School of Medicine

This study examines associations of son preference, desire for sons more than daughters, with abortion history among ever-married women in Nepal (N=9837), using nationally representative 2011 Nepal Demographic and Health Survey data. Logistic regression analyses, adjusted for sociodemographics and other relevant covariates, assessed associations between son preference (assessed by self-report and number/sex of children) and abortion history. Having multiple sons rather than 0 or 1 son was associated with abortion history, where presence/absence of daughters and self-reported son preference was not. Son preference appear to play a role in abortion decision-making in Nepal that cannot be captured by self-report.