Analyzing the Effectiveness and User Experience of AI-Based Creative Writing Mobile Learning Integrated with Local Wisdom (AI-CreWLi)

Abstract

The development of artificial intelligence (AI) technology in education has opened up new opportunities to create more interactive, adaptive, and contextual learning, including in the development of creative writing skills. This study aims to analyze the effectiveness and user experience of the application of AI-CreWLi (AI-Based Creative Writing Mobile Learning Integrated with Local Wisdom) in improving students' 21st century creative writing skills. The study used an explanatory quantitative approach with a pretest-posttest design and a Likert scale questionnaire to assess user perception. A total of 78 students participated in creative writing learning using the AI-CreWLi application for eight weeks. The results showed a significant improvement in the experimental group using AI-CreWLi, with the average score increasing from 73.91 to 85.59, while the control group only increased from 66.09 to 75.13. The Levene test showed homogeneous data variance (Sig. > 0.05), while the paired t-test and ANOVA confirmed a significant difference between the two groups (Sig. = 0.000). Linear regression analysis yielded a coefficient of β = 1.173 (Sig. = 0.000) with a standardized Beta value of 0.807, indicating a strong influence of the use of AI-CreWLi on learning outcomes. In addition, students' perception of the application showed values of R = 0.786 and R² = 0.601, which means that 60.1% of the variation in students' positive perceptions was explained by the use of AI-CreWLi. The perception ANOVA test (F = 47.836; Sig. = 0.000) also showed a significant influence on 21st century skill development. In conclusion, the application of AI-CreWLi has proven to be effective in enhancing creative writing skills and strengthening 21st century skills through AI-based learning that is adaptive, contextual, and integrated with local wisdom values. These findings support the development of innovative learning models that are in line with the Independent Learning – Independent Campus policy.