Adoption of Artificial Intelligence among Pre-Service Teachers of Administration Program in Indonesia: Confirmatory Composite Analysis
Choirul Anam *
Office Administration Education Study Program, Faculty of Economics and Business, Universitas Negeri Malang, Jl. Semarang No. 5 Malang, Indonesia.
Madziatul Churiyah
Office Administration Education Study Program, Faculty of Economics and Business, Universitas Negeri Malang, Jl. Semarang No. 5 Malang, Indonesia.
Sigit Permansah
Office Administration Education Study Program, Faculty of Economics and Business, Universitas Negeri Malang, Jl. Semarang No. 5 Malang, Indonesia.
Della Rulita Nurfaizana
Office Administration Education Study Program, Faculty of Economics and Business, Universitas Negeri Malang, Jl. Semarang No. 5 Malang, Indonesia.
Wama Halfi Al Imani
Office Administration Education Study Program, Faculty of Economics and Business, Universitas Negeri Malang, Jl. Semarang No. 5 Malang, Indonesia.
Siti Nurun Chumairoh
Office Administration Education Study Program, Faculty of Economics and Business, Universitas Negeri Malang, Jl. Semarang No. 5 Malang, Indonesia.
*Author to whom correspondence should be addressed.
Abstract
Aims: This study examined the determinants of the Adoption of Artificial Intelligence (ATAI) among pre-service teachers of administration programs in Indonesia by validating the effects of Technological Readiness (TR), Digital Literacy (DL), and Institutional Support (IS) through Confirmatory Composite Analysis within the Partial Least Squares Structural Equation Modeling (PLS-SEM) framework.
Methodology: A quantitative survey-based design applying PLS-SEM was employed to identify and validate both reflective and composite constructs of AI adoption. The research was conducted across teacher education institutions in Indonesia from January to June 2025. A total of 103 pre-service teachers enrolled in administration programs across Indonesian teacher education institutions participated in this study. Data were collected through a five-point Likert questionnaire measuring TR, DL, IS, and ATAI. Analyses were performed using SmartPLS 4 with the two-stage embedded and repeated indicators approaches. Measurement validity and reliability were confirmed via outer loadings, Average Variance Extracted (AVE), Composite Reliability (CR), and HTMT ratio. Structural relationships were tested using bootstrapping (5,000 resamples) with path coefficients, effect sizes (f²), R², and Q² statistics.
Results: DL strongly affected ATAI (β = 0.565; P = 0.000; f² = 0.401), while IS had a smaller but significant effect (β = 0.211; P = 0.014; f² = 0.056). TR showed no direct effect (β = –0.106; P = 0.137; f² = 0.021) but influenced DL (β = –0.227; P = 0.002; f² = 0.092). IS was the strongest predictor of DL (β = 0.630; P = 0.000; f² = 0.680). The model demonstrated moderate predictive accuracy, with substantial explained variance for ATAI and DL.
Conclusion: DL and IS are the principal predictors of AI adoption among pre-service teacher of administration program, while TR operates indirectly through DL. Strengthening digital literacy and institutional support is essential to promote sustainable AI integration in teacher education.
Keywords: Artificial intelligence adoption, confirmatory composite analysis, digital literacy, institutional support, technological readiness, pre-service teachers