After more than a decade of the global call for gender parity in educational access, gender gaps in
educational outcomes still exists, especially in sub-Saharan Africa. This incidence could be attributed
to an uneven allocation of household resources towards the schooling of boys and girls. This paper
sheds light on gender bias in a household’s educational expenditures using a ‘Hurdle or Two-Part
Model’ estimation strategy. Using household-level data from the GLSS6, the study explores two
potential sources of gender bias: (i) bias in the decision to enroll/keep boys and girls in school; and (ii)
bias in the decision of education expenditure conditional on enrolling both boys and girls in schools.
The study established a significant pro-male bias in both the enrollment decision and in the decision of
how much to spend conditional upon enrolling boys and girls in school among the cohort of individuals
in the post-secondary schooling age while only the former source of bias is present in the seniorsecondary
schooling age cohort. Strikingly, however, no gender bias is found in the positive expenditure
decision for boys and girls belonging to the basic education school-going age cohort, even though a
pro-male bias operates in the conditional educational expenditure decision for children in the primary
school-going age group. Contrary to earlier findings, we find a pro-female biasin a household’s decision
of how much to spend conditional upon enrolling boys and girls for the cohort of children in the junior
secondary schooling age. These findings have important implications for policy action.