The Effectiveness of AI-Based Tutors in Teaching Reading Comprehension: A Case Study Using Duolingo for EFL College Students in Iraq

The present research aimed at investigating the efficacy of AI-based tutors on the improvement of reading comprehension skills by Iraqi EFL college students using Duolingo as a digital tool for learning. Both the research and the participants were from the Department of English, College of Basic Education, University of Maysan, during the academic year 2024-2025. In total, 60 second-year undergraduate students were involved in the study, split into experimentation (n = 30) where students use Duolingo for reading exercises while control (n = 30) still continue in traditional classroom instructions. In assessing the reading comprehension level of students, a quasi-experimental design with pre- and post-tests was employed. The AI-based adaptive learning system in Duolingo includes individualized exercise, real-time feedback, and game-based reading tasks. The quantitative data were analyzed with SPSS 26, with paired sample t-tests and ANOVA to measure significant differences in students' performance. Furthermore complemented by qualitative dimension, that is, semi-structured interviews with five instructors from the English department to elicit their views regarding an AI-assisted reading instruction. Experimental findings indicate that statistically significant improvements in reading speed, vocabulary retention, and inferencing skills were achieved by the experimental group relative to the control group (p