2026 ASEE Annual Conference & Exposition

Background. International partnerships increasingly aim to build artificial intelligence (AI) capacity in engineering programs across developing regions, yet evidence-based guidance for designing such partnerships remains scarce. The 2025 United States Agency for International Development (USAID) funding freeze disrupted partnerships across African higher education, exposing the fragility of collaborations selected on reputation or governance type rather than on empirical assessment of institutional readiness. Engineering educators seeking international collaboration need a way to identify institutions positioned for productive, sustainable partnership.
Purpose. This study identifies institutional and contextual factors that predict AI integration and partnership readiness in Nigerian universities, with particular attention to engineering and computing programs as primary sites of AI adoption. The goal is to give international engineering education partners evidence-based criteria for partner selection grounded in institutional profiles rather than governance assumptions.
Design and Method. Comparative content analysis of 45 Nigerian universities stratified by type (15 Federal, 15 State, 15 Private) was conducted from May to October 2025. AI integration was assessed across six dimensions (AI Infrastructure, Curriculum Integration, Research Initiatives, Industry Partnerships, International Collaborations, and Policy Frameworks) using a 10-point coding framework with dimension-specific anchored descriptors. Inter-rater reliability yielded Cohen's kappa of 0.79 to 0.85 across dimensions. Multiple regression identified predictive factors and Pearson correlation with Bonferroni correction examined relationships among dimensions.
Results. The regression model explained 30% of variance in AI integration scores, R²=0.30, F(4,40)=4.29, p=0.006. Institution age was the strongest predictor, β=0.43, p=0.016, followed by South-West regional location, β=0.31, p=0.029. Institution type, tested via joint F-test on dummy variables, was not a significant predictor when other factors were controlled, F(2,40)=0.30, p=0.744. A strong correlation between international collaborations and industry partnerships, r=0.74, p<0.001, indicated mutually reinforcing external networks. Additional analysis showed institution age correlated moderately with Research Initiatives, r=0.40, p=0.006, and Research Initiatives correlated strongly with overall AI integration, r=0.89, p<0.001, supporting a resource-based mechanism in which age operates through accumulated faculty research capacity.
Conclusions. These findings challenge assumptions that governance structure alone determines partnership readiness. International engineering education partners should assess specific institutional profiles rather than selecting partners based on institution type. Three evidence-informed strategies are proposed: tiered consortium models, sector-specific partnerships aligned with engineering workforce needs, and hybrid digital-physical approaches that address infrastructure constraints.