Abstract
This study investigates the relationship between students’ prompt literacy and the development of their scientific writing skills within an AI-supported learning environment. Conducted as part of a larger quasi-experimental research project, the study involved 6th-grade students engaged in science lessons designed using the ChatGPT-supported 5E instructional model. While students received guidance on effective AI interaction, specific attention was given to improving their ability to write purposeful and structured prompts to support scientific writing tasks. To evaluate students’ prompt literacy, a researcher-developed Prompt Writing Rubric was used to assess clarity, specificity, scientific relevance, and structural quality of the prompts generated by students. Scientific writing performance was measured using a validated rubric focusing on organization, vocabulary use, scientific reasoning, and content accuracy. Findings indicate a significant correlation between students’ ability to formulate effective prompts and their improvement in scientific writing. Students who demonstrated higher scores in prompt writing were also more likely to produce coherent, well-structured, and scientifically accurate texts. Qualitative data gathered through classroom observations and student reflections further highlighted how structured prompt training enhanced students’ metacognitive awareness and engagement during AI-supported tasks. This study provides practical implications for integrating prompt literacy into AI-mediated science education. The findings suggest that teaching students how to communicate effectively with AI systems—through well-crafted prompts—can serve as a catalyst for developing deeper scientific understanding and more effective written expression.
Details
Presentation Type
Theme
KEYWORDS
AI, Prompt, Scientce, Writing
