Showing papers by "Duncan Thomas published in 1991"

How does mother's education affect child height ?

Abstract: Many studies have demonstrated that parental education has a significant positive impact on child health This paper attempts to identify the mechanisms through which maternal education affects one indicator of child health-height conditional on age and sex Using data from the 1986 Brazilian Demographic and Health Survey, it is shown that almost all the impact of maternal education can be explained by indicators of access to information-reading papers, watching television, and listening to the radio In urban areas, whether the mother is semi-literate accounts for some of the education effect There are also significant interactions between maternal education and the availability of community services indicating that education and health services are substitutes but education and the availability of sewerage services are complements Very little of the maternal education effect is transmitted through income

Efficiency of cohort sampling designs: some surprising results.

Abstract: Cohort sampling designs are proposed which one would intuitively expect to be more efficient than nested case-control sampling. Two of these designs start with a nested case-control sample and distribute controls to sampled risk sets other than those for which they were picked. The third design has the goal of maximizing the number of distinct persons in a nested case-control sample. Simulation results show surprisingly little gain, and more often a loss in efficiency of these new designs relative to nested case-control sampling. This is due to the sampling-induced covariance between score terms. We conclude that the often stated intuition that nested case-control sampling does not make good use of sampled individuals' covariate histories is false.

Gender Differences in Household Resource Allocations

Abstract: Using household survey data from the United States, Brazil, and Ghana, the authors examine the relationship between paternal education and child height, an indicator of health and nutritional status. In all three countries, the education of the mother has a bigger effect on her daughter's height; paternal education, in contrast, has a bigger impact on the son's height. There are apparently, differences in the allocation of household resources depending on the gender of the child and these differences vary with the gender of the parent. In Ghana, relative to other women, the education of a woman who is better educated than her husband has a bigger impact on the height of her daughter than her son. In Brazil, women's nonlabor income has a positive impact on the health of her daughter but not her son's health. If relative education of parents and nonlabor income are indicators of power in a household bargaining game, then these results suggest that gender differences in resource allocations reflect both technological differences in child rearing and differences in the preferences of parents.

Wages, Schooling and Background: Investments in Men and Women in Urban Brazil

Abstract: Numerous empirical regularities of the urban Brazilian labor market are documented using the 1982 Pesquisa Nacional por Amostra de Domicilios (PNAD). Adopting a semi-parametric approach special attention is paid to the shape of the relation between wage and years of schooling. The authors focus on heterogeneity in this relationships: by gender region choice of sector of employment age-cohort and race. They also examine the role of family background in wage determination both directly and through its influence on labor market participation choices. The effect of education on both (the logarithm of) wage and sectoral choices is neither linear nor smooth. Returns to post-primary schooling tend to be far higher than returns to less schooling. This is true for both market and self-employment work. The differential is greater in the Northeast than the South and has been increasing over time. Furthermore additional years of schooling sometimes have either no effect or even a negative impact on expected wages. The heterogeneity in the shape of the wage-education function suggest that the lack of smoothness cannot be explained by credentialism but rather indicates self-selection among those who choose to continue at school. Family background factors are found to play a significant role in wage determination though of far less magnitude than own education. Unobserved characteristics of those entering the wage vs self-employment sectors or not participating in the labor force at all are also found to affect log-wages in plausible ways suggesting that workers are sorting according to their comparative advantages. (authors)

Public Policy and Anthropometric Outcomes in the Cote d'Ivoire

Abstract: Household survey data from the Cote d'Ivoire are used to examine the impact of public policies on child height, child weight for height and adult body mass index. Economic adjuntment programs in the 1980s were accompanied by reduced availability and quality of health care services and increases in relative food prices. The health of Ivorians was probably adversely affected by these changes.

Like Father, Like Son: Gender Differences in Household Resource Allocations

Abstract: Using household survey data from the United States, Brazil and Ghana, we examine the relationship between parental education and child height, an indicator of health and nutritional status. In all three countries, the education of the mother has a bigger effect on her daughter's height; paternal education, in contrast, has a bigger impact on his son's height. There are, apparently, differences in the allocations of household resources depending on the gender of the child and these differences vary with the gender of the parent. In Ghana, relative to other women, the education of a woman who is better educated than her husband has a bigger impact on the height of her daughter than her son. In Brazil, women's non-labor income has a positive impact on the health of her daughter but not on her son's health. If relative education of parents and non-labor income are indicators of power in a household bargaining game, then these results suggest that gender differences in resource allocations reflect both technological differences in child rearing and differences in the preferences of parents.