Evaluation of Problem Based Gamification Learning (PBGL) Model on Critical Thinking Ability with Artificial Intelligence Approach Integrated with ChatGPT API: An Experimental Study

The purpose of this research is to evaluate the effectiveness of Problem Based Gamification Learning (PBGL) model integrated with Artificial Intelligence (AI) through ChatGPT API in improving participants' critical thinking skills. The study used an experimental design with two groups, namely control and experimental, each consisting of 120 students (240 in total) from two countries (Timor Leste and Indonesia) studying basic Python programming. The main issues were complex system integration with ChatGPT API, technical challenges in AI implementation for automated feedback, and maintaining test consistency and fairness. Results showed the experimental group had a significant improvement in critical thinking skills, with an average score improvement of 23.57 points compared to 9.04 points in the control group, showing a mean difference of 14.53 points. Content Validity Ratio (CVR) testing showed good content validity with CVR values > 0.75 for all items. Construct Validity (CV) testing confirmed the instrument measured the desired construct with significant correlation coefficients. Test-retest Reliability testing showed excellent reliability with a Pearson correlation coefficient of 0.960. The findings confirm that the integration of AI in gamified learning can provide automatic feedback and additional relevant information, improving critical thinking and motivation of participants. The contribution of the research is the development and testing of an AI-supported PBGL model, which can serve as a reference in digital learning curriculum design and improve critical thinking skills. Practical implications show that the integration of AI in learning enriches the material and provides flexibility to participants and teachers. Overcoming the technical challenges and adaptability of AI models will be key to increasing the effectiveness of PBGL in various educational contexts in the future. This research provides new insights in the development of more effective and adaptive learning methods in the digital era.