Abstract
This paper examines the relationship between women's status and fertility in two regions of rural Bangladesh. Based on individual and household-level survey data, women's status is measured through four constructs. The covariates of these four aspects of women's status vary considerably and confirm the view that women's status is conceptually and operationally complex. For all aspects, however, variation between villages accounts for the largest share of explainable variance. Proxy measures of status do not provide uniform relationships with all facets of status. Further, the paper shows that women's status is an important determinant of fertility; of the variance in total children ever born that can be explained by factors other than age, nearly 30 per cent is due to direct measures of women's status; this is as much as can be explained by all other socio-economic variables combined. Thus, models of fertility that rely solely on proxy measures of women's status will be underspecified. In addition, measurement of women's status that does not account for the bias that women's status and fertility are simultaneously determined in patriarchal societies will misstate the direction, and underestimate the effects, of status on fertility. Lastly, different dimensions of women's status influence fertility differently – in terms of magnitude, direction and statistical significance.