Bridging the gap: an analysis of determinants contributing to the gender digital divide in Southern African countries, using the AfroBarometer data (2018)

Gebers, Isabella

Bridging the gap: an analysis of determinants contributing to the gender digital divide in Southern African countries, using the AfroBarometer data (2018)

Thesis / Dissertation

2025

Publisher

University of Cape Town

Department

School of Economics

Faculty

Faculty of Commerce

Abstract

In this paper, I examine the factors that contribute to the gender digital divide across eleven Southern African countries. The ongoing digitalization of the world presents this divide as a critical challenge that prevents equality in Least Developed Countries. Despite efforts and policies that attempt to address the barriers to female's access and usage of digital resources, a digital gender gap persists. This is largely the case in Southern Africa, where females are shown to be less likely than men to use technologies. Using the 2018 AfroBarometer data for eleven Southern African countries (Botswana, Lesotho, Madagascar, Malawi, Mauritius, Mozambique, Namibia, South Africa, Zambia, Zimbabwe, and Eswatini), my investigation includes an analysis through descriptive statistics and through empirical methods. My preliminary findings through the descriptive analysis illustrate the diverse range of the divide across the countries. In all eleven countries the difference between males and females in internet usage is evident. Further findings demonstrate the impact of the rural/urban divide on internet usage and how educational levels affect access to computers. My empirical approach was layered: first, a multivariate logistic regression looked at the determinants of internet usage, followed by a nonlinear decomposition technique that broke down these determinants into different effects. The findings from these approaches were consistent with the descriptive statistics. Variables such as education and employment were noted as significant contributing factors. Additionally, these results also identified household size and cultural implications behind the digital divide. The results were in favour of males having greater access to and usage of internet and mobile technology. To assess the robustness of my results, I have also included a sensitivity analysis of the decomposition results through models that incorporate different assumptions, data variations and an Instrumental Variable technique to check for non-randomness in the sample. These results support the consistency of the baseline models and enhance the robustness of the original decomposition approach used. However, I do acknowledge that the paper does have limitations due to the nature of the data. My suggestion is to investigate the digital divide further through other datasets that are more recent. Given the results, my recommendations for policy would be to focus on increasing female employment and addressing the skills gap. To bridge the gender digital divide, these should focus on enhancing females' economic empowerment, particularly through education and capability-enhancing opportunities. This is essential to embed equal, inclusive prospects for females.

Keywords

Digital divide, gender

Reference:

Collections

Masters

Full item page