Analysis of Early Childhood Teachers' AI Adoption Levels: Stage-Specific Perceptions and Differences by Teacher Characteristics Based on Rogers' Diffusion of Innovation Theory

This study analyzes early childhood teachers' adoption and utilization of artificial intelligence (AI) based on Rogers’ Diffusion of Innovation Theory, focusing on differences by teacher characteristics (institution type, class taught, age, qualification). A survey was conducted with 395 teachers from early childhood education institutions in Seoul, Gyeonggi, and Incheon. Data were analyzed using SPSS 27.0 through descriptive statistics, ANOVA, t-tests, x² tests, and qualitative content analysis. Findings reveal that AI adoption was highest at the Decision stage and lowest at the Implementation stage. Among persuasion factors, perceived relative advantage was high, while perceived complexity was low. Second, adoption and utilization levels differed significantly by teacher characteristics, with higher levels observed among public kindergarten teachers, those teaching older children, teachers in their 20s–30s, and those with kindergarten certification. Third, teachers in private/home childcare settings, teaching infants, aged 50+, or holding only childcare certification showed lower adoption but higher AI use in caregiving contexts. Fourth, qualitative responses highlighted both enthusiasm for AI and the need for practical training and ethical guidance. This study offers foundational insights for differentiated AI training policies and effective AI integration in early childhood education