본문 바로가기
  • Home

Analyzing Sentiment and Syntactic Complexity in L2 Writing: A Computational Approach

  • Modern English Education
  • Abbr : MEESO
  • 2024, 25(), pp.143-154
  • Publisher : The Modern English Education Society
  • Research Area : Humanities > English Language and Literature > English Language Teaching
  • Received : May 26, 2024
  • Accepted : May 31, 2024
  • Published : June 20, 2024

Jung, YeonJoo 1

1부산대학교

Accredited

ABSTRACT

This study investigated the effect of sentiment in second language (L2) writing prompts on the syntactic complexity and emotionality in essays written by English as a Foreign Language (EFL) learners. The research analyzed a dataset of 1,004 essays written by Korean adolescent students to examine the relationship between the sentiment scores of writing prompts and learner essays. The study used computational tools to measure various dimensions of syntactic complexity. The findings indicated that the sentiment of writing prompts is a significant predictor of the emotionality scores of L2 students' essays. The positive correlation suggested that more positive prompts are more likely to elicit overall more positive essays. Additionally, the study identified significant differences in baseline sentiment values between writing prompts and learner essays. Even when responding to highly negative prompts, L2 students tended to use neutral sentiment in their essays. This study demonstrated how emotional prompts can impact cognitive load, resulting in simpler syntactic structures. It also highlighted the potential risks in educational assessment due to construct-irrelevant variance. Among the indices used to measure syntactic complexity, mean length of clauses (MLC) and coordinate phrases per clause (CP/C) showed significant effects influenced by the emotionality of writing prompts and learner essays. The implications of this study are discussed.

Citation status

* References for papers published after 2023 are currently being built.