ChatGPT as a Pedagogical Tool for Developing Literal Reading Comprehension Skills in Pakistani ESL Undergraduates: A Quasi-Experimental Study

Abstract

The integration of Artificial Intelligence in education represents a paradigm shift in teaching and learning, offering unprecedented opportunities to enhance instructional effectiveness. The objective of this study is to find out the effect of Chat GPT as a pedagogical tool on undergraduate students’ English literal reading comprehension skills, involved three sub-skills; contextual meaning, identification of the main idea, and identification of supporting details. The study employed a quasi-experimental design  with experimental and control group. The sample size consisted on 78 undergraduates who were enrolled in BS four years degree programme in the Department of Computer Science,in a public sector university in Sindh, Pakistan where English Comprehension and Composition was taught as a compulsory subject in the first semester .  Purposive sampling technique was used to select the respondents of the study. An experimental group (n = 39) was taught using ChatGPT as a pedagogical tool and  control group (n = 39) was taught in the traditional method. A reading comprehension test as pretest was used to collect the data from the experimental and control group. After ChatGPT intervention of 12 week (90 min per session in each week) to the experimental group, a post test same as pre- test was conducted. The SPSS Version 26 was used to analyse the data with the help of paired-samples t-tests and independent-samples t-tests. The experimental group showed statistically significant and meaningfully greater gains in all three sub-skills (p < .001) than did the control group. The results are solid empirical evidence of the instructional potential of ChatGPT in improving ESL reading comprehension skills in Pakistani undergraduates.