Hierarchical Characteristics and Cultivation Mechanism of Teachers’ Data Literacy in the Context of Digital Transformation

Abstract

Efficient data acquisition, intelligent analysis, and adaptive decision-making are fundamental capabilities for modern digital engineering systems. To address the limitations of homogeneous teacher training under digital transformation, this study proposes a hierarchical diagnosis and progressive cultivation framework for teacher data literacy based on multimodal assessment and latent profile analysis. A context-embedded evaluation tool integrating situational judgment, behavioral retrospection, and ethical dilemma tasks was developed and applied to 1,263 teachers to identify four distinct competency levels. Based on the diagnosed developmental characteristics, a four-stage cultivation mechanism incorporating intelligent recommendation, competency dashboards, and micro-certification was established and iteratively optimized through design-based research. Experimental results demonstrate that the proposed framework achieves an upward competency migration rate of 60.7%, significantly outperforming conventional training approaches, while the competency dashboard provides the greatest independent contribution to performance improvement. The hierarchical and evidence-driven strategy effectively enhances personalized learning trajectories and supports scalable digital capability development. Beyond educational applications, the proposed multimodal profiling and adaptive optimization framework provides a practical reference for intelligent information processing, data fusion, human-centered decision support, and adaptive sensing architectures, offering potential guidance for data-driven optimization and intelligent management in electromagnetic wave analysis, antenna-enabled sensing systems, and propagation-oriented engineering applications.