The paper analyses poverty in South Africa, using the GHS-data of 2018. STATSSA designates three poverty lines, the upper-bound poverty line is employed in the methodology, which makes the analysis more detailed and unique. The methodology uses descriptive statistics, multivariate analysis and inferential approaches specifically the binary-logistic regression model. The results of the model indicate that socioeconomic factors such as income, and household size, on the household level and head of household characteristics such as gender, marital status, age and population group were significant determinants of poverty. Policies that target girl education would go a long way in addressing poverty.