Reproductive Preferences and Subsequent Behaviour: The Sri Lankan Experience

Abstract

On the basis of data collected in the 1982 Sri Lanka Contraceptive Prevalence Survey and a 1985 follow- up survey, the Sri Lanka Contraceptive Survey, reproductive preferences of Sri Lankan women and the reliability of their preferences in predicting fertility behaviour are examined. Wanted family size, ideal family size, desire for another child, and preferred waiting time to the next birth was used as fertility preference indicators. A preference for small families was widely evident, but few women wished to remain childless or to have only one child. The proportion who wanted no more children rose dramatically as the number of living children increased, and at a family size of three is moderate son preference, this was not clearly related to use contraception. A significant minority of women who wanted to cease childbearing used no contraception, and the husband’s disapproval of family planning was strongly related to non-use of contraception. The main finding at the aggregate level is that 1982 statements on desire for another child were highly accurate as predictors of subsequent aggregate fertility. The close correspondence of aggregate level preferences and behaviour during 1982-85 was, however, partly the result of many counterbalancing inconsistencies on the part of individuals. Ability to control unwanted births among Sri Lankan women has increased over according to their expectations. Perhaps among those who wanted additional children many deferred childbearing while seeking better economic opportunities. Even though inconsistencies existed, statements on reproductive preferences were fairly reliable; the number of additional children wanted in 1982 was related to subsequent fertility, which in turn modified women’s preferences in 1985. When the subsequent fertility of women who wanted no more children in 1982 was examined, more of those who believed that their husbands wanted additional children than of those who believed otherwise reported births between the surveys.