A Quantitative Study On The Improvement Of Students’ Reading Literacy By AI-Assisted English Reading Comprehension Training Platform

Abstract

Reading literacy remains a critical foundation for academic achievement and lifelong learning, particularly in the context of second language acquisition. Despite growing access to digital tools, many learners struggle to develop higher-order comprehension skills due to one-size-fits-all instruction and limited personalized feedback. With the rapid advancement of Artificial Intelligence (AI), adaptive learning platforms have emerged as promising solutions to bridge this gap. This study presents a quantitative investigation into the effectiveness of an AI-assisted English reading comprehension training platform in improving students’ reading literacy at the secondary school level. A quasi-experimental research design was implemented over a 12-week period, involving 120 students from Grade 9 across two comparable classrooms. The experimental group used an AI-based platform featuring adaptive text difficulty, real-time feedback, and skill-focused comprehension exercises, while the control group received conventional reading instruction. Pre- and post-test assessments were administered using a standardized English Reading Comprehension Test (ERCT) aligned with CEFR-B1 levels, evaluating key sub-skills such as vocabulary acquisition, inference-making, summarization, and main idea identification. Statistical analyses, including paired sample t-tests and ANCOVA, revealed that students using the AI-assisted platform demonstrated significantly greater gains in reading scores compared to their peers in the control group (p < 0.01). Notably, the experimental group showed marked improvement in inference (+18%) and vocabulary (+15%) sub-skills. Student feedback collected through Likert-scale surveys also indicated high levels of satisfaction with the AI platform’s usability, engagement, and effectiveness. The results suggest that AI-assisted reading platforms can offer personalized, data-driven instruction that meaningfully enhances student outcomes in reading literacy. These findings have implications for curriculum design, teacher training, and the broader integration of AI in language education. Further longitudinal and multi-contextual studies are recommended to explore sustained impacts and scalability.