A Study on the Influence Mechanism of Rural English Teachers’ AI Teaching Self-Efficacy from the Perspective of TPACK Theory

Ziying Ling; Anqi Wen; Xinyun Li; Yujian Huang; Yongyi Mo

doi:10.14201/fde.24230

Authors

Ziying Ling School of Foreign Studies, South China Normal University, Guangzhou, China Author https://orcid.org/0009-0002-6241-3911
Anqi Wen School of Foreign Studies, South China Normal University, Guangzhou, China Author https://orcid.org/0009-0009-2904-4799
Xinyun Li School of Foreign Studies, South China Normal University, Guangzhou, China Author https://orcid.org/0009-0009-5752-9515
Yujian Huang School of Foreign Studies, South China Normal University, Guangzhou, China Author
Yongyi Mo School of Foreign Studies, South China Normal University, Guangzhou, China Author

DOI:

https://doi.org/10.14201/fde.24230

Keywords:

TPACK Theory, Artificial Intelligence, Self-Efficacy, Influence Mechanism, Rural English Teachers

Abstract

Based on the Technological Pedagogical Content Knowledge (TPACK) theoretical framework, this study employs a mixed-methods approach—combining questionnaire surveys and semi-structured interviews—to systematically investigate the current development status and internal influence mechanisms of TPACK capabilities and Artificial Intelligence (AI) teaching self-efficacy among 115 rural primary and secondary school English teachers in western Guangdong. The findings indicate that the overall TPACK capabilities and AI teaching self-efficacy of rural English teachers are at an upper-middle level. Across the TPACK dimensions, teachers exhibit an uneven developmental profile characterized by relatively solid foundational theoretical knowledge but weak higher-order capabilities in the deep integration of technology and subject disciplines. Regarding AI teaching self-efficacy, their mastery of basic AI knowledge significantly surpasses their confidence in higher-order practical applications such as teaching assessment. Furthermore, linear regression analysis confirms that TPACK capability has a significant positive predictive effect on AI teaching self-efficacy, explaining 41.2% of its variance. This study not only provides empirical evidence for enhancing rural English teachers' AI teaching integration capabilities and overcoming technology application bottlenecks, but also offers vital practical references for promoting the equitable digital development of rural education and innovating teacher training models.