This study investigates the potential of Large Language Models (LLMs), specifically GPT-4, in analyzing classroom dialogue. Traditional qualitative methods are time-consuming and labor-intensive. The researchers compared manual coding of classroom dialogues from middle school math and Chinese classes with GPT-4 outputs, evaluating time efficiency, inter-coder agreement, and reliability. Results showed significant time savings and high coding consistency between the model and human coders, with minor discrepancies, highlighting LLMs' potential in teaching evaluation.

Yun Long, Haifeng Luo, Yu Zhang