Identifying Students’ Writing Styles by Using Computational Linguistic Approach

Abstract

The purpose of this study was to identify stable writing styles of students for implementing the cognitive learning system of writing education. According to the previous research, writing styles of writers could be reflected by the linguistic features of writing such as ambiguity, cohesion, lexical diversity, and syntactic complexity. This study explored the method of capturing students’ writing styles by investigating the similarity of linguistic features between drafts and comments written by the same students in a peer review system. The computational linguistic tool, Coh-Merix, was used for analyzing the linguistic features of drafts and comments produced by 41 undergraduate students. The results of this study showed that there was similarity between drafts and comments in the measures of ambiguity, lexical diversity, and syntactic complexity, whereas there was difference in the measures of cohesion.