Education Society
Students' Different Ways of Using AI Assistants Are Reflected in Essay Writing
A writing assistant based on generative AI is not just one tool for students, but it is used in very different ways – and these ways may be connected to the quality of the essay. This is suggested by a recent study stored in the arXiv service, which developed an AI assistant aimed at college students for writing argumentative essays.
The study specifically examined the dynamics of interaction: what students request from the AI at different stages of the writing process and how usage progresses during a two-hour writing session. The data consisted of 1,282 log files from 32 undergraduate students, revealing every interaction step between the student and the assistant.
Two data analysis methods were used in the analysis. The so-called sequential pattern mining searched for recurring action sequences, and k-means clustering grouped similar usage patterns together. This resulted in two different student groups based on AI usage.
The first group utilized the AI assistant particularly for planning the essay's framework and refining its structure. The second group's activity differed from this, and students used the assistant in clearly different ways, although a more detailed description of the emphases remains open in the summary. The study also examined how these different usage patterns related to the quality of the essays.
The study highlights that a pedagogically designed AI assistant is not merely a text generator, but its role in writing depends on how the student interacts with it. By understanding different usage methods, AI tools can be designed in the future to better support both the writing process and the development of writing skills.
Source: Examining Student Interactions with a Pedagogical AI-Assistant for Essay Writing and their Impact on Students Writing Quality, ArXiv (AI).
The study specifically examined the dynamics of interaction: what students request from the AI at different stages of the writing process and how usage progresses during a two-hour writing session. The data consisted of 1,282 log files from 32 undergraduate students, revealing every interaction step between the student and the assistant.
Two data analysis methods were used in the analysis. The so-called sequential pattern mining searched for recurring action sequences, and k-means clustering grouped similar usage patterns together. This resulted in two different student groups based on AI usage.
The first group utilized the AI assistant particularly for planning the essay's framework and refining its structure. The second group's activity differed from this, and students used the assistant in clearly different ways, although a more detailed description of the emphases remains open in the summary. The study also examined how these different usage patterns related to the quality of the essays.
The study highlights that a pedagogically designed AI assistant is not merely a text generator, but its role in writing depends on how the student interacts with it. By understanding different usage methods, AI tools can be designed in the future to better support both the writing process and the development of writing skills.
Source: Examining Student Interactions with a Pedagogical AI-Assistant for Essay Writing and their Impact on Students Writing Quality, ArXiv (AI).
This text was generated with AI assistance and may contain errors. Please verify details from the original source.
Original research: Examining Student Interactions with a Pedagogical AI-Assistant for Essay Writing and their Impact on Students Writing Quality
Publisher: ArXiv (AI)
Authors: Wicaksono Febriantoro, Qi Zhou, Wannapon Suraworachet, Sahan Bulathwela, Andrea Gauthier, Eva Millan, Mutlu Cukurova
December 27, 2025
Read original →