Motivated by Machines? A Meta-Analysis of GenAI’s Influence on ESL/EFL Learning Motivation

• In Event: Teaching and Learning with GenAI

Abstract

This meta-analysis synthesizes 47 effect sizes from 24 empirical studies (N = 7,075) to examine the effects of generative AI (GenAI) interventions on ESL/EFL learners’ motivation. Using a three-level model, findings revealed a moderate, significant overall effect (g = 0.611, p < .001). Stronger effects emerged for Ideal L2 Self, L2 Learning Experience, and self-efficacy, while intrinsic, extrinsic, and Ought-to L2 Self motivations showed non-significant effects. Moderator analysis based on the usefulness framework identified three key factors: multimodal vs. single-mode formats, in-class vs. after-class use, and EFL vs. ESL contexts. These results offer theory-informed insights and practical implications for tailoring GenAI-supported language learning to different motivational dimensions and instructional settings.

Authors

• Meng Xiong, The Chinese University of Hong Kong

• Timothy Teo, The Chinese University of Hong Kong

• YaoYao Ji, Chinese University of Hong Kong