Abstract:

The rapid digitalisation of Nigerian higher education has increased the use of Artificial Intelligence (AI) tools and digital assessments, making it essential to understand the factors that shape students’ performance in technology-driven assessments. Although previous studies have examined variables such as computer anxiety, information literacy, and AI usage independently, few have integrated these constructs into a unified digital profile or explored their combined predictive effects on digital assessment performance. These gaps limit the comprehensive understanding of digital readiness within AI-supported assessment environments in Nigeria. This study explores the impact of artificial intelligence (AI) interaction and students’ digital profiles on academic performance in digital assessments among Nigerian Undergraduates. The quantitative study adopts a cross-sectional survey design. The study population comprised undergraduate students in Southwestern Nigeria. The multi-stage random sampling was employed to select five universities in Southwestern Nigeria. The stratified sampling technique was employed to select two different universities within the Southwestern region of Nigeria, with school type and location serving as strata. The study selects 432 undergraduate students using simple random sampling techniques. The data was collected using validated questionnaires and official digital assessment results. Pearson correlation and multiple regression analyses examined individual and joint predictive effects. Results revealed that AI interaction significantly improved digital assessment performance (r = 0.42, p < 0.001), representing a medium effect size (Cohen’s d = 0.91). Information literacy emerged as the strongest positive predictor (β = 0.36, p < 0.001; f² = 0.26), while computer anxiety negatively affected performance (β = –0.28, p < .01; f² = 0.18). Demographic characteristics contributed modestly (β = 0.14, p < 0.05). The combined model was significant (F(4, 385) = 26.47, p < 0.001) and explained 41% of the variance in digital assessment performance, indicating a large overall effect. The study recommends expanding AI-enhanced learning environments, strengthening digital literacy training, and implementing targeted interventions for students with high computer anxiety. In conclusion, AI interaction and students’ digital profiles significantly shape academic performance in digital assessments, highlighting the need for equitable, data-driven digital learning strategies in Nigerian universities.

Keywords:

Artificial intelligence, digital profiles, academic performance, digital assessments, Nigerian undergraduates.