Design and Evaluation of an Adaptive AI-Powered Learning Companion for Enhancing Digital Literacy in Nigerian Secondary Schools

Abstract

The rapid digitization of education in the 21st century has necessitated the integration of Artificial Intelligence (AI) into learning environments to foster inclusivity and adaptability. Despite significant progress globally, Nigerian secondary schools remain constrained by a lack of adaptive learning systems capable of personalizing instruction in line with students’ diverse cognitive and socio-economic contexts. This study presents the design and evaluation of an AIpowered Adaptive Learning Companion (ALC) aimed at enhancing digital literacy among secondary school students in Nigeria. The system integrates Natural Language Processing (NLP) and Reinforcement Learning (RL) algorithms to tailor instructional content dynamically based on learners’ performance patterns and interaction behaviors. A quasi-experimental design was employed, involving 300 students across three Nigerian states. Test Instruments Reported: Cronbach’s Alpha: 0.882 (strong reliability), One-Way ANOVA: F(1,298) = 47.813, p < 0.001 (statistically significant); Partial Eta-Squared (η²): 0.14 (large effect size) and TwoWay ANOVA: Institution × Group interaction insignificant for ENG (p = 0.06), significant for Overall_Mean (p = 0.021).The results demonstrated a significant improvement in learners’ digital literacy scores, engagement rates, and learning retention compared to traditional instruction. The study concludes that AI-driven adaptive systems represent a transformative approach to equitable and scalable education delivery in developing contexts.