Son and daughter preferences in Benighat, Nepal: implications for fertility transition
Niraula, B.B.; Morgan, S.P.
Social Biology 42(3-4): 256-273
1995
ISSN/ISBN: 0037-766X PMID: 8738551 DOI: 10.1080/19485565.1995.9988906Document Number: 291237
This study is similar to a 1979 study by Karki in a rural district and Pokhara town in the hill region of Nepal. The study was conducted in 1988 in Benighat and represents well the central hill society and all major caste-ethnic groups. Findings in both studies were similar. Findings indicate that the prime motivation for having a son was old age support (38.9%), followed by lineage (35.3%). This agricultural subsistence society uses intergenerational transfers as a solution to old age dependency. The Hindu code of conduct reinforces the transfer of family name through male offspring. Other reasons for desiring sons included religious beliefs, agricultural labor, and companionship. Villagers identified the successful and powerful families in Benighat as having many sons. Daughters were desired for religious reasons (41.6%), help around the house (22.5%), and lineage (15.2%). Hindus assign great value to marrying a virgin daughter. Young girls help a lot with caring for livestock and fetching water, fuel, and fodder. Findings indicate that 377 wives reported a desire to have more sons than daughters, and 296 wives reported they would have an equal number of sons and daughters if they could start all over. Responses to questions about satisfaction with the number of children indicated substantial sex preference. Satisfaction was higher at higher parities. Women who were dissatisfied tended to report wanting another child. Findings from logistic models indicate that the odds of being satisfied were 5.5 times greater if there were two sons and 8.8 times greater if there were one son and one daughter than if there were two daughters. Those with two sons were 70% more likely to be using contraception. There was some support for the hypothesis that son preference was greater among the landless, but formal tests indicated that there were few differences across four landholding groups and five castes.