<scp>AI</scp>‐Facilitated Article Revisions for Primary School Students With Writing Difficulties: Effects of a Large Language Model‐Based <scp>PDRPE</scp> Approach on Writing Performance, Attitude and Anxiety

ABSTRACT Background Primary school is key to developing writing skills. However, students may face challenges in identifying and revising articles due to weak language skills, organisational thinking, comprehension, and analytical skills. Therefore, improving the writing skills of primary school students with writing difficulties has become important. Objective Conventional writing instruction often lacks immediate, targeted feedback, whereas large language models (LLMs) offer personalised support to help students with writing difficulties improve their writing quality. Method This study proposed an LLM‐supported PDRPE (plan, draft, revise, present, and evaluate) approach and explored its impact on the writing performance, learning attitude, and writing anxiety of primary school students with writing difficulties. Through screening, 56 participants were recruited. Among them, 29 students in the experimental group adopted the LLM‐based PDRPE approach, while 27 students in the control group used the C‐PDRPE approach. Results and Conclusions The results showed that although there was no significant difference in the number of words, the LLM‐based PDRPE approach better facilitated students' writing performance regarding the basic competence index (including the level of detail, expression, rhetoric, and vocabulary, and transitions) and other hard measurement indexes (including the number of sentences, typo rate, and punctuation error rate). Additionally, the LLM‐based PDRPE approach better improved students' writing attitude and alleviated writing anxiety. Moreover, the interview results indicated that the LLM‐based PDRPE approach is effective in supporting the revision process. Therefore, this study provides valuable references and insights for future research on the application of LLM and effective instruction for students with writing difficulties.

Authors

Institutions

Publication Details

Journal
Journal of Computer Assisted Learning
Published
2025-09-21
DOI
https://doi.org/10.1111/jcal.70131
Primary Topic
Writing and Handwriting Education
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

AI‐Facilitated Article Revisions for Primary School Students With Writing Difficulties: Effects of a Large Language Model‐Based PDRPE Approach on Writing Performance, Attitude and Anxiety

Mingyi Li, Gwo‐Jen Hwang, Xinli Zhang, Lailin Hu et al.
Journal of Computer Assisted Learning
Writing and Handwriting Education
article

AI‐Facilitated Article Revisions for Primary School Students With Writing Difficulties: Effects of a Large Language Model‐Based PDRPE Approach on Writing Performance, Attitude and Anxiety

Mingyi Li, Gwo‐Jen Hwang, Xinli Zhang, Lailin Hu, Ru Huang, Yun‐Fang Tu, Yuchen Chen, Ruihua Zhang
article en

Abstract

ABSTRACT Background Primary school is key to developing writing skills. However, students may face challenges in identifying and revising articles due to weak language skills, organisational thinking, comprehension, and analytical skills. Therefore, improving the writing skills of primary school students with writing difficulties has become important. Objective Conventional writing instruction often lacks immediate, targeted feedback, whereas large language models (LLMs) offer personalised support to help students with writing difficulties improve their writing quality. Method This study proposed an LLM‐supported PDRPE (plan, draft, revise, present, and evaluate) approach and explored its impact on the writing performance, learning attitude, and writing anxiety of primary school students with writing difficulties. Through screening, 56 participants were recruited. Among them, 29 students in the experimental group adopted the LLM‐based PDRPE approach, while 27 students in the control group used the C‐PDRPE approach. Results and Conclusions The results showed that although there was no significant difference in the number of words, the LLM‐based PDRPE approach better facilitated students' writing performance regarding the basic competence index (including the level of detail, expression, rhetoric, and vocabulary, and transitions) and other hard measurement indexes (including the number of sentences, typo rate, and punctuation error rate). Additionally, the LLM‐based PDRPE approach better improved students' writing attitude and alleviated writing anxiety. Moreover, the interview results indicated that the LLM‐based PDRPE approach is effective in supporting the revision process. Therefore, this study provides valuable references and insights for future research on the application of LLM and effective instruction for students with writing difficulties.

Journal of Computer Assisted LearningVol. 41(5)
The University of Sydney (AU), Wenzhou University (CN), National Taiwan University of Science and Technology (TW), Soochow University (TW), National Taichung University of Education (TW), Yuan Ze University (TW)
Openalex Percentile: Top 72%
Writing and Handwriting Education
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.