The Effect of ChatGPT-Assisted Argument Development on Students Argumentative Speaking at MAN 2 Kota Padang

Mafrani, Risya (2026) The Effect of ChatGPT-Assisted Argument Development on Students Argumentative Speaking at MAN 2 Kota Padang. Skripsi thesis, Universitas Islam Negeri Imam Bonjol Padang.

	Text (COVER-DAFTAR ISI) RISYA MAFRANI_NIM.2214050078_COVER-DAFTAR ISI.pdf - Published Version Download (2MB)
	Text (BAB I) RISYA MAFRANI_NIM.2214050078_BAB I.pdf - Published Version Download (842kB)
	Text (BAB III) RISYA MAFRANI_NIM.2214050078_BAB III.pdf - Published Version Download (843kB)
	Text (BAB V-DAFTAR PUSTAKA) RISYA MAFRANI_NIM.2214050078_BAB V-DAFTAR PUSTAKA.pdf - Published Version Download (739kB)
	Text (FULLTEXT) RISYA MAFRANI_NIM.2214050078_FULLTEXT.pdf - Published Version Restricted to Repository staff only Download (7MB)

Abstract

This study examines the impact of ChatGPT-assisted argument development on the argumentative speaking skills of students at MAN 2 Padang City. This skill requires students not only to speak fluently but also to articulate clear claims, logical reasoning, supporting evidence, and coherent idea organization aspects that often pose challenges for EFL students when transforming initial opinions into fully formed oral arguments. The study employed a quantitative approach using a quasiexperimental pre-test/post-test control group design. The sample consisted of 60 students (30 in the experimental class and 30 in the control class) selected via cluster sampling. The experimental class received instruction in debate-based argumentative speaking utilizing ChatGPT as an argument development tool, while the control class received conventional instruction. Argumentative speaking proficiency was measured using a rubric adapted from the Toulmin Argumentation Model (covering claims, reasons/warrants, evidence/data, backing, rebuttals, and organization). The results indicate that the mean score of the experimental class rose from 26.77 (pre-test) to 43.83 (post-test) an increase of 17.06 points whereas the control class saw an increase of only 3.70 points (24.43 → 28.13). A pairedsamples t-test revealed significant improvement in both groups (p < .001), and an independent-samples t-test on post-test scores showed a significant difference favoring the experimental class, t(58) = 13.195, p = .000, with a mean difference of 15.70 points. These findings demonstrate that ChatGPT-assisted argument development has a significant positive effect on students' argumentative speaking skills, particularly in organizing ideas, presenting evidence, and formulating rebuttals. It is recommended that ChatGPT be integrated as a scaffolding tool in speaking instruction, provided its use remains guided and students retain ownership of the ideas they express orally.

Actions (login required)

View Item