Abstract
The objective of this work is to analyze the determinants of inequalities in the school performance of 56 392 pupils at the end of primary school in 2 603 schools located in 647 districts in 12 countries of Eastern and Southern Africa (Botswana, Kenya, Lesotho, Malawi, Mozambique, Namibia, Swa-ziland, Tanzania, Uganda, Zambia and Zimbabwe) based on the third 2007 Southern and Eastern Africa Consortium for Monitoring Educational Quality (SACMEQ III) survey. The work is struc-tured around three chapters. The first chapter studies the influence of individual characteristics of the student and the school on school performance, as well as the important role of regional characteristics. In the second chapter, we seek to analyze how the inspection of schools and their accessibility by students a˙ect the eÿciency of schools. The last chapter proposes to study the socio-economic factors and schooling conditions a˙ecting performance and inequalities in access between girls and boys, taking into account their socio-economic backgrounds. To do this, we used di˙erent econometric approaches, namely a multilevel model in the first chapter, a non-parametric frontier model in chapter two, and a spatial econometric model in chapter three. The results show that the achievements of pupils, the eÿciency of institutions and inequality in access to school between girls and boys are highly heterogeneous in Southern and Eastern Africa. The results also indicate that the students who registered relatively high scores are located in rich urban areas with access to the means of transportation. Moroever, deep inequalities in schooling exist due to deficiency of transportation, road infrastructure, educational and health conditions particularly in districts located in rural areas and in poor neighborhoods in large cities. We also show that the variables a˙ecting school inspection missions, the use of the teaching langage at home and security of the school environment play an
Collections
Unless otherwise noted, the license for the item is described as Attribution-NonCommercial-NoDerivates.