
Collaborative programming problem-solving in augmented reality: Multimodal analysis of effectiveness and group collaboration
Author(s) -
Cheng-Yu Chung,
Nayif Awad,
I-Han Hsiao
Publication year - 2021
Publication title -
australasian journal of educational technology
Language(s) - English
Resource type - Journals
eISSN - 1449-5554
pISSN - 1449-3098
DOI - 10.14742/ajet.7059
Subject(s) - augmented reality , computer science , task (project management) , human–computer interaction , multimedia , task analysis , workspace , collaborative learning , artificial intelligence , knowledge management , management , robot , economics
Although numerous studies have demonstrated different ways that augmented reality (AR) can assist students to understand the learning content via contextualised visualisation, less explored is its effect on collaborative problem-solving (CPS) in computer programming. This study aims to investigate how AR affects a CPS in a programming task. We designed a mobile app that could visualise computer programming in AR and non-AR 3D images. The app could involve two participants working together on a programming problem face to face in the same workspace. We conducted a within-subjects experiment to compare their AR experience to the non-AR experience and collected multimodal usage data about the task performance, verbal communication, and user experience. The analysis showed that the participants in the AR experience had higher task performance and more insightful communication than the non-AR. The participants also had positive attitudes toward the use of AR in classroom instructions. In a semi-structured interview, the participants reflected that AR helped them engage in the content and analyse the task easier. Based on this study, we discuss several challenges and implications for future instruction designers.Implications for practice or policy:AR can improve student engagement in a collaborative problem-solving task.AR has the potential to promote and improve group communication in collaborative work.Instruction designers may need to carefully align the characteristics of AR with the task content especially when physical models are rarely used in the learning content.