Exploring the potential of artificial intelligence to enhance writing skills
Academic paper writing is crucial in the education of medical students, especially for those whose first language is not English. A recent study investigated the effectiveness of using large language models, specifically ChatGPT, to improve the English academic writing skills of non-native English speakers.
Study Overview
A group of 25 third-year medical students from China participated in the study, which consisted of two stages. Initially, the students were tasked with writing a mini paper. Subsequently, they revised the paper using ChatGPT within a two-week period. The evaluation of the mini papers focused on structure, logic, and language, with assessments conducted manually and through AI scoring using ChatGPT-3.5 and ChatGPT-4 models. Feedback on the experience of using ChatGPT was collected through a questionnaire.
Results and Findings
After utilizing ChatGPT for writing assistance, there was a significant improvement in both manual and AI scoring. The study indicated a notable increase in scores, highlighting the potential of large language models in supporting academic writing. Statistical analysis showed no significant difference between manual scoring and ChatGPT-4 scoring, suggesting the utility of ChatGPT-4 in assisting teachers with grading.
Feedback from the questionnaire revealed positive responses from students, with improvements in writing quality and language skills noted.
Role of Large Language Models in Education
Large language models (LLMs) like ChatGPT are AI tools that excel at generating and understanding text. Trained on vast amounts of textual data, LLMs have demonstrated their capabilities in various tasks such as question answering, machine translation, and writing. ChatGPT, released by Open AI in 2022, has found applications in the medical field, offering personalized learning experiences and enhancing problem-solving abilities.
Research has shown that LLMs can improve students' proficiency in writing, organization, coherence, grammar, and vocabulary. AI-mediated instruction positively impacts English learning achievement and self-regulated learning, making them valuable tools in educational settings.
Overall, the study underscores the efficacy of large language models in enhancing the English academic writing skills of non-native speakers in medical education. It sheds light on the potential of AI tools to aid in educational evaluation processes, particularly in non-English speaking environments.
