Using ChatGPT to Teach English for Academic Purposes: Perspectives from the Classroom

Abstract

Much attention has recently been given to the implications, potentials and pitfalls of ChatGPT in higher education. Thus far these discussions have taken a largely theoretical and exploratory perspective on what the strengths and weaknesses of ChatGPT are and how it could be used to either disrupt or enhance teaching and learning in higher education (Gimpel, et al., 2023; Kasneci, et al., 2023; Rudolph et al., 2023). Currently, there are few studies that have provided empirical evidence about the practical implementation of ChatGPT in the classroom, how it is being used and received by students, and its impact on student learning outcomes (Firat, 2023). This paper begins to address this gap by describing the ways in which ChatGPT was used in an English for Academic Purposes class for bachelor students in the Spring Semester of 2023 and assessing how it was received and used by students. Part one of the paper describes the curriculum design and the specific ways in which ChatGPT was incorporated into the syllabus, the learning outcomes, the assessments, and in-class activities. At the end of a semester a set of open-ended reflective questions was used to collect data from the students and gather insights into how students had experienced using ChatGPT in the class, what they saw as its strengths and weaknesses, and how they planned to use it in the future. Part two of the paper therefore uses a thematic analysis approach to consider the reflections of 26 students and identify patterns and themes in the students’ responses. Seven key themes emerged from this analysis which included: generating ideas, improving essay structure, editing, too vague/unreliable, prompting, limiting creativity, and critical thinking. The data suggests that students largely understood both the strengths and weaknesses of ChatGPT as a writing tool, however, they were not able to use it in a way that significantly improved their writing. The ways in which students planned to use ChatGPT in the future varied significantly with the most common plans being to use it to generate new ideas and edit texts. The paper concludes by discussing the implications of these findings. Keywords: AI, ChatGPT, higher education, students, learning outcomes