John Casterline (Department of Sociology, Ohio State University)
04/07/2008, at noon in room 6050 ISR-Thompson.
The prevention of unwanted births has long been a fundamental justification for investment of public and private resources in family planning services. Unwanted childbearing is assumed to have detrimental consequences – for the child and for its family and larger community – that are distinctive, substantial, and potentially long-term. There is, however, surprisingly little empirical research that offers a solid scientific basis for this assumption, especially in low-income non-Western societies. We examine the impact of child wantedness on child survival and schooling in rural Bangladesh via analysis of longitudinal data collected in the period 1982-2007. The sample of children is large as compared to samples in previous research, and child wantedness is measured prospectively and on a sex-specific basis. Two analytical strategies to remove confounding effects of unmeasured factors are employed: models with fixed effects for sibling set, and a “natural experiment” provided by the random assignment of child sex. The regression estimates indicate that unwanted births suffer higher mortality during infancy (odds ratios in excess of 2.0) and complete fewer years of schooling. Higher mortality is characteristic of births unwanted either because the desired family size has been exceeded or because the child is the wrong sex. Effects on schooling, in contrast, are concentrated among children who are unwanted due to their sex (females).