Development and validation of the Artificial Intelligence Literacy Scale for Teachers (AILST)

The integration of AI in education highlights the significance of Teachers' AI Literacy (TAIL). Existing assessment tools, however, are hindered by incomplete indicators and a lack of practicality for large-scale application, necessitating a more systematic and credible evaluation method. This study is based on a systematic literature review and aimed to develop the Artificial Intelligence Literacy Scale for Teachers (AILST). A random sampling method was used to collect 604 valid samples, which were analyzed using Exploratory Factor Analysis (EFA), Confirmatory Factor Analysis (CFA), and the Random Forest Model (RFM). Through this process, the scale was refined to 36 items, focusing on AI perception, knowledge and skills, applications and innovation, and ethics. EFA identified four primary factors and eliminated incongruent items for theoretical coherence. CFA confirmed the robust fit of the AILST structure, with indices such as the Absolute Fit Index (AFI), Incremental Fit Index (IFI), and Parsimonious Fit Index (PFI) meeting standard criteria. RFM was used to rank the characteristics of the four elements of TAIL and their subordinate indicators, further validating their importance. This study presents a validated AILST with good reliability and validity, offering a refined tool for assessing TAIL and demonstrating strong theoretical and practical value.