kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
ChatGPT-4’s Effectiveness in Providing Feedback on Argumentative Writing in Higher Education: A Case Study
KTH, School of Electrical Engineering and Computer Science (EECS).
KTH, School of Electrical Engineering and Computer Science (EECS).
2024 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
Abstract [sv]

Att ge och få effektiv återkoppling på argumenterande uppsatser är avgörande för studenters utveckling men är ofta utmanande med ökande klasstorlekar och lärarnas ökande arbetsbörda. Denna studie undersöker hur väl ChatGPT-4 kan ge återkoppling på argumenterande uppsatser inom högre utbildning och jämför dess effektivitet med återkoppling från andra studenter. Med en fallstudie-experimentell metod utvärderade sju universitetsstudenter de valda akademiska texterna. Deras återkoppling jämfördes noggrant med ChatGPT-4:s återkoppling, med hjälp av etablerade kriterier som inkluderade beskrivning, identifiering, motivering och konstruktiv återkoppling. Resultaten visar att ChatGPT-4 vanligtvis ger mer detaljerad och konsekvent återkoppling än studenterna, vilket indikerar att det kan vara ett användbart alternativ eller ett komplement för återkoppling i utbildningssammanhang. Studien visar också hur generativa AI-verktyg som ChatGPT-4 potentiellt kan införas i utbildningssystem för att förbättra studenternas lärande och göra återkopplingsprocessen mer effektiv. Genom att jämföra mänsklig och AI-genererad återkoppling bidrar studien till diskussionen om generativa AI:s roll i utbildning och belyser både dess styrkor och begränsningar.

Abstract [en]

Providing and receiving effective feedback on argumentative essays is crucial for student development but is often challenging with increasing class sizes and teacher workloads. This paper examines how well ChatGPT-4 can give feedback on argumentative essays in higher education and compares its effectiveness to feedback from other students. Using a case study experimental method, seven university students evaluated the selected academic texts. Their feedback was carefully compared to that of ChatGPT-4, using the established criteria which included description, identification, justification, and constructive feedback. The findings show that ChatGPT-4 usually provides more detailed and consistent feedback than peers, indicating that it could be a useful alternative or additional tool for feedback in educational settings. The study also shows how generative AI tools like ChatGPT-4 can potentially be incorporated into educational systems to improve student learning and make the feedback process more efficient. By comparing human and AI-generated feedback, the study adds to the discussion on generative AI's role in education, pointing out both its strengths and limitations.

Place, publisher, year, edition, pages
2024. , p. 8
Series
TRITA-EECS-EX ; 2024:236
Keywords [en]
Peer feedback, ChatGPT, academic writing
National Category
Computer and Information Sciences
Identifiers
URN: urn:nbn:se:kth:diva-349956OAI: oai:DiVA.org:kth-349956DiVA, id: diva2:1881812
Supervisors
Examiners
Available from: 2024-08-08 Created: 2024-07-03 Last updated: 2024-08-08Bibliographically approved

Open Access in DiVA

fulltext(1166 kB)453 downloads
File information
File name FULLTEXT01.pdfFile size 1166 kBChecksum SHA-512
3e67704543b9a0c39c9df6a77240212193cfd252bf03b9b01f7a3d9e7d877ad01bcfa02de7fadf957ca7d5caed603f6dc37e9f6b0b95ce75652d0f8df59d34f2
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Computer and Information Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 454 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 440 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf