This study tests the effectiveness of integrating ChatGPT into debate sessions to strengthen critical thinking and argumentation skills among undergraduate diplomacy students in Colombia. One hundred and sixty-two participants were randomly assigned to an experimental group using ChatGPT during debates and a control group engaging in traditional debates. Pretest and post-test assessments measured participants’ skills using adapted versions of the Watson-Glaser Critical Thinking Appraisal and a custom argumentation rubric. Multivariate analysis of covariance (MANCOVA) found higher scores in the experimental group for combined learning outcomes. Univariate analyses of covariance (ANCOVAs) showed improvements in understanding complex concepts, critical thinking, and argumentation skills for the experimental group. Structural equation modeling (SEM) noted direct and indirect effects of ChatGPT on these skills. The study also analyzed participants’ perceptions through qualitative data from post-test open-ended questions. These findings highlight ChatGPT’s potential in fostering essential skills in resource-constrained educational settings, while noting implementation challenges in developing nations. This research adds to the growing literature on AI-powered tools in education, particularly in economically emerging regions.
Read full abstract