Enhancing Chinese language teaching proficiency in primary schools: The role of AI driven personalized learning
Year: 2025
Type: article
Study design: Observational Study
Abstract: This study examines how AI-driven personalized learning improves the Chinese language proficiency of primary school students in Shenzhen, China. Through a mixed-method study involving 378 participants from 10 schools, the study showed that there was a statistically significant improvement (p<0.01) in language achievement, especially for lower achiever and non-native Mandarin speakers. The results showed a positive trend: the mean value of artificial intelligence application (AU) was 3.777 (SD 1.182), the mean value of personalized learning (PL) was 3.769 (SD 1.146), and the mean value of language ability improvement (CLI) was 3.794 (SD 1.165). The impact on variable shown that: 1) Personalized learning has a positive impact on the improvement of Chinese proficiency (β = 0.817, R2 = 0.668) indicated that personalized learning significantly improved the language learning effect. 2) AI utilization has a positive impact on improvement of Chinese proficiency (β = 0.762, R2 = 0.580). The results showed that the use of artificial intelligence was significant for Chinese learning, and regression weight (0.762) and statistical significance (p < 0.001). These findings highlight the effectiveness of customized instruction and AI-assisted learning in improving language education outcomes. 3) AI utilization has a positive impact on Personalized Teaching. (β =0.803, R2 =0.644). In line with Sustainable Development Goals of Quality education) and Reducing inequalities, the study provides a scalable model for linguistically diverse urban schools. The findings demonstrate that AI personalization can simultaneously improve academic performance and bridge learning gaps, advocating balanced technology adoption and complementing humanist pedagogy. These findings have important implications for Chinese wisdom education policies in society and similar technology-involved regions around the world.
Authors: Ruoying Wang, Supot Rattanapun, Rawi Buaduang, Alan Rober White, Tubagus Darodjat
Institution: Rajamangala University of Technology Krungthep
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4414549096
FWCI:
Cites:
Cited by:
Related to: