ChatGPT’s pedagogical approach and the potential for hidden curriculum through school context proxies
American Journal of STEM Education · 16.06.2026 · 24 · 157-176
DOI: 10.32674/nyqmx528 ↗Kurz gefasst
Fachartikel in der Zeitschrift American Journal of STEM Education, online erschienen am 16.06.2026. Verfasst von Laurie Rubel und 1 weiteren Autor:innen. Die Arbeit nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
This study investigates ChatGPT-4's pedagogical approach through mathematics lesson plans and examines how school context descriptors influence its output. We generated 45 individual eighth-grade probability lesson plans across three conditions: a baseline prompt, a "Successful School" context, and a "Struggling School" context. Content analysis of the lesson plans and 368 teacher-action verbs revealed that the LLM favored a direct instruction model (across all conditions) and shifted in specific ways in response to school descriptors. "Successful School" prompts generated lessons with additional mathematics concepts and advanced technology integration, whereas "Struggling School" prompts produced lessons with more monitoring of students’ progress, reinforcement activities, and attention to socioemotional support. These findings suggest that LLMs may replicate patterns of educational stratification.
Zitieren
APA
Rubel, L., & Goidel, S. (2026). ChatGPT’s pedagogical approach and the potential for hidden curriculum through school context proxies. American Journal of STEM Education, 24, 157-176. https://doi.org/10.32674/nyqmx528
Vancouver
Rubel L, Goidel S. ChatGPT’s pedagogical approach and the potential for hidden curriculum through school context proxies. American Journal of STEM Education. 2026;24:157-176. doi:10.32674/nyqmx528
BibTeX
@article{Rubel2026,
author = {Rubel, Laurie and Goidel, Shimrit},
title = {ChatGPT’s pedagogical approach and the potential for hidden curriculum through school context proxies},
journal = {American Journal of STEM Education},
year = {2026},
volume = {24},
pages = {157--176},
doi = {10.32674/nyqmx528}
}
EUVIMED