Enhancing Equitable Access to Secondary Education in Malawi Using Generative AI : A Pilot Study on the Agriculture Subject
Stanley Ndebvu et al.
Abstract
Statistics reveal that inadequate teaching and learning resources are a significant challenge in Africa's education sector, with various educational reports highlighting shortages of qualified teachers, textbooks, and high student–teacher ratios. These challenges hinder progress toward achieving Sustainable Development Goal 4 and the African Agenda 2036, which advocate for inclusive, equitable, and quality education for all. To address these issues, we developed a Llama‐powered chatbot aligned with the Malawian School Certificate of Education (MSCE) curriculum. The system allows students to ask subject‐specific questions and receive personalized, interactive feedback, including auto‐generated notes, multiple‐choice assessments, and performance tracking. We fine‐tuned a LLaMA 2 3B on a curated corpus of textbooks, pamphlets, and national exam questions from the MSCE syllabus. We used this application to conduct a study investigating how artificial intelligence (AI) can improve access to quality education for marginalized communities in Malawi. The study involved a purposefully sampled 200 secondary school students and 20 teachers, with 93% of the participants indicating that they would continue to use the tool and would recommend it to their peers. 82% indicated that they are satisfied with the chatbot as a learning tool. The results suggest that AI solutions in education have the potential to significantly enhance access to personalized learning resources and curb geographical and time constraints to teaching and learning.
Evidence weight
Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40
| F · citation impact | 0.50 × 0.4 = 0.20 |
| M · momentum | 0.50 × 0.15 = 0.07 |
| V · venue signal | 0.50 × 0.05 = 0.03 |
| R · text relevance † | 0.50 × 0.4 = 0.20 |
† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.