The Effects of Prompt Scaffolding on Learning to Write Arguments with ChatGPT
Open accessAbstract
Although ChatGPT can be useful for writing, it is not certain whether ChatGPT is beneficial for learning to write particularly when ChatGPT carries out the task instead of learners.To enhance the positive effect of ChatGPT in learning to write, teachers should provide students appropriate scaffoldings on creating prompts, which are crucial in interacting with ChatGPT.This study aims to investigate the effects of prompt scaffoldings on learning to write arguments with ChatGPT.In this study, 179 Korean high school students participated, and they were assigned into experiment and control groups.This study found that the experiment group that received prompt scaffoldings outperformed the control group in regards to self-efficacy, interest, knowledge, and performance in argumentative writing.This study implies that ChatGPT can be beneficial for learning only when learners effectively interact with it.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Studies used AI interventions to address common barriers such as limited teacher feedback, low student motivation for revision, or insufficient opportunities for creative expression.
“These studies used AI interventions to address common barriers such as limited teacher feedback, low student motivation for revision, or insufficient opportunities for creative expression [81-87].”
From A Systematic Review of Generative AI in K–12: Mapping Goals, Activities, Roles, and Outcomes via the 3P Model (2025) · cites this paper as [86]
The Procedure describes an AI-assisted argumentative-writing intervention, and the Discussion reports improved student interest with prompt scaffolding. The Introduction says ChatGPT can provide personalized feedback and help students express ideas. The paper does not identify limited teacher feedback, low motivation specifically for revision, or insufficient opportunities for creative expression as barriers the intervention addressed. (LLM, conf=0.82)
Experimental — accuracy not yet measuredLarge language models support literary writing, story crafting, and creative text generation; students iteratively revise content, adjust style, and sharpen narrative logic, often under teacher guidance.
“Large language models support literary writing, story crafting, and creative text generation; students iteratively revise content, adjust style, and sharpen narrative logic-often under teacher guidance to rework initial plots toward intended effects [84,86,139,144,145].”
From A Systematic Review of Generative AI in K–12: Mapping Goals, Activities, Roles, and Outcomes via the 3P Model (2025) · cites this paper as [86]
The Procedure and Data collection sections describe students using ChatGPT to develop argumentative essays, with teacher-led activities, prompt scaffolding, and feedback on their interactions. They do not address literary writing, story crafting, style or narrative-logic revisions, or reworking plots toward intended effects. Page numbers are not available in our copy. (LLM, conf=0.86)
Experimental — accuracy not yet measuredAI provides strong scaffolds for core practices, markedly improving the quality and conventionality of learner outputs-for example, enhancing grammatical accuracy and argumentative coherence in writing, and supporting problem-solving and experimental design in science and mathematics-thereby helping learners meet baseline practice demands more efficiently.
“First, AI provides strong scaffolds for core practices, markedly improving the quality and conventionality of learner outputs-for example, enhancing grammatical accuracy and argumentative coherence in writing [86,87,145], and supporting problem-solving and experimental design in science and mathematics [77,88,93,103,147]-thereby helping learners meet baseline practice demands more efficiently.”
From A Systematic Review of Generative AI in K–12: Mapping Goals, Activities, Roles, and Outcomes via the 3P Model (2025) · cites this paper as [86]
The Performance section reports a significant improvement in argumentative-writing scores for students given ChatGPT prompt scaffolding compared with students using ChatGPT without it. The Data analysis section says scores covered content, form, and expression, but does not report separate gains in grammatical accuracy or argumentative coherence; the paper also does not establish that students met writing demands more efficiently. The science and mathematics examples have separate citations and are not attributed to this source. (LLM, conf=0.86)
Experimental — accuracy not yet measuredAffective gains often translate into identity-level shifts: students' self-efficacy and domain confidence increase-particularly in programming, mathematics, and writing-with corresponding reductions in learning anxiety.
“These affective gains often translate into identity-level shifts: students' self-efficacy and domain confidence increase-particularly in programming [95], mathematics [77,103], and writing [86]-with corresponding reductions in learning anxiety [80,141].”
From A Systematic Review of Generative AI in K–12: Mapping Goals, Activities, Roles, and Outcomes via the 3P Model (2025) · cites this paper as [86]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
The publisher hasn’t deposited this paper’s reference list.
Research works printed but not linked: 7
- Bozkurt, A. (2023). Generative artificial intelligence (AI) powered conversational educational agents: The inevitable paradigm shift. Asian Journal of Distance Education, 18(1), 198-204.
- Goldman, S. R., Greenleaf, C., Yukhymenko-Lescroart, M., Brown, W., Ko, M. L. M., Emig, J. M., ... & Britt, M. A. (2019). Explanatory modeling in science through text-based investigation: Testing the efficacy of the project READI intervention approach. American Educational Research Journal, 56(4), 1148-1216.
- Ibrahim, H., Asim, R., Zaffar, F., Rahwan, T., & Zaki, Y. (2023). Rethinking homework in the age of artificial intelligence. IEEE Intelligent Systems, 38(2), 24-27.
- Rahman, M. M., & Watanobe, Y. (2023). ChatGPT for education and research: Opportunities, threats, and strategies. Applied Sciences, 13(9), 5783.
- Lo, L. S. (2023). The CLEAR path: A framework for enhancing information literacy through prompt engineering. The Journal of Academic Librarianship, 49(4), 102720.
- Muñoz, S. A. S., Gayoso, G. G., Huambo, A. C., Tapia, R. D. C., Incaluque, J. L., Aguila, O. E. P., ... & Arias- Gonzáles, J. L. (2023). Examining the impacts of ChatGPT on student motivation and engagement. Social Space, 23(1), 1-27.
- Shell, D. F., Murphy, C. C., & Bruning, R. H. (1989). Self-efficacy and outcome expectancy mechanisms in reading and writing achievement. Journal of educational psychology, 81(1), 91.
Cited by
Sorted by citations
Prompt Engineering in Education: A Systematic Review of Approaches and Educational Applications
Qian, Y. (2025) · Journal of Educational Computing Research · doi:10.1177/07356331251365189 · Cited by 91 · see citing works
“Jang et al. (2024) demonstrated this with ChatGPT in K – 12 education, where students iteratively re fi ned prompts to improve their argumentative writing.”
AI in the classroom: Exploring students’ interaction with ChatGPT in programming learning
Güner, H., & Er, E. (2025) · Education and Information Technologies · doi:10.1007/s10639-025-13337-7 · Cited by 117 · see citing works
“On the other hand, scaffolding, a pedagogical approach that involves providing temporary support to learners as they develop new skills, might involve instructors offering sample prompts and clear examples of how to interact with the AI (Jang et al., 2024).”
Lin, X., & Tan, H. (2025) · Systems · doi:10.3390/systems13100840 · Cited by 22 · see citing works
Bai, S., Yeung, S. S., & Lo, C. K. (2026) · Interactive Learning Environments · doi:10.1080/10494820.2026.2649547 · Cited by 16 · see citing works
Silva, F. G., & Aranha, E. H. D. S. (2025) · Anais do ... Simpósio Brasileiro de Informática na Educação/Anais do Simpósio Brasileiro de Informática na Educação · doi:10.5753/sbie.2025.12807 · Cited by 6 · see citing works
“Essa abordagem combina Socratic prompting , scaffolding de dicas e verificac¸˜ao conceitual com casos de teste internos, al´em de explicac¸˜oes con-ceituais gen´ericas, validac¸˜ao algor´ıtmica rigorosa, perguntas reflexivas e tom motivador [Ding et al. 2024, Jang et al. 2024].”
Leveraging generative AI to facilitate peer feedback in collaborative argumentation learning
Chang, Y., Liu, Q., Lu, Y., & Miao, E. (2026) · International Journal of Educational Technology in Higher Education · doi:10.1186/s41239-026-00586-w · Cited by 4 · see citing works
Zhao, Y., Bae, H., Park, J., & Walton, D. (2025) · Proceedings. · doi:10.22318/icls2025.294906 · Cited by 1 · see citing works
Chen, X., Fu, Y., & Zhang, Z. H. (2026) · International Conference on Educational and Information Technology · doi:10.1109/iceit68991.2026.11521366
“Evidence from digital learning shows prompt-based scaffolding can improve achievement [12], and emerging work in LLM-mediated learning suggests structured prompt scaffolds can shape learners’ products and engagement in ChatGPT-supported tasks [13].”
Supporting systems thinking in environmental science education with large language model chatbots
Nguyen, H., Valdez, E., Pei, J., Ludovise, S., & Santagata, R. (2026) · International Journal of Science Education · doi:10.1080/09500693.2026.2695951
- Authors
- Jang, J., Eun, S., Lee, H., Choi, J., & Cho, Y. H.
- Affiliations
- Seoul National University (KR)
- Venue
- Proceedings of the International Conference of the Learning Sciences
- Published
- 2024-06-10
- Pages
- 1502-1505
- ISSN
- 1814-9316 (ISSN-L)
- 1819-0138
- OpenAlex
- W4399489410
- Cited by
- 9
- References
- 0
- Citations by year
- 2026: 4 · 2025: 5
- FWCI
- 3.43
- Type
- conference paper
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- Proceedings. (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Information Systems › Artificial Intelligence in Education
- Social Sciences › Arts and Humanities › Language and Linguistics › EFL/ESL Teaching and Learning
- Social Sciences › Psychology › Developmental and Educational Psychology › Innovative Teaching and Learning Methods
- Keywords
- ChatGPT, argumentative writing, AI-assisted writing, self-efficacy, student interest, writing performance, high school students, human-AI interaction
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
