Predicting Online English Language Learning Outcome From Learner-, Instructor-, and Course-Level Factors

• In Event: Investigating Ways of Evaluating, Measuring, Predicting, and Profiling Online Attitudes and Behaviors

Abstract

Online language learning poses great challenges on both learners and instructors. While most studies of online language learning used students’ self-perceived data and focused on higher education, this study provides a comprehensive examination of factors predicting online high-school English language learning outcome from learner, instructor, and course level using data from the learning management system (LMS). Using hierarchical linear modeling with 919 student data nested in 8 course courses taught by 13 instructors, our findings suggested that students’ login times and durations in the LMS significantly predicts their academic achievement. Furthermore, having more project-based assignments and text resources such as instructor guide helped enhance student learning. Implications for online language course design, teaching practices, and research are discussed.

Authors

• Binbin Zheng, Michigan State University

• Chin-Hsi Lin, The University of Hong Kong

• Jemma Bae Kwon, Michigan Virtual