Language Teachers' Efficacy in Using Generative Language AI in Instruction: A Mixed-Method Explanatory Sequential Study

Therese Marie Bedonia, Marijo D. Chua

Artificial Intelligence (AI), particularly generative language AI, is rapidly transforming education. This study explored the efficacy of language teachers in integrating generative language AI into instruction within a large schools division in the Philippines during the 2024–2025 school year. Using a mixed-method explanatory sequential design, data were collected through surveys and focus group discussions. Quantitative data were analyzed using mean, standard deviation, and the Mann-Whitney U test, while qualitative data underwent thematic analysis using Lichtman’s 3C’s model. Findings showed that language teachers demonstrated high efficacy in using generative AI, particularly in four domains: adoption, application, adaptation, and ethical consideration. Notably, there were no significant differences in efficacy based on demographic factors such as age, curricular category, educational attainment, sex, and teaching experience. The qualitative phase revealed four themes: AI as a catalyst for teaching efficiency, a driver of teacher productivity and work-life balance, a means of democratizing access to educational resources, and a double-edged sword in writing instruction. The meta-inference demonstrated alignment between the quantitative and qualitative results, highlighting teachers’ readiness and optimism toward AI integration. However, challenges such as student over-dependence on AI and limited internet connectivity remain significant obstacles. To support effective AI adoption, stakeholders must improve access to technology and implement strategies to mitigate negative impacts.