Virtual Interaction Technology and English Learning Performance: A Mixed-Method Analysis Using SEM and fsQCA

As virtual interaction technologies (VIT), including virtual reality (VR) and simulation-based platforms, become more prevalent in higher education, their role in enhancing English language instruction warrants systematic investigation. This study explores how VIT influences undergraduate students’ engagement and English learning performance in non-native English-speaking settings, using China as a case study. A total of 416 university students from Guangzhou participated in a structured survey (53.1% male, 46.9% female; majority aged 18–21). Partial Least Squares Structural Equation Modeling (PLS-SEM) was applied to examine the relationships among three core predictors—application degree of VIT, teacher proficiency in using VIT, and classroom environment support—alongside two mediators (student engagement and learning motivation) and one outcome variable (English learning performance). Additionally, fuzzy-set Qualitative Comparative Analysis (fsQCA) was used to uncover alternative causal configurations leading to high academic performance. Results indicate that all three predictors significantly influence engagement and motivation, which subsequently improve English learning outcomes. Among all factors, classroom environment support emerged as the most influential. The fsQCA results identified three distinct configurations capable of producing high performance, highlighting the complementary strengths of linear and configurational approaches. These findings offer practical insights for enhancing digital pedagogy through combined technological, instructional, and environmental strategies in English language education.