Cited by 4Discourse Analysis in Language StudiesArtificial Intelligence in EducationLinguistics and Discourse Analysis

The DOI matches its Crossref record; 2 of the 2 references it cites were found in a registry or an index.

Cohesion and Coherence in Essays Generated by ChatGPT: A Comparative Analysis to University Students’ Writing

Open access
Ismail, Hanaa Youssef Shaarawy·2023·CDELT Occasional Papers in the Development of English Education

Abstract

ChatGPT is a natural language processing tool that can assist in many language-related tasks (ChatGPT, personal communication, May 20, 2023). It can also produce human-like writing. Cohesion and coherence are major requirements of any piece of writing, and it is assumed that ChatGPT can generate successful essays regarding these language requisites. Adopting Halliday and Hasan’s (1976) theory of cohesion and van Dijk and Kintsch’s (1983) cognitive situation model to achieve coherence, the researcher investigates cohesion and coherence in essays produced by ChatGPT compared to essays produced by students with respect to the same language features. The current study revolves around two major research questions: how are cohesion and coherence manifested in essays produced by ChatGPT, and how do these same language features resemble or differ from essays written by students? To answer these questions, students’ essays from first and fourth years in the English Department at one of the Egyptian private universities and essays produced by ChatGPT on the same prompts were assessed by the two courses’ instructors, qualitatively analyzed by ChatGPT and quantitatively analyzed using Coh-Metrix software version 3.0, and then compared. Findings revealed that ChatGPT can produce human-like essays not only in terms of cohesion and coherence, but also regarding lexical choices and advanced grammatical structures, which poses a great challenge of how to avoid having students plagiarizing from ChatGPT. Accordingly, the article concludes with some recommendations that utilize ChatGPT to enhance learners’ linguistic proficiency without sacrificing the integrity, honesty and reliability of the learning process.

Overview

Integrity signals

Registry

Crossref · 10.21608/opde.2023.325331

Cites

2 of 2 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Verification

Claims the paper makes

We couldn’t retrieve this paper’s full text, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

Cited by

Sorted by citations

See all 4

Metadata

Authors
Ismail, H. Y. S.
Venue
CDELT Occasional Papers in the Development of English Education
Published
2023-07-01
Volume
83
Issue
1
Pages
143-165
ISSN
  • 1110-2721 (ISSN-L)
  • 2735-3591
OpenAlex
W4388613120
Cited by
4
References
2
Citations by year
2025: 3 · 2024: 1
FWCI
10.95
Type
journal article
Language
en
Open access
Locations
CDELT Occasional Papers in the Development of English Education (publishedVersion)
Topics
  • Social Sciences › Linguistics and Language › Discourse Analysis in Language Studies
  • Social Sciences › Education › Artificial Intelligence in Education
  • Social Sciences › Linguistics and Language › Linguistics and Discourse Analysis
Keywords
ChatGPT, cohesion, coherence, AI-generated text, student writing, Halliday and Hasan cohesion theory, Coh-Metrix, word choice, grammatical complexity, academic integrity, plagiarism, EFL writing

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review