Abstract
In this article, methods are presented for measuring the level and age pattern of sterility from incomplete birth histories, such as those that can be collected in demographic surveys of women who may not yet have reached the end of their reproductive span. The characteristics of the methods are examined in a simulation study that demonstrates that estimates based on information about fertility subsequent to a given age are more robust to variations in reproductive determinants, sample size, and sampling variation than the other measures, which include the frequently employed length of the open birth interval. In an illustrative analysis of World Fertility Survey data from sub-Saharan Africa, sterility was found to be high in Cameroon, intermediate in Lesotho and Sudan, and low in Ghana and Kenya relative to an English historical population.
