Development, Network Analysis, and Validation of the Ethical-Spiritual Algorithmic Trust Calibration Scale (ES-ATCS):...

The purpose of this study is to design and evaluate the psychometric properties of the Ethical-Spiritual Algorithmic Trust Calibration Scale (ES-ATCS) among teachers. The primary aim was to develop a reliable and valid instrument to understand how teachers navigate trust, ethical accountability, and spiritual coherence when engaging with AI-driven educational technologies. The study was conducted in two main phases. Phase 1 comprised item generation, 12 specialist expert reviews (Lawshe CVR cutoff = 0.56), and pilot testing (n = 35 teachers), which reduced the item pool through CVR/I-CVI filtering and impact-score analyses. Phase 2 involved a cross-sectional sample of 666 teachers, which was randomly split into two halves for EFA and CFA. EFA and exploratory graph analysis suggested a coherent six-factor structure accounting for 63.20% of total variance, with the Spiritual Coherence Perception factor explaining 12.26% of the variance. Iterative CFA supported a final 48-item first- and second-order six-factor model with acceptable fit (RMSEA < .08; CFI, TLI > .90; SRMR < .08) and standardized loadings > .40. Measurement invariance was acceptable across gender and teaching experience. Reliability (Cronbach's α .898-.953, McDonald's ω .848-.953, CR .898-.954) and stability were strong: ICCs ranged from .755 to .853. Convergent (AVE .501-.940) and discriminant validity (Fornell-Larcker) were acceptable. Network EGA identified 43 nodes and 209 edges with six communities; centrality indices identified salient items. The 48-item ES-ATCS is culturally sensitive and psychometrically sound for measuring teachers' ethical-spiritual trust in AI.