Please use this identifier to cite or link to this item:
|Title:||Making Students' Ideas Visible through Coding a Scientific Computational Model|
de Deus Lopes, Roseli
|Publisher:||International Society of the Learning Sciences|
|Citation:||Fuhrmann, T., Fernandez, C., Blikstein, P., & de Deus Lopes, R. (2021). Making Students' Ideas Visible through Coding a Scientific Computational Model. In de Vries, E., Hod, Y., & Ahn, J. (Eds.), Proceedings of the 15th International Conference of the Learning Sciences - ICLS 2021. (pp. 1017-1018). Bochum, Germany: International Society of the Learning Sciences.|
|Abstract:||This study investigates the ideas of 5th-grade students while creating computational models for diffusion. Results illustrate that translating ideas into code can be a strategy to make them explicit. However, there is tension between designing blocks and learning; limited numbers of blocks interfere with students' expressing their ideas, while an open environment can make it hard to converge into acceptable explanations. We argue that creating domain-specific-blocks for modeling needs to be a thoughtfully designed process. This study investigates students' ideas while creating computational models for the phenomenon of diffusion. Seven 5th-grade students participated in individual sessions to observe an experiment and create a model to explain the experiment using domain-specific programming-blocks. Results illustrate that designing a model, coding, and translating ideas into computer-programs advance students' sensemaking about diffusion. Students' reasoning about particles' properties and behaviors evolved from a generic macro-level to a detailed micro-level description. Modeling using domain-specific-blocks is an excellent strategy to explore students' reasoning processes; translating ideas into code unpacks their thinking and reveals inadequate conceptions. There is a tension between the design of blocks and students' learning; a limited number of blocks might interfere with students' expressing their ideas, while the open endedness of the environment makes it hard to converge to an acceptable explanation. We argue that creating domain-specific blocks for modeling needs to be a thoughtful and carefully designed process.|
|Appears in Collections:||ISLS Annual Meeting 2021|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.