Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • harvard-cite-them-right
  • 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
Prompt Engineerings påverkan på studenters lärande: en mixad metodstudie
University of Borås, Faculty of Librarianship, Information, Education and IT.
University of Borås, Faculty of Librarianship, Information, Education and IT.
2025 (Swedish)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
Prompt engineerings effect on student learning : a mix method study (English)
Abstract [sv]

Denna studie undersökte hur användandet av instruktioner påverkar studenters tillämpning av prompt engineering gentemot studenter utan instruktioner och ifall det har en effekt på deras utveckling och lärande. Utifrån detta syfte formulerades en frågeställning: Hur förbättrar instruktioner för prompting av Large Language Models studenters lärande. För att ta reda på studenters tillvägagångssätt användes en mixad metod i syfte att analysera åtta studenter. Dessa studenter delades in i två olika grupper, en kontrollgrupp utan instruktioner och en experimentgrupp med instruktioner. Intervjufrågorna utgick från en semistrukturerad form och skapades utifrån syftet. Studien använde sig av kvalitativa och kvantitativa ansatser i syfte att beräkna tiden och kvaliteten av respektive intervju. Efter intervjuerna utformades beskrivningskategorier baserat på svaren i den kvalitativa studien, samt att tidtagningen från det kvantitativa testet sammanställdes. Detta lade grunden till studiens resultat. Resultatet från undersökningen visar att studenters lärande ökar när de fick ta del av instruktioner, eftersom det ökade förståelsen för att göra LLM medveten om exempelvis sina tidigare erfarenheter och vilken kunskapsnivå de redan besatt. Hypotesen i arbetet var att experimentgruppen som fick ta del av instruktioner skulle genomföra testet snabbare. Detta visade sig inte stämma utan gruppen som fick ta del av instruktioner tog i snitt längre tid på sig att genomföra testet än kontrollgruppen som inte fick ta del av instruktioner. Detta kunde bero på att experimentgruppen med instruktioner arbetade mer iterativt och analyserade svaren som genererats mer än studenterna i kontrollgruppen. Dock behöver ytterligare forskning göras kring användandet av instruktioner om prompt engineering vid promptande av Large Language Models och hur det påverkar studenters lärande, kunskapsutveckling och undervisning. 

Abstract [en]

This study aimed to investigate how the usage of instructions affects the usage of prompt engineering with students that were given instructions against students without instructions. To answer this the following questions was created: How does instructions improve students learning capabilities with prompting of Large Language Models. To find out the students' approach a mixed method was used, with the purpose of analysing eight students. The students were divided into two groups, one control group without instructions and one experiment group with instruction. The interview question was based on a semi-structured form which was created based on the purpose. The study used a qualitative and quantitative approach in purpose of calculating time and quality of each interview. Based on the answers of the qualitative study were descriptive categories created in order to analyze the result. The quantitative test was analyzed by collecting the time it took the respondents to complete the test. The result of this study shows that students' learning capabilities improved when given instructions of the usage of prompt engineering. Since it improved their understanding of for example making the LLM’s aware of their previous experience and knowledge. The hypothesis was that the experiment group that were given instructions would complete the tasks quicker than the group without instructions. However this was not the case, and the group with instructions took longer to complete the different tasks. This could be because they were working more iteratively and analyzing the generated answers more carefully than the group without instructions. In conclusion more research should be done on how the usage of instructions affects students' learning, development of education with prompt engineering.

Place, publisher, year, edition, pages
2025.
Keywords [en]
rompt engineering, Large language models, ChatGPT, Prompting, Students, Learning
Keywords [sv]
Prompt engineering, Large language models, ChatGPT, Prompting, Studenter, inlärning
National Category
Computer and Information Sciences
Identifiers
URN: urn:nbn:se:hb:diva-34280OAI: oai:DiVA.org:hb-34280DiVA, id: diva2:2000283
Subject / course
Informatics
Available from: 2025-09-26 Created: 2025-09-23 Last updated: 2025-09-26Bibliographically approved

Open Access in DiVA

2025KANI23(12024 kB)44 downloads
File information
File name FULLTEXT01.pdfFile size 12024 kBChecksum SHA-512
fefc2acfb1e3d1b2a4fb5dc486438b9c17e9cdf440544ea0fbe4849da6abd4518b2e260d07cf650c384b85fef5a92d6f835213e8e9e6120d76e0783e548794d6
Type fulltextMimetype application/pdf

By organisation
Faculty of Librarianship, Information, Education and IT
Computer and Information 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: 2273 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • harvard-cite-them-right
  • 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