The Predictive Effects of Sociobiographical Variables, English Learning Confidence, and Digital Competence on AI‐Mediated Informal Digital Learning of English (AI‐IDLE)
Guangxiang Liu, Xian Zhao
Abstract
ABSTRACT As powerful artificial intelligence (AI) represented by ChatGPT creates new opportunities for creative and self‐directed language learning beyond traditional classrooms, this study examines AI‐mediated informal digital learning of English (AI‐IDLE) as an emerging area of research. Building upon the proactive language learning theory, we investigated the extent to which students’ participation in AI‐IDLE can be predicted by sociobiographical variables, English learning confidence, and digital competence. A refined hierarchical regression modeling approach was adopted to analyze survey data collected from 1,012 Chinese college English‐as‐a‐foreign‐language (EFL) learners. Initial correlation analysis and linear regression analysis identified that among the eight sociobiographical variables examined, only ethnicity and number of digital devices met the inclusion criteria for further analysis. Hierarchical regression further revealed that English learning confidence was the strongest predictor of AI‐IDLE, while digital competence and number of digital devices also significantly predicted AI‐IDLE with medium to large effect sizes. Additionally, although ethnicity emerged as a statistically significant predictor, its practical impact was limited. These findings have important implications for pedagogical practice and highlight the need for future research to identify conditions that enable students to become productive and confident language learners in AI‐mediated environments.