Teachers with higher AI literacy are significantly more likely to adopt and effectively integrate large language models into their K-12 classrooms, but success depends heavily on hands-on professional development, institutional support, and clear ethical guidelines.
Objective: The main goal of this study was to investigate the relationship between AI literacy and the adoption of large language models (LLMs) among K-12 teachers in North Texas, identify key factors influencing LLM usage, and determine which elements of professional development programs teachers perceive as most effective for improving AI literacy. The research aimed to provide evidence-based recommendations for designing targeted professional development initiatives that would equip teachers with the knowledge and skills necessary to effectively integrate AI technologies into their instructional practices.
Methods: This mixed-methods study employed both quantitative and qualitative approaches to comprehensively examine AI literacy and LLM adoption among educators. The quantitative component involved administering a survey adapted from the Meta AI Literacy Scale (MAILS) to 452 K-12 teachers across North Texas school districts. The survey assessed various dimensions of AI literacy including practical application skills, conceptual understanding, ability to detect AI-generated content, and ethical awareness. Additionally, the survey measured key technology adoption constructs based on the Unified Theory of Acceptance and Use of Technology (UTAUT) framework, including performance expectancy, effort expectancy, social influence, and facilitating conditions. The qualitative component consisted of semi-structured interviews with 20 purposively selected teachers from the survey participants, representing diverse grade levels, subject areas, and district types. These interviews explored teachers' lived experiences with AI integration, barriers and facilitators to adoption, and professional development needs. The study utilized established theoretical frameworks including UTAUT and Technological Pedagogical Content Knowledge (TPACK) to guide the analysis and interpretation of findings.
Key Findings: The study revealed several significant findings that illuminate the complex relationship between AI literacy and technology adoption in educational settings. A strong positive correlation (r = 0.59, p < .001) was found between overall AI literacy and LLM usage frequency, with the "Use & Apply AI" subscale showing the strongest relationship (r = 0.65, p < .001). Teachers demonstrated varied proficiency across AI literacy dimensions, scoring highest in AI Ethics (M = 3.04) but lowest in Detecting AI (M = 1.90), indicating a significant gap in critical evaluation skills. The regression analysis explained 52% of the variance in LLM usage, with AI literacy, performance expectancy, and facilitating conditions emerging as the strongest predictors. Qualitative findings revealed that while teachers recognized AI's potential benefits for personalizing instruction and streamlining administrative tasks, they faced substantial barriers including technical complexity, insufficient professional development, and ethical concerns about data privacy and algorithmic bias. Teachers overwhelmingly preferred hands-on, interactive professional development over traditional lecture-based formats, emphasizing the importance of peer collaboration and ongoing mentorship. Institutional support varied significantly across districts, with some teachers receiving robust backing while others operated in environments with unclear policies and limited resources.
Implications: These findings have significant implications for educational policy, practice, and teacher preparation programs. The research demonstrates that AI literacy serves as a critical gateway to effective technology integration, supporting the need for comprehensive professional development programs that emphasize practical application skills alongside ethical considerations. For policymakers, the study highlights the urgent need for standardized AI literacy competencies, strategic funding for AI integration initiatives, and robust ethical frameworks to address data privacy and algorithmic bias concerns. Educational institutions should prioritize hands-on training programs, establish professional learning communities, and ensure equitable access to AI resources across diverse school settings. The findings also suggest that pre-service teacher education programs must incorporate AI literacy and ethical considerations into their curricula to prepare future educators for an AI-driven educational landscape. The identification of performance expectancy as a key predictor indicates that demonstrating clear instructional benefits is crucial for promoting adoption, while the importance of facilitating conditions underscores the need for sustained institutional support and infrastructure investment.
Limitations: Several limitations should be considered when interpreting these findings. The study was geographically limited to North Texas districts, which may limit the generalizability of results to other regions with different technological infrastructure, policy frameworks, or demographic characteristics. The research relied heavily on self-reported data, which may be subject to response bias, particularly from teachers who volunteered to participate and may have had pre-existing interest in or positive attitudes toward AI technologies. The cross-sectional design provides a snapshot of current practices but cannot establish causal relationships or track changes in AI literacy and adoption over time. Additionally, the rapidly evolving nature of AI technologies means that findings may have limited longevity as new tools and applications emerge. The study's focus on K-12 education may not extend to other educational levels, and the emphasis on LLMs specifically may not capture the full spectrum of AI applications in education.
Future Directions: The study identifies several important avenues for future research that would advance understanding of AI integration in education. Longitudinal studies are needed to track how AI literacy and adoption patterns evolve over time, providing insights into the sustainability of professional development interventions and long-term impacts on teaching practices and student outcomes. Comparative research across different geographic regions, educational levels, and institutional contexts would help identify universal principles versus context-specific factors that influence AI adoption. Future studies should explore the direct impact of AI-integrated instruction on student learning outcomes, engagement, and critical thinking skills across various subjects and grade levels. Research into specific factors such as ethical implementation strategies, administrative leadership styles, and the effectiveness of different professional development models would provide more targeted guidance for practitioners. Additionally, interdisciplinary collaboration with AI researchers, ethicists, and cognitive scientists could yield valuable insights into designing more effective and responsible AI educational tools and implementation strategies.
Title and Authors: "Enhancing Educator AI Literacy for Effective Use of Large Language Models: Exploring Teacher Usage and Professional Development" by Doreen Mayrell.
Published On: May 2025
Published By: University of Louisiana Monroe (Doctoral Dissertation)