Please use this identifier to cite or link to this item: https://repository.isls.org/handle/1/10601
Title: Identification of Different Patterns in Solving Collaborative Jigsaw Puzzle Tasks Using Hidden Markov Models
Authors: Xu, Qizhi
Ge, Zongting
Zhu, Mengxiao
Dou, Ziwei
Wang, Jing
Keywords: CSCL
Issue Date: 2024
Publisher: International Society of the Learning Sciences
Citation: Xu, Q., Ge, Z., Zhu, M., Dou, Z., & Wang, J. (2024). Identification of Different Patterns in Solving Collaborative Jigsaw Puzzle Tasks Using Hidden Markov Models. In Clarke-Midura, J., Kollar, I., Gu, X., & D'Angelo, C. (Eds.), Proceedings of the 17th International Conference on Computer-Supported Collaborative Learning - CSCL 2024 (pp. 83-90). International Society of the Learning Sciences.
Abstract: To assess learner’s collaborative problem-solving (CPS) skills, it is necessary to identify their behavior patterns in completing collaborative tasks. These patterns are concealed within the collaboration and problem-solving processes, and can only be analyzed through modeling of the detailed process data. This study utilized Hidden Markov Models (HMMs) to analyze 15,767 instances of process data generated by 63 students in an online collaborative jigsaw puzzle task. With the goal of comparing low and high performing teams, we identified three hidden states in the low-scoring group, including Warm-up, Action, and Strategy Execution and Validation states, and four hidden states for the high-scoring group, including the three states for the low-scoring group plus a new Strategy Optimization state. Furthermore, the results showed that even though both groups initiated the task with the Warm-up state, the high-scoring group exhibited more complex transitions between states than the low-scoring group.
Description: Long Paper
URI: https://doi.org/10.22318/cscl2024.849100
https://repository.isls.org//handle/1/10601
Appears in Collections:ISLS Annual Meeting 2024

Files in This Item:
File SizeFormat 
CSCL2024_83-90.pdf669.34 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.