Digital practice platforms can support repeated coding practice, but their value in English-medium undergraduate courses depends on how feedback, language demands, and assessment are aligned. This study analysed 293 valid assessment records from four successive offerings of an introductory data science course in a Sino-foreign cooperative programme. The cohorts represented Chinese-medium traditional teaching (2021), bilingual platform-supported teaching (2022), English-medium instruction with increased textbook and assessment difficulty (2023), and English-medium instruction with data-informed assessment adjustment (2024). Welch's one-way analysis of variance showed substantial cohort differences in overall grades, F(3, 160.14) = 48.03, p < .001, η² = .331. The 2022 cohort exceeded the 2021 cohort by 5.47 points (p = .011; Hedges' g = 0.51). The overall mean then fell by 18.18 points in 2023 before recovering by 15.38 points in 2024 (p < .001; g = 1.68). The overall pass rate recovered from 69.44% to 97.22%. Correlations between formative and final-examination grades remained positive but were numerically lower in the later cohorts. The results support a coordinated model in which digital practice, language scaffolding, formative tasks, and summative assessment difficulty are treated as one instructional system rather than independent reforms.

Tianzhen Chen. Digital Practice Platforms and Assessment Adjustment in English-Medium Data Science Education: A Four-Cohort Study. Frontiers in Educational Research (2026), Vol. 9, Issue 9: 67-74. https://doi.org/10.25236/FER.2026.090910.

[1] Letteri I, Vittorini P. Enhancing student feedback in data science education: Harnessing the power of AI-generated approaches[J]. International Journal of Artificial Intelligence in Education, 2025, 35: 3071-3094. DOI: 10.1007/s40593-025-00492-8.

[2] Castellanos-Reyes D, Leppard T. Online project-based learning to foster students' course choices in data science: A longitudinal case study using Sankey visualization[J]. International Journal of Educational Technology in Higher Education, 2025, 22: 63. DOI: 10.1186/s41239-025-00560-y.

[3] Messer M, Brown N C C, Kölling M, Shi M. Automated grading and feedback tools for programming education: A systematic review[J]. ACM Transactions on Computing Education, 2024, 24(1): 1-43. DOI: 10.1145/3636515.

[4] Børte K, Lillejord S, Chan J, Wasson B, Greiff S. Prerequisites for teachers' technology use in formative assessment practices: A systematic review[J]. Educational Research Review, 2023, 41: 100568. DOI: 10.1016/j.edurev.2023.100568.

[5] Kuklick L, Greiff S, Lindner M A. Computer-based performance feedback: Effects of error message complexity on cognitive, metacognitive, and motivational outcomes[J]. Computers & Education, 2023, 200: 104785. DOI: 10.1016/j.compedu.2023.104785.

[6] Yuan L, Fang F, Hu G. Are students prepared and supported for English medium instruction in Chinese higher education to promote educational equality?[J]. System, 2024, 125: 103423. DOI: 10.1016/j.system.2024.103423.

[7] Aizawa I. Tracking the first-year experience in English medium instruction: A pre-post study of transitional challenges[J]. English for Specific Purposes, 2024, 73: 20-32. DOI: 10.1016/j.esp.2023.09.002.

[8] Zhou S, Fung D, Thomas N. Towards deeper learning in EMI lectures: The role of English proficiency and motivation in students' deep processing of content knowledge[J]. Journal of Multilingual and Multicultural Development, 2025, 46(7): 1749-1764. DOI: 10.1080/01434632.2023.2248078.

[9] Vlachopoulos D, Makri A. A systematic literature review on authentic assessment in higher education: Best practices for the development of 21st century skills, and policy considerations[J]. Studies in Educational Evaluation, 2024, 83: 101425. DOI: 10.1016/j.stueduc.2024.101425.

[10] Sembey R, Hoda R, Grundy J. Emerging technologies in higher education assessment and feedback practices: A systematic literature review[J]. Journal of Systems and Software, 2024, 211: 111988. DOI: 10.1016/j.jss.2024.111988.