Econometric Analysis of Gender Disparities in STEM Enrollment and Performance in Sub-Saharan African Secondary Schools

Gender disparities in Science, Technology, Engineering, and Mathematics (STEM) remain among the most entrenched structural inequalities in Sub-Saharan African (SSA) secondary education, with far-reaching consequences for human capital development and inclusive economic growth. Despite incremental progress in achieving gender parity at the primary level, significant participation and performance gaps re-emerge and intensify within STEM tracks at the secondary level, constraining the region's long-term development trajectory. This study presents a rigorous, multi-method econometric analysis of gender gaps in STEM enrollment and examination performance across five SSA countries—Kenya, Tanzania, Uganda, Ghana, and Nigeria—using a pooled dataset of 1,423,807 upper secondary students spanning 2018 to 2023. The analytical framework integrates binary logistic regression and probit models to estimate the probability of STEM enrollment as a function of gender and a comprehensive set of socioeconomic and institutional covariates, alongside the Oaxaca-Blinder decomposition technique to partition the gender performance gap in national examinations into explained (endowment) and unexplained (structural) components. Two-way panel fixed-effects models with country and year controls provide additional robustness. Descriptive results confirm that female students are 23.4 percentage points less likely to enroll in STEM tracks than their male counterparts. Logistic regression analysis yields a fully controlled log-odds estimate for the female indicator of −0.812 (Average Marginal Effect = −18.4 percentage points; p < 0.001). Oaxaca-Blinder decomposition reveals that 38.6% of the observed performance gap of 0.41 standard deviations is attributable to measurable endowment differences, while the remaining 61.4% is structurally unexplained, implicating stereotype threat, implicit teacher biases, gendered curricula, and cultural norms. The share of female STEM teachers emerges as a significant moderator of female performance. The study concludes with evidence-based policy recommendations directed at teacher recruitment and gender-responsive training, curriculum reform, conditional cash transfer programmes targeting STEM-enrolled girls, and equitable STEM infrastructure investment across rural and urban schools.