Investigating the Role of AI-Powered Tools in Reducing Foreign Language Anxiety among Bangladeshi EFL Learners: An Illustrative Mixed-Methods Study

Read full text Summarise with VIRAT
Authors: Md. Ismaile Hossain
Cited by: 0
References: 16

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Recent growth in conversational artificial intelligence, automated speech-evaluation systems, and generative feedback tools has renewed interest in whether such technologies can reduce foreign language anxiety in English language learning. This article presents a journal-formatted illustrative mixed-methods manuscript based on a transparently simulated pilot dataset designed for manuscript prototyping in a Bangladeshi university EFL context. A quasi-experimental design was modelled with 148 undergraduates across an experimental group (n = 74) using ChatGPT, ELSA Speak, and Grammarly-supported low-stakes rehearsal tasks for eight weeks and a comparison group (n = 74) receiving conventional speaking practice. Quantitative measures included an adapted eight-item foreign language anxiety scale and a six-item speaking self-efficacy scale; qualitative data consisted of 12 illustrative semi-structured interviews. The simulated analyses showed that the AI-supported group demonstrated a larger reduction in anxiety (Mpre = 3.52, SD = 0.46; Mpost = 2.83, SD = 0.45) than the comparison group (Mpre = 3.45, SD = 0.52; Mpost = 3.31, SD = 0.49), with a significant post-test between-group difference, t(146) = -5.09, p < .001. Speaking self-efficacy also improved more strongly in the experimental group. In regression modelling, perceived feedback quality and self-paced control were the strongest negative predictors of post-test anxiety, while the qualitative themes highlighted private rehearsal, repeatable feedback, and increased control over learning pace. The other issues were still that there were some instances of distrust towards automated feedback and a potential of overreliance. The paper posits that AI-based technologies are less helpful when used to replace teachers but as low-stakes rehearsal partners that can help to mitigate, but not avoid, anxiety related to speaking.

Fields of Study & Subjects

EFL/ESL Teaching and Learning English Language Learning and Teaching Artificial Intelligence in Education

Publication details

Journal / source
International Journal For Multidisciplinary Research
Volume
8
Issue
2
Publication date
21 March 2026
Language
EN