John Martin (Visiting Research Associate, School of Advanced Research; Professor Emeritus, Department of Anthropology, Arizona State University)
Abstract. The ratio of male to female births [(m/f) x 100] is reported to vary from 100 or less among national populations in southern Africa to 105-6 among national populations in Europe, Asia, and in the United States. In northern sub-Saharan Africa sex ratios at birth (SRb) are reported to be high as 108 to 110. Differences this large between national and regional populations are not likely to result from chance. Such variations have been attributed to genetic differences and to such biological and social/behavioral factors as maternal age, parity (the number of mother’s previous births), polygyny as against monogamy, the presence or absence of a male partner in the household, and the frequency of unprotected sexual intercourse in the woman’s fertile period.
Here we present data on the SRb among 5,723 births to married mothers in southern Africa (sex ratio=100.1), and introduce another variable: the presence or absence of one or more other fertile females in the mother’s household. Analyses of the data with contingency tables, factorial designs, and weighted logistic regressions of the log odds ratios over the X variables show that when we control for the interactions of mothers’ residential arrangements with marriage forms, the presence or absence of husbands in the household, frequency of sexual intercourse, maternal age, and parity, the mother’s residential arrangements prove to be critical. When the interactions of maternal residential and marital forms are controlled, the associations of marriage forms with sex ratios lose their significance. We also find no significant associations between sex ratios and the presence or absence of a husband in the household.
But, these analyses further show that when mothers are under 30 years of age, parity is negatively related to the SRb. However, among births to mothers who were not sharing their households with other fertile women, the SRb rises significantly between maternal ages 30-34 and remains high (≥130) thereafter, not significantly varying with increases in age or parity. However, among births to mothers who were sharing their households with other women of fertile age, the negative relationship of the SRb with parity continues after age 30 to the end of reproductive life. Overall, the sex ratio among births to mothers not sharing their households with other fertile women is 109.1 while among births to mothers who were, it is 82.6 (p<0.0001, OR=130.1).
Previous research on the female reproductive cycle – in particular, on developments during the follicular phase – suggest that the cited variations in SRb stem from the interactions of ages and parities, and mothers’ residential arrangements with shared, probably genetically controlled, processes in the female reproductive tract interacting with sperm swimming behavior.