Evaluating ChatGPT’s effectiveness in enhancing argumentative writing: A quasi-experimental study of EFL learners in Pakistan

A crucial area of ongoing research is the potential of artificial intelligence to improve learning outcomes in diverse educational settings, given its increasing global integration into education. This study assesses the effectiveness of ChatGPT, an artificial intelligence tool, in improving the argumentative writing skills of Pakistani students pursuing the Secondary School Certificate (SSC) in English as a Foreign Language (EFL) environment. The focus of this study is to assess the impact of ChatGPT on the skills of argumentative writing of SSC students to evaluate to what extent they improve the key components of argumentative writing through pre-test and post-test, and to identify the specific types of errors committed by the learners in Pakistani context. This study adopted a quasi-experimental approach, using pre- and post-tests to measure changes in students' argumentative writing abilities after three months of ChatGPT interaction. The data from the pre-test and post-test is analyzed by using key components of argumentative writing by Tolumin (1958) and the framework of error analysis by Ellis (2002). The findings demonstrate significant progress in students' argumentative writing, with a corresponding decrease in error types after implementing the ChatGPT intervention. The study's results indicate a significant improvement in the argumentative writing abilities of secondary school students using ChatGPT, suggesting its usefulness in supporting error correction and enhancing writing skills in English as a Foreign Language education.