International Journal of
Information and Education Technology

Abstract

This paper presents a Virtual Reality (VR)-based framework for evaluating English language proficiency in an immersive learning environment. Based on a dataset of 176,911 participants, we analyzed reading, listening, speaking, and pronunciation activities, as well as vocabulary acquisition, writing exercises, and grammar correction. The translation tasks and language-learning results are complemented by behavioral indicators, including gesture frequency, reaction time, interaction depth, and session length. Multivariate regression results indicated a moderate proficiency level and weak associations between behavioral engagement and language outcomes. Neither the device type nor interaction metrics had a significant effect on overall proficiency level, and mediation testing further revealed no indirect effects via response latency. These results suggest that VR language skill learning does not depend on the quantity of behavioral activity and highlight the significance of task design and meaningful engagement. This framework is designed to help build future adaptive, scalable language assessment systems using immersive technologies to support fair, data-driven English language learning.