Modeling K12 Teachers’ Online Teaching Competency and Its Predictive Relationship with Performance—A Mixed-Methods Study Based on Behavioral Event Interviews

This study constructs and validates a multidimensional online teaching competency model for K12 teachers through an integrated mixed-methods design. Combining behavioral event interviews (<i>n</i> = 38) with large-scale psychometric evaluation (<i>n</i> = 4378), we identified six hierarchically organized competency dimensions encompassing 29 measurable elements. The model differentiates between 12 discriminative competencies and 17 baseline competencies, further categorized into explicit (knowledge, technical, instructional, management) and implicit (achievement orientation, individual traits) dimensions. Exploratory and confirmatory analyses validated the model's robust multidimensional structure (CFI = 0.923, TLI = 0.914, RMSEA = 0.042). Structural equation modeling revealed significant competency-performance linkages, with 10 of 12 hypothesized paths attaining statistical significance (<i>p</i> < 0.05). Management competencies emerged as the strongest predictor of both process (β = 0.37) and outcome performance (β = 0.29), followed by instructional competencies (β = 0.31 and 0.24 respectively). The model provides empirically grounded guidance for developing online teaching norms, competency-based teacher training programs, and performance evaluation systems.