Evidence-Based Insights from a Bilingual Big Data Analytics Course: A Comparative Study of Domestic and International Students in Transportation Engineering

Abstract

As transportation engineering education increasingly relies on data-driven methods, students are expected to develop competencies in programming, analytical reasoning, and interpretation of heterogeneous traffic datasets. This study examines learning outcomes and student experiences in a dual-language, dual-cohort course on big data analytics at Beijing Jiaotong University, delivered as a Chinese-taught class for domestic undergraduates (CN) and an English-medium instruction class for international students (EN) with largely identical curricular content. The analysis integrates three years of CN performance records (2023–2025) and two years of EN performance records (2024–2025), supplemented by a pre-course survey of the EN cohort on demographics, prior preparation, and learning motivation. Results show that the CN cohort achieved a slightly higher median score but with substantially greater variance, whereas EN scores were more tightly clustered with no failures among active students. Survey evidence indicates that most EN students reported limited programming and analytics foundations but strong career-oriented motivation. Correlation patterns further suggest that project-based assessment and collaborative tasks helped translate assignment effort into stable final outcomes in the EN class. Based on these findings, the study discusses practical implications for bilingual technical courses, including early-stage code scaffolding, motivation-sensitive project design, and process-oriented assessment to support learners with diverse backgrounds.