Research 

Given the low levels of educational standards in the developing world, the World Education Forum adopted the Dakar Framework for Action (DFA) in 2000, calling for quality 'Education for All' children of school-going age. Heeding this call, many sub-Saharan African countries instituted Free Primary Education (FPE) policies. Lesotho instituted the FPE programme in 2000 on a grade-by-grade basis; first abolishing school fees in grade one, and then in successive higher grades each following year. This thesis consists of a short introductory chapter, three self-contained analytical chapters which empirically evaluate the importance of the FPE policy and family factors on education in Lesotho, and the summary chapter. It first examines the effect of the FPE policy on primary school enrolment in Chapter 2 using household-level data for before and after the policy. A difference-in-differences strategy is employed to tease out the FPE effect. This exploits the variations in enrolment rates over time and across grade-groups (i.e. grades covered versus those not-yet covered) created by the implementation strategy of the programme. The findings demonstrate that the policy significantly increased enrolment of primary school-age children by at least 9.3 percentage points (or 13.2 percent). There is also evidence that this policy disproportionately raised enrolment levels of children from poor households and that of boys (the historically disadvantaged group), thereby bridging the gender- and wealth-related educational (enrolment) inequalities. In Chapter 3, the thesis draws on grade six pupils' standardised maths and (English) reading test scores from 2000 and 2007 to analyse changes in educational achievement and educational inequality, and the determinants of such changes. The analysis of the data shows that educational achievement increased significantly for both low- and high-ability pupils over the period of analysis. Nonetheless, this increase in achievement was accompanied by a significant rise in educational inequality, especially in reading test scores. The analysis further shows that these changes are statistically related to policy measures taken under the FPE programme. In particular, the results show that pupil-teacher ratio is negatively correlated with changes in reading performance of low-ability pupils, while teacher effort (i.e. subject-testing frequency and teaching hours per week) and grade repetition have a positive influence on changes in educational achievement. These results suggest that the fall in pupil-teacher ratio between 2000 and 2007 has helped increase educational achievement. The analysis, however, reveals that much of the increase in educational achievement and educational inequality is unexplained by both school and pupils' family characteristics, which suggests that there could be other unobserved family and school factors that influence achievement and inequality. Therefore, in Chapter 4 of the thesis, I shift focus from FPE policy effects and look at the impact of family factors on human capital accumulation. Specifically, Chapter 4 examines the effect of a child's order of birth on human capital development (i.e. enrolment, educational attainment, and schooling progression) using family-fixed effects models. Birth order has received much attention in the economics and psychology literature. Contrary to much of the evidence from developing countries, I find that birth order has a strong negative effect on human capital development. These birth order effects are pronounced in large families, and families with first-born girls, thereby revealing the strong girls' education preferences in Lesotho. Turning to potential pathways of these effects, I find that birth order effects are not propagated through family wealth, but mainly through birth- (or child-) spacing. These results suggest that there are some intra-household inefficiencies that could explain the changes in educational achievement and inequality. 

Thesis (PhD)--University of Cape Town