, and an efficient method to fit the parameters is presented. Then, this dispersion model is utilized as a generative model within an analysis-by-synthesis spectral unmixing algorithm. Further, a technique for inverse rendering using a convolutional neural network to predict parameters of the generative model is introduced to enhance performance and speed when training data is available. Results achieve state-of-the-art on both infrared and visible-to-near-infrared (VNIR) datasets, and show promise for the synergy between physicsbased models and deep learning in hyperspectral unmixing in the future. 

   more » « less
3. Cognitive and Social Benefits Among Underrepresented First-Year Biology Students in a Field Course: A Case Study of Experiential Learning in the Galápagos

   Mason, N. A. (November 2018, Frontiers)

   Student attrition is a persistent challenge in the life sciences, particularly among underrepresented minorities, first-generation students, and women. Experiential learning through short-term study abroad opportunities diversifies curricula by immersing students in nontraditional academic environments. However, most experiential learning and study abroad opportunities are primarily available to upper-division undergraduates. Here, we present a qualitative analysis of an experiential learning opportunity offered exclusively to first-year U.S. undergraduate students from underrepresented demographics. We performed ethnographic observations of a 10-day field component in the Galápagos Islands and analyzed self-reported survey results and field journals. Students consistently reported strong cognitive gains in their understanding of basic evolutionary concepts. Most students also benefited socially, although we observed higher variation in selfreported social gains. Our findings suggest that immersive field courses increase scientific literacy and promote social cohesion among students. We speculate that experiential learning opportunities may improve retention of underrepresented minorities in the life sciences, and we encourage future studies to further examine the short-term and long-term impacts of study abroad on student cognition and retention. 

   more » « less
4. Hybrid Feature Extraction Model to Categorize Student Attention Pattern and Its Relationship with Learning

   https://doi.org/10.3390/electronics11091476  

   Poudyal, Sujan; Mohammadi-Aragh, Mahnas J.; Ball, John E. (May 2022, Electronics)

   The increase of instructional technology, e-learning resources, and online courses has created opportunities for data mining and learning analytics in the pedagogical domain. A large amount of data is obtained from this domain that can be analyzed and interpreted so that educators can understand students’ attention. In a classroom where students have their own computers in front of them, it is important for instructors to understand whether students are paying attention. We collected on- and off-task data to analyze the attention behaviors of students. Educational data mining extracts hidden information from educational records, and we are using it to classify student attention patterns. A hybrid method is used to combine various techniques like classifications, regressions, or feature extraction. In our work, we combined two feature extraction techniques: principal component analysis and linear discriminant analysis. Extracted features are used by a linear and kernel support vector machine (SVM) to classify attention patterns. Classification results are compared with linear and kernel SVM. Our hybrid method achieved the best results in terms of accuracy, precision, recall, F1, and kappa. Also, we correlated attention with learning. Here, learning corresponds to tests and a final course grade. For determining the correlation between grades and attention, Pearson’s correlation coefficient and p-value were used. 

   more » « less
5. Exploring science teaching in interaction at the instructional core

   https://doi.org/10.1002/tea.21790  

   Tekkumru‐Kisa, Miray; Akcil‐Okan, Ozlem; Kisa, Zahid; Southerland, Sherry (June 2022, Journal of Research in Science Teaching)

   Abstract Recent instructional reforms in science education aim to change the way students engage in learning in the discipline, as they describe that students are to engage with disciplinary core ideas, crosscutting concepts, and the practices of science to make sense of phenomena (NRC, 2012). For such sensemaking to become a reality, there is a need to understand the ways in which students' thinking can be maintained throughout the trajectory of science lessons. Past research in this area tends to foreground either the curriculum or teachers' practices. We propose a more comprehensive view of science instruction, one that requires attention to teachers' practice, the instructional task, and students' engagement. In this study, by examining the implementation of the same lesson across three different classrooms, our analysis of classroom videos and artifacts of students' work revealed how the interaction of teachers' practices, students' intellectual engagement, and a cognitively demanding task together support rigorous instruction. Our analyses shed light on their interaction that shapes opportunities for students' thinking and sensemaking throughout the trajectory of a science lesson. The findings provide implications for ways to promote rigorous opportunities for students' learning in science classrooms. 

   more » « less