Abstract

This research aimed to: 1) develop the components and indicators of a blended learning instructional model based on Collaborative Game-Based Learning with Generative AI (the BCG-AI model) to enhance AI literacy, and; 2) examine the model fit of these components and indicators with empirical data. The research employed a two-phase mixed-methods research design. Phase 1 developed the components and indicators of the BCG-AI model based on structured interviews with 12 experts selected through purposeful sampling based on specific expertise criteria. Phase 2 examined the model fit of these components and indicators with empirical data through questionnaires. The sample consisted of 301 university lecturers in Thailand, determined using a 10:1 parameter ratio and selected through stratified random sampling. The research instruments were structured interview and questionnaires, with the validation results achieved an IOC score of 1.00 for all items and a reliability coefficient (Cronbach’s alpha) of 0.97. Data were analyzed using confirmatory factor analysis (CFA) via AMOS software. In terms of findings: 1) the expert interviews identified five components of the BCG-AI model (1. Instructor, 2. Content, 3. Environment, Technology and Media, 4. Learning Management, and 5. Measurement and Evaluation, comprising 30 indicators in total); 2) the model fit indices showed strong alignment with the empirical data: χ² = 119.662, df = 101, χ²/df = 1.185, p = 0.099, GFI = 0.970, CFI = 0.997, NFI = 0.983, RMR = 0.015, and RMSEA = 0.027. These results validated the model’s structural integrity and effectiveness for enhancing AI literacy in higher education, supporting Thailand’s ‘4.0 Vision’ to transfirm the country into a global digital hub.

This work is licensed under a Creative Commons Attribution 4.0 License.