Self-efficacy, anxiety, AI utilization, and grammar knowledge among Indonesian and Taiwanese EFL learners

Engineering notes

Engineering notes will be added by the aipentium editorial team.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为大语言模型、生成式AI、ChatGPT相关技术、计算机视觉、深度学习等高价值论文补充中文说明。

Original abstract

This dataset supports a study on the relationships among self-efficacy, foreign language classroom anxiety, artificial intelligence application utilization, and grammar knowledge among undergraduate EFL learners, with gender examined as a moderator. Participants were 155 fourth-semester students enrolled in English education programmes at three higher education institutions, two in Indonesia and one in Taiwan. Data were collected through an online questionnaire covering self-efficacy, anxiety, and AI utilization, together with a proctored grammar test adapted from the TOEFL structure and written expression section. The dataset contains four measured constructs. Self-efficacy is represented by 11 items adapted from general self-efficacy frameworks. Anxiety is represented by 12 items adapted from the Foreign Language Classroom Anxiety Scale, covering communication apprehension, test anxiety, and fear of negative evaluation. AI utilization is represented by 15 items developed from the Technology Acceptance Model with an added readiness dimension, referring to tools such as ChatGPT, Grammarly, and QuillBot. Grammar knowledge is represented by a summed score across 40 dichotomously scored test items. All questionnaire items used a four-point response format, and reverse-scored items were recoded before analysis. Twenty-six additional columns hold the standardized product indicators used to model the two interaction terms. Also included are two SmartPLS 4 output files: the PLS algorithm results reporting loadings, reliability, average variance extracted, discriminant validity, R-square, f-square, and variance inflation factors; and the bootstrapping results reporting path coefficients, t statistics, p values, and indirect and total effects. No personally identifying information was collected. Ethical clearance was granted by the institutional research ethics committee, and all participants gave informed consent before data collection.

Links and sources

Comments