KEBERKESANAN KAEDAH FLIPPED CLASSROOM BERASASKAN CHATGPT DALAM PENGATURCARAAN: ANALISIS TERHADAP PENCAPAIAN GURU PELATIH

Abstract

Integrating artificial intelligence (AI) technology into instructional practices is one of the seven core pillars outlined by the Ministry of Education Malaysia (MOE) to strengthen digital literacy among educators and students. Therefore, this quasi-experimental study was conducted to identify the effectiveness of a ChatGPT-based flipped classroom approach on trainee teachers’ achievement in programming. The study involved 78 trainee teachers from three intact classes. Using cluster sampling, one class (n = 26) received instruction through conventional teaching methods, another class (n = 23) was taught using the flipped classroom approach based on instructional videos, while the third group (n = 29) was taught using the flipped classroom approach based on ChatGPT. The instrument used to measure students’ achievement was the Computer Programming Achievement Test (CPAT), which had been adapted and evaluated to ensure its validity and reliability. One-way ANOVA and the Post Hoc Tukey HSD were used to analyze the data. The analysis revealed that the group exposed to the ChatGPT-based flipped classroom approach achieved significantly higher mean scores than both the conventional and video-based flipped classroom groups. These findings indicate that incorporating ChatGPT into the flipped classroom setting can significantly improve trainee teachers’ achievement in programming, highlighting a promising strategy for enhancing instructional outcomes in computer science education.