Parental attitude and gender bias in the schooling and health outcomes of children: experimental and survey evidence from Bangladesh

This dissertation compiles three complete and allied papers, focusing children’s schooling and health with special emphasis on parental attitudes and gender bias. The first paper examines the effects of the Female Secondary School Stipend Program (FSSSP) in Bangladesh on the schooling of girls, and its subsequent impact on the education of their younger siblings. We find that the education of older siblings has a positive effect on the schooling of younger siblings, and that the effect is slightly greater in magnitude on younger brothers than on younger sisters. When we address the endogeneity of education of older siblings, we find that the gender composition of older siblings generally has no effect on the schooling attainments of younger siblings. The results indicate a 10 per cent increase in the schooling of younger siblings due to the FSSSP. The results suggest that stipend programs could bring both short- and long-term gains, not only via direct benefits to the affected children, but also via indirect benefits to their siblings. The second paper identifies parental attitude towards different-gendered children using an experimental approach. The study was conducted in villages of two districts in Bangladesh, with randomly selected households that had at least two school-aged children (6-18 years) of different genders. In the experiment employed for the study, parents in the households were given an endowment to benefit an anonymous girl or boy at a local or nearby school. The households were randomly assigned to one of four mutually exclusive groups that corresponded to different experiment conditions. The results suggest that there is no systematic inherent bias in parental attitude towards the gender of a child, neither father nor mother is systematically biased for/against the gender of a child, and no significant differences are found between father’s and mother’s behaviors. We also find that joint decisions elicit more biased choices compared to individual choices, but again find no systematic bias toward one gender or the other. The results suggest that subjects revealed their true preferences under the experimental set-up. The third paper investigates how parents’ inherent gender bias is associated with their decisions regarding the schooling and health of their own son and daughter. We focus on five indicators for education, viz., years of schooling, grade for age, enrolment status, education expenditure, and test score; and three indicators for health, viz., incidence of illness, and access to formal treatment and treatment cost, in the case of illness. Although the game outcome suggests that on average there is no systematic inherent bias among parents, inherently biased parents nevertheless allocate resources in a discriminatory manner. The results suggest that boy-biased parents are more likely to enrol their sons in school and to spend more on their sons’ education, and that they are less likely to enrol their daughters in school and spend less on their daughters’ education. The health indicators suggest that boy-biased parents are less likely to seek formal treatment and tend to spend less when a daughter is sick.