Evaluation of MagicSchool AI-Generated Elementary Teaching Materials: A Self-Study

Abstract

Recent developments in Artificial Intelligence (AI) have resulted in many possible tools, such as MagicSchool AI, that educators could use in their practice. Possible outcomes of using AI in their practice could be reduced teacher burnout, reduced time for developing materials, and increased time for other aspects of their role as an educator. Thus, the purpose of this mixed methods self-study is to investigate my experiences using MagicSchool AI to generate elementary classroom materials for a two-week integrated instructional unit, the quality of the AI-generated materials, and my pedagogical decisions surrounding using those materials in my personal teaching practices. I collected the following data: AI-generated lesson plans, lesson materials, and assessments; evaluation of AI-generated materials; personal journal entries; and recorded and transcribed critical friend conversations. Quantitative data analysis included descriptive statistics, and qualitative data analysis included open coding and axial coding to identify themes. The findings address experiences utilizing MagicSchool AI to generate classroom materials, professional development opportunities, best AI practices for using in the elementary classroom, and my pedagogical decisions based on the quality of the materials. Implications for future practice are discussed for pre-service and in-service teachers as they use AI to generate classroom materials and future research.