Students rated ChatGPT-generated narrative endings equal to or higher than student-written endings on setting, themes, clarity of language and tone¶
Claim · All claims
Evidence · 1 study · causalr1·q2
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? In a comparative survey, undergraduate students rated ChatGPT-generated texts higher than student-written texts on setting, themes and topics, clarity of language, and tone, and equal on characterization. → Andrea Roxana Bellot 2025
Evidence¶
Andrea Roxana Bellot 2025¶
Andrea Roxana Bellot, Mar Gutiérrez-Colón Plana, Katarzyna Agnieszka Baran. (2025). Redefining Literature Education: The Role of ChatGPT in Undergraduate Courses. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-025-00497-3
q2 · i? · causal · r1
Survey comparison of the experimental group (ChatGPT texts) against the control group (student texts) across Questions 3–10 using Likert-scale means analyzed in JASP. Setting means 4.381 vs 3.773; "ChatGPT excelled in maintaining a consistent and appropriate tone (mean of 4), which was significantly higher than the control group's mean (3.273)". Only descriptive means are reported.
"The setting was significantly better appreciated in the texts generated by ChatGPT (mean of 4.381) as opposed to those written by students (mean of 3.773), showcasing a stronger ability of ChatGPT to vividly convey place and time."
Discussion¶
Related Claims¶
- Students analyzing ChatGPT-generated endings were somewhat more likely than peer-text readers to believe the ending could be Orwell's original — related
- Students perceive AI scoring as accurate for grammar, mechanics, and organization but skeptical of its ability to evaluate diverse styles, complex writing, tone, and emotion — related
- Students trusted the plot plausibility of peer-written alternative endings more than ChatGPT-generated endings — reports the opposite
- Both student- and AI-text groups similarly affirmed contextual and character integration, with greater uncertainty in the ChatGPT-text group — related
- Students were far more likely to accept peer-written alternative endings as valid conclusions to the novel than ChatGPT-generated endings — reports the opposite
Page metadata
| Field | Value |
|---|---|
status |
review |
generated |
by process:wiki-ingest on 2026-10-27 |
trust |
unverified — no one has confirmed this page's content |
Frontmatter is the page's machine-readable record — see the Schema & Guide.