An Investigation into the Role of Large Language Models (LLMs) in Cultivating and Fostering Some of the EFL 21st Century Skills among Al Madinah International University Students

Abstract

This research investigates the impact of Large Language Models (LLMs) on the development of 21st-century skills, particularly critical thinking and creativity, among English as a Foreign Language (EFL) students at Al Madinah International University. Given the global importance of English proficiency and the educational potential of LLMs like ChatGPT and Gemini, this study fills a gap in existing research by directly investigating the empirical effect of LLMs on the four key skills (critical thinking, creativity, communication, and collaboration) beyond basic language tasks. The study employed a quasi-experimental design, with an experimental group (n = 17) utilising LLMs and a control group (n = 17) employing traditional teaching methods. A comprehensive 21st-century skills test, validated for expert review, internal consistency (Cronbach's Alpha for critical thinking: 0.704, creativity: 0.906, and overall: 0.885), and discriminant validity (significant differences between high and low scores at α≤0.01), was administered as pre- and post-tests. Test-retest reliability further confirmed the results (critical thinking: 0.767, creativity: 0.713, overall: 0.894). Statistical analysis, using the Mann-Whitney U Test and Wilcoxon Signed Ranks Test, showed significant differences (α ≤ 0.01) in favour of the experimental group on post-test scores for critical thinking and creativity. The substantial effect sizes indicate a strong positive impact of integrating LLMs on these skills. These findings suggest that LLMs can effectively enhance critical thinking and creativity in EFL education, offering practical insights for educators and policymakers seeking to incorporate advanced technological solutions into language-learning curricula.