kth.sePublications KTH
Operational message
There are currently operational disruptions. Troubleshooting is in progress.
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
Design and Evaluation of an LLM Assistant: Supporting teachers in formative assessment for programming education
KTH, School of Industrial Engineering and Management (ITM), Learning.
2025 (English)Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Large Language Models or LLMs have been employed in many different fields, including education. These AI tools are used, among other things, for answering students’ questions and generating teaching material. Assisting teachers in assessment is an interesting topic that does not get as much attention in research. In this master’s thesis, which was done as a part of the K-ULF collaboration at KTH, we explore using LLMs to assist teachers in formative assessment. One goal of this thesis is to determine ways that LLMs can facilitate formative assessment for secondary school programming teachers in Sweden. Another goal is to examine teachers’ views on the efficiency and usefulness of formative assessment of students provided by AI. In the end of this thesis, we conclude that the teachers’ views on using LLMs for assisting their work in formative assessment are mostly positive. The technology can facilitate their work by giving them a readable and well-summarized overview of the students work process, and possibly pointing out things they might have missed themselves. Different opinions from the teachers were noted on different models.

Abstract [sv]

Stora språkmodeller, även kallade LLMer, har nyttjats inom många olika områden, inklusive utbildning. Dessa AI-verktyg används bland annat för att besvara elevers frågor och generera undervisningsmaterial. Att hjälpa lärare i bedömningar är ett intressant ämne som inte får lika mycket uppmärksamhet i forskningen. I denna masteruppsats, som skrevs i samarbete med K-ULF-gruppen på KTH, utforskar vi hur LLMer kan stödja lärare i formativ bedömning. Ett av målen med denna avhandling är att fastställa hur LLMer kan underlätta formativ bedömning för lärare i programmering på gymnasienivå i Sverige. Ett annat mål är att undersöka lärarnas syn på effektiviteten och nyttan av formativ bedömning som tillhandahålls av AI. I denna uppsats drar vi slutsatsen att lärarnas syn på att använda LLMer för att stödja deras arbete med formativ bedömning mestadels är positivt. Teknologin kan underlätta deras arbete genom att ge dem en läsbar och väl sammanfattad översikt över elevernas arbetsprocess, och eventuellt påpeka saker de själva kan ha missat. Lärarna hade olika åsikter om de olika modellerna.

Place, publisher, year, edition, pages
2025. , p. 50
Series
TRITA-ITM-EX ; 2025:500
Keywords [en]
Pedagogy, didactics, AI, formative assessment, LLM, teachers, tool, sociocultural theory
Keywords [sv]
Pedagogik, didaktik, AI, formativ bedömning, LLM, lärare, verktyg, sociokulturell lärandeteori
National Category
Engineering and Technology Educational Sciences
Identifiers
URN: urn:nbn:se:kth:diva-371821OAI: oai:DiVA.org:kth-371821DiVA, id: diva2:2007787
Subject / course
Technology and Learning
Educational program
Master of Science in Engineering - Engineering and of Education
Presentation
2025-09-09, 00:00
Supervisors
Examiners
Available from: 2025-10-21 Created: 2025-10-21 Last updated: 2025-10-21Bibliographically approved

Open Access in DiVA

fulltext(1725 kB)110 downloads
File information
File name FULLTEXT01.pdfFile size 1725 kBChecksum SHA-512
85c1deb693c2f193995453050245ae296f1c1f51ce4499062183c60b081cac6220fb53c63cec0819a0e8a2ec26b548207ef64fd9429a4c297cd3e412ceb3e508
Type fulltextMimetype application/pdf

By organisation
Learning
Engineering and TechnologyEducational Sciences

Search outside of DiVA

GoogleGoogle Scholar
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: 503 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