Search

This paper provides an overview of evidence on education in South Africa provided by household survey data, with a particular focus on large national surveys such as the 1993 South African Living Standards Survey and the annual October Household Survey. These surveys indicate that racial gaps in schooling persist in South Africa, although they have declined steadily over time. There is essentially no gender gap in schooling, however, with almost identical schooling outcomes for men and women in all racial groups. The racial gap in schooling can be largely attributed to a high rate of grade repetition for Africans, with only small differences in enrollment rates across racial groups. School quality has an important effect on both grade attainment and adult economic outcomes and is clearly an important component of the racial gap in schooling. Survey data demonstrate a large effect of schooling on earnings beginning in late primary years, with a higher rate of return to schooling for Africans than for whites. (Contains 22 references.) (Author/SM)

This article investigates differential investment by men in children by using a biosocial perspective to examine their financial expenditures on their genetic children and stepchildren from both current and former relationships. It uses a sample of 635 children age 0-12 years drawn from the Albuquerque Men Study 1990-93, which involved interviews that examined marital, reproductive, and employment histories and expenditures on children. This article also examines selection efforts by restricting the sample to men with children in more than one category (e.g., men with both genetic children and stepchildren in current unions), explaining that selection effects are important. Results indicate that men with both genetic offspring and step-offspring in current relationships invest similarly in children, but the differences between offspring from current and previous unions are greater when comparing men with children in both categories. The results are consistent with the biosocial model advanced in the paper. (Contains 73 references.) (SM)

This study examined the relationship between family structure, expenditures on education, and children's educational outcomes for black South Africans, using the nationally representative 1995 October Household Survey. The analyses focused on 28,215 individuals, ages 10 to 24 years, who had not completed secondary schools. The findings indicated that although enrollment levels were high for most ages, schooling advancement rates were well under one grade per year, suggesting high rates of grade repetition. Controlling for background factors, family structure was highly correlated with educational outcomes. The strongest effects were seen for children living with neither of their genetic parents, who were less likely to be enrolled in school, had completed fewer grades, were older for their grade if enrolled, and had less money spent on their school fees and school-related transportation costs than children living with both genetic parents. Children who lived with single mothers were also disadvantaged for most measures. Family structure played a strong role in the probability that a child was enrolled in school; additional effects were evident, although diminished in strength, for outcomes affecting only enrolled students, age delays for grade, and financial expenditures on schooling. In addition, past academic progress influenced expenditures on school; children who were behind in school for their age (indicative of previous grade repetition) had less money spent on their schooling, above and beyond the effects of family structure on schooling expenditures. The findings suggest that family structure is an important contributor to educational inequality in South Africa, although there are important caveats regarding self-selection into different family types, as well as issues of school quality. (Contains 63 references.) (Author/KB)

College enrollment rates of blacks have historically trailed those of whites, although the actual size of the racial gap has fluctuated in recent decades. Using logistic regression and a decomposition analysis, this study investigated the extent to which this college-going gap is due to racial disparities in socioeconomic family backgrounds, academic performance, and expectations and values about education. Sizable differences are found between blacks and whites in socioeconomic family background and academic performance, as well as interactions between race and these variables. However, expectations and values about education are similar between the groups and contribute little to racial disparities in college attendance. (Contains 2 figures, 4 tables, and 43 references.) (SLD)

Large household surveys are used to analyze links between schooling inequality and earnings inequality in Brazil and South Africa, countries that have long had among the highest levels of income inequality in the world. The studies were the 1995 South African October Household Survey (32,000 households) and the 1995 Brazilian Pesquisa Nacional de Amostra de Domicilios (about 85,000 households). Although the countries have similar earnings inequality, South Africa has much lower inequality in schooling. The contribution of schooling to earnings inequality is very similar in the two countries, however, due to the convex relationship between schooling and earnings. If the countries traded schooling distributions or returns to schooling, there would be little effect on earnings inequality. Both countries demonstrate strong relationships between parents' schooling and children's schooling, a key component of the intergenerational transmission of inequality. It is significant, however, that the penalty for having poorly educated parents is much smaller in South Africa. The results suggest that even large improvements in schooling maybe associated with inertia in earnings inequality in developing countries. (Contains 6 tables, 8 figures, and 18 references.) (Author/SLD)

This paper tests three explanations for the high educational expectations of Asian American high school students living in the United States: (1) favorable socioeconomic and background characteristics; (2) demonstrated academic ability; and (3) cultural values conducive to education. The focus is on differences in the relevance of these explanations across Asian American ethnic groups. Data from the National Education Longitudinal Study of 1988 (NELS:88), for 1988 through 1992, were used with five multivariate regression models to explain differences in educational expectations between Asian American ethnic groups and Whites, both for the base year and for changes over time. Much diversity is found in the forces that shape educational expectations and effect changes in them across Asian American ethnic groups. The educational expectations of groups that are well assimilated into U.S. society are principally influenced by socioeconomic and demographic factors, while parental expectations explain much of children's high educational expectations for recent immigrant refugees. Appendixes contain charts of estimated coefficients of linear regressions models predicting educational expectations for the first and second follow-ups of the NELS:88. (Contains 5 tables and 56 references.) (SLD)

This document provides information about an undergraduate course that introduces college students to major social, economic, and political influences that have affected the demographic structure of the nation's population over the last four decades. The course permits students, working in small teams, to investigate the ways in which changes in race relations, family living arrangements, the status of women, and the nation's industrial structure affect particular birth cohorts, population subgroups, and geographical areas. In addition to a description of the course, the document also provides illustrative examinations, team assignments, class discussion questions, and additional exercises. Lists of data sets, which students use in computerized census tabulations, also are included. (DB)