Computer Science > Computers and Society
[Submitted on 14 Aug 2024]
Title:New Curriculum, New Chance -- Retrieval Augmented Generation for Lesson Planning in Ugandan Secondary Schools. Prototype Quality Evaluation
View PDFAbstract:Introduction: Poor educational quality in Secondary Schools is still regarded as one of the major struggles in 21st century Uganda - especially in rural areas. Research identifies several problems, including low quality or absent teacher lesson planning. As the government pushes towards the implementation of a new curriculum, exiting lesson plans become obsolete and the problem is worsened. Using a Retrieval Augmented Generation approach, we developed a prototype that generates customized lesson plans based on the government-accredited textbooks. This helps teachers create lesson plans more efficiently and with better quality, ensuring they are fully aligned the new curriculum and the competence-based learning approach.
Methods: The prototype was created using Cohere LLM and Sentence Embeddings, and LangChain Framework - and thereafter made available on a public website. Vector stores were trained for three new curriculum textbooks (ICT, Mathematics, History), all at Secondary 1 Level. Twenty-four lessons plans were generated following a pseudo-random generation protocol, based on the suggested periods in the textbooks. The lesson plans were analyzed regarding their technical quality by three independent raters following the Lesson Plan Analysis Protocol (LPAP) by Ndihokubwayo et al. (2022) that is specifically designed for East Africa and competence-based curriculums.
Results: Evaluation of 24 lesson plans using the LPAP resulted in an average quality of between 75 and 80%, corresponding to "very good lesson plan". None of the lesson plans scored below 65%, although one lesson plan could be argued to have been missing the topic. In conclusion, the quality of the generated lesson plans is at least comparable, if not better, than those created by humans, as demonstrated in a study in Rwanda, whereby no lesson plan even reached the benchmark of 50%.
Current browse context:
cs.CY
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.