TeamLearn - Teamwork analytics for training collaborative problem solving in professional higher education
TeamLearn studies how digital technologies, multimodal learning analytics and automated feedback can be employed to study collaborative problem solving and team-based learning in higher education.
Photo: Jason Goodman/ Unsplash
About the project
Student teams from medical interprofessional, nursing and legal education are studied while they solve complex problems, for example, a patient diagnosis, a critical intensive care situation or a complex court case. The project uses design-based research and develops a multimodal learning analytics dashboard (MAD), which visualizes multimodal data as feedback on the team activities. Multimodal data includes, e.g. micro-location sensors, audio, video, object use logs, and affective states. The feedback is shared with students and teachers in interventions aimed at supporting collaborative problem solving and team-based learning.
The project has four objectives
- Develop methodologies and indicators for the real-time feedback and guidance of teamwork in collaborative problem solving in professional education;
- Adapt, test, and implement a learning tool for providing automated feedback based on multimodal analytics on teamwork during learning collaborative problem solving;
- Generate knowledge about the effects of interventions and learning designs based on multimodal teamwork analytics on the collaborative problem-solving process and outcomes;
- Develop and disseminate knowledge and instruments based on multimodal teamwork analytics to support collaborative problem-solving and teamwork across professional education contexts.
The project runs four years from December 2021 until November 2025. The project combines learning research, computer sciences and learning analytics, and includes Norwegian partners and collaboration with universities from Australia, USA, Finland and the Netherlands.
The project has four phases
- Phase 1: Generating teamwork indicators. This phase includes exploring relevant stakeholders’ experiences with collaborative problem solving and teamwork activities, as well as their needs for support. Based on these outcomes, we will identify observable indicators that provide information about the quality of team’s collaborative problem solving.
- Phase 2: Co-design and trials of multimodal learning analytics tool and learning scenarios. Indicators from phase 1 are used to adapt and pilot a multimodal teamwork analytics dashboard (MAD) developed at Monash University. In three higher education courses (law, nursing, interprofessional health care), researchers and teachers co-design learning scenarios in which MAD is piloted to support students’ collaborative problem solving through automated feedback.
- Phase 3: Scaled empirical studies and implementation. In two iterations, we examine the use of automated MAD-based self-assessment and guidance, and its influence on collaborative problem solving process and team performance in the three empirical cases. The first iteration involves a post-event intervention, where MAD-based feedback is provided in a debriefing after the CPS session. The second iteration follows the same procedure, with MAD-based feedback provided in real time during CPS sessions.
- Phase 4: Dissemination and exploitation activities
- Centre for Experiential Legal Learning, Faculty of Law, UiO
- Institute of Health and Society, Faculty of Medicine, UiO
- Institute for Nursing and Healthcare, University of South-East Norway
- Centre for Learning Analytics, Monash University, Australia
Research and development groups involved
Scientific Advisory Board
- Sten Ludvigsen, University of Oslo, Norway
- David William Shaffer, University of Wisconsin-Madison, USA
- Alyssa Wise, New York University Steinhard, USA
- Davinia Hernandez-Leo, University of Pamplona, Spain
- Klas Karlgren, Karolinkska Institute, Sweden
Cristin returned 'not found'