Survey Data on Generative AI Use and Learning Agency among Pre-service Teachers

It includes 1,320 pre-service teachers from six universities in Sichuan Province, Shaanxi Province, Guangxi Zhuang Autonomous Region, and Guangdong Province, China. The data were collected using a 5-point Likert scale through an online questionnaire, with all items set as mandatory; therefore, there are no missing values.

The variables include demographic and learning background information, generative AI use in education, human-AI collaboration quality, AI dependence, learning agency, AI ethics literacy, teacher guidance, and school norms. The dataset can be used to examine the dual-path mechanism through which generative AI use affects learning agency: promoting learning agency through human-AI collaboration quality and weakening learning agency through AI dependence.