HIV among women: does education matter more than we previously thought?

Abstract

Women remain disproportionately affected by HIV in sub-Saharan Africa. Although there is unanimous agreement on the positive impact of schooling in reducing the pandemic, measuring the extent of this impact remains empirically difficult. Using data from the 2018 round of Zambia’s Demographic and Health Survey (DHS), we took advantage of the free primary education reform that abolished school fees for grades one to six in 2002 to obtain an exogenous variation in women’s education levels. We estimate a three-equation model to assess consistent estimates of the impact of education on a woman’s probability to be HIV positive. When the problems of sample selection and endogeneity are not addressed, we find that the effect of education on HIV status is greatly underestimated. After controlling for these two sources of bias, the effect having a secondary education on the risk of being seropositive doubles when compared to the uncorrected results. This result suggests that women acquire agency through education to prevent HIV infection. Pathways to these effects include contraceptive use, the number of lifetime sexual partners and marital status. Hence, policy makers and practitioners in Zambia should invest substantial efforts in promoting girls’ education in order to reduce the prevalence of HIV/AIDS among women.

Keywords:

• Education
• HIV testing
• HIV status
• women
• endogeneity
• sample selection bias

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Demographic and Health surveys data are publicly available online at https://www.dhsprogram.com/data/