Measuring Acceptance of Age-Tiered AI Literacy Guidebooks: A Developmentally Informed Study of K-12 Students and Teachers

Texto completo

As generative artificial intelligence (AI) becomes increasingly integrated into K-12 education, there is a pressing need for AI literacy initiatives to incorporate learning resources that are both age-appropriate and ethically grounded. This study examined acceptance of the AI Learning Application Guidebook, developed in Elementary and Advanced editions for learners aged 9–12 and 13–18, respectively. The Elementary edition features scaffolded, platform-based, non-generative activities, whereas the Advanced edition includes authentic applications, ethical reasoning exercises, and supervised engagement with generative AI. The guidebooks were evaluated in Taiwanese classrooms by 831 participants (794 students and 37 teachers). For the student sample, split-sample exploratory and confirmatory factor analyses supported a four-factor model encompassing Performance Expectancy, Effort Expectancy, Perceived Playfulness, and Behavioral Intention. Measurement invariance was supported across the two student editions, while teacher data provided preliminary cross-role psychometric evidence rather than fully powered invariance conclusions. Group-specific models showed the largest estimated path coefficient between perceived playfulness and behavioral intention in both cohorts. The performance-expectancy-to-behavioral-intention path was statistically significant in the Advanced cohort but nonsignificant in the Elementary cohort, although substantial overlap between perceived playfulness and behavioral intention limits interpretation of these structural paths. Qualitative findings highlighted participants' perceived need for age-sensitive, multimodal, bilingual, and ethically explicit materials. The study did not measure AI literacy achievement, learning gains, actual adoption, or sustained use; therefore, the findings indicate short-term acceptance and perceived readiness for further use after initial guided exposure, not implementation effectiveness or learning outcomes. The study provides a context-specific material-level operationalization of established acceptance constructs and student-based psychometric evidence to inform early-stage refinement of age-tiered AI literacy materials.