There is a newer version of the record available.

Published November 28, 2024 | Version v1
Presentation Embargoed

ENHANCING CLIMATE CHANGE LITERACY THROUGH AGENT-FACILITATED COLLABORATIVE LEARNING: A CROSS-CULTURAL COMPARISON

Description

Climate change is a critical global issue. We designed a literacy-oriented climate change curriculum implemented in a computer-supported collaborative learning environment. The curriculum, focusing on “”the impact of human activities on climate change,”” included dialogue-based activities with the conversational agent Clair (i.e., Collaborative Learning Agent for Interactive Reasoning). Clair monitors student-student chat discussions, and intervenes as a teacher would to ask questions and prompt the students to engage in a more productive dialogue.


The study involved 176 students aged 14 to 16 from Taiwan (96), Germany (29), and the Netherlands (51) and conducted a cross-cultural comparative analysis. Analysis showed that Clair intervened less frequently in Taiwanese dialogues, averaging 1-2 times, compared to 3-4 times in Germany and the Netherlands. This difference likely stems from cultural distinctions in education; Taiwanese students maintain a high degree of formality and engage less in open interactions, whereas European students’ more open classroom interactions led to more frequent agent interventions.


Preliminary data from Germany indicate that Clair’s interventions not only improved teamwork and collaboration but also enhanced students’ ability to identify reliable climate change sources and understand its complex impacts, leading to more comprehensive solutions. However, post-intervention findings suggest an overly optimistic view of nature’s resilience among students, deviating from scientific consensus. These insights help the educational community better understand and address the impact of cultural differences on instructional effectiveness, aiming for a more personalized and inclusive educational environment.

Files

Embargoed

The files will be made publicly available on November 28, 2027.

Reason: This presentation contains unpublished data that will be part of a future journal publication. Files will be available after publication.