Leveraging Machine Learning Approach to Identify the Predictors of Informal Digital Learning of English Behaviours Among EFL Learners

Background Study

With the rapid transition to remote learning necessitated by the closure of traditional educational infrastructures globally, the arena of informal digital learning of English (IDLE) has received much attention, particularly among English as a Foreign Language (EFL) learners in China.

Objective

This study explores how demographic variables (gender, age, grade, major, and background) along with confidence, desire, online self-efficacy, attitudinal belief, and intention to learn English predict IDLE behaviours among EFL learners in IDLE contexts.

Methods

Utilising a comprehensive dataset, the research incorporates machine learning algorithms (e.g., Random Forest, Support Vector Machine, Logistic Regression, Decision Tree, Gradient Boosting Decision Tree and Adaptive Boosting (AdaBoost)) to analyse psychological, behavioural and demographic predictors of IDLE behaviours. Participants included 2, 055 EFL learners in China.

Results

The study finds that EFL learners' confidence, desire, online self-efficacy, attitudinal belief, intention to learn English and IDLE behaviours display a moderate level. Moreover, confidence and desire act as the strongest predictors of IDLE behaviours, whereas demographic variables (gender, age, grade, major and background) predict the minimum of IDLE behaviours.

Conclusion

By understanding these predictors, educational strategies can be better tailored to enhance digital education outcomes.