The effect of sex preference on contraceptive use and fertility in rural south India.

The aim of this study is to examine the influence of sex preference on contraceptive use in the southern state of Karnataka in India. Two sets of data are used in the analysis. One set is cross-sectional data collected during 1985-90 among all currently married women aged 15-44 years in rural areas representing more and less developed regions of Chickmagalore Belgaum Bijapur Dharwad and Gulbarga districts. The other data set comprises data from family planning evaluation surveys conducted during 1982-89 in rural areas of Belgaum Bellary Bijapur North Kannada and Shimoga districts among a sample of sterilization and IUD users. Analysis focuses on the trends in the number of living children by sex among users and nonusers. Trends are described for the sex distribution of all children and the sex of the last child at initiation of contraceptive use. Analysis focuses on the sex of a desired child among nonusers. The methods of Arnold are used to estimate contraceptive prevalence in the absence of sex preference. The extent of influence of sex preference on contraceptive use is calculated as the difference between actual and adjusted contraceptive prevalence rates. Contraceptive prevalence was estimated to be 62% in Chickmagalore (1985) 39% in Bijapur (1987) 37% in Dharwad (1989) 46% in Belgaum (1990) and 33% in Gulbarga (1990). Reversible method use was very low (under 3%) in all districts. Findings reveal that acceptors had more male children in all districts. Among nonusers the mean number of female living children in all districts was equal to or slightly higher than the mean number of male children. Cross-sectional data reveal that variation among the number of female children was greater than variation among the number of male children among acceptors. The reverse was true among nonacceptors. The analysis of sex ratios among last children reveals that more contraceptive users had a male as a last child. It is suggested that the number of living children among sterilization acceptors was lower among acceptors in some districts. It is concluded that in the absence of sex preference contraceptive prevalence would be higher in all districts. Son preference is found to have a greater impact on usage in areas of low prevalence.