women are delaying marriage because of shifting gender roles, it is important to document these patterns of behavior and understand what the potential implications are both for the individuals and for the larger society.
In recent years, few demographers have heeded van de Walle’s (1993) appeal to explore the process and timing of marriage for its own sake. Yabiku and colleagues’ (Yabiku, 2003; Yabiku et al., 2002) analysis of the effect of community variables on the timing of marriage in a region of Nepal experiencing rapid social change is a notable exception.1 There is, however, a large descriptive literature. Although lacking much in the way of explanatory variables, this research documents trends and differentials in the age of first union among women, with a particular focus on the practice of early marriage in the developing world (see, e.g., Choe, Thapa, and Achmad, 2001; Heaton, Forste, and Otterstrom, 2002; Jejeebhoy, 1995; Rashad and Osman, 2003; Singh and Samara, 1996; Westoff, 2003).
This interest in early marriage reflects the concern of human rights and reproductive health advocates, who in putting “child marriage” on the international agenda have emphasized the potentially harmful consequences for young women of marrying too early. Researchers at the International Center for Research on Women (2004) highlight these possible problems in a rather dramatic fashion:
Child brides are robbed of the ordinary life experiences other young people take for granted. Many are forced to drop out of school. Their health is at risk because of early sexual activity and childbearing. They cannot take advantage of economic opportunities. Friendships with peers are often restricted. Child marriage deprives girls of basic rights and subjects them to undue disadvantage—and sometimes violence. Countries with a high percentage of child marriage are more likely to experience extreme and persistent poverty, and high levels of maternal and child mortality.
While focus on marriage prior to age 18, the internationally established age of adulthood, has gained prominence, research has yet to establish the
There are also several studies that predate van de Walle’s call for further research. Fricke, Syed, and Smith’s (1986) analysis of marriage timing strategies in Pakistan is noteworthy as is Lesthaeghe, Kaufman, and Meekers (1989) investigation of nuptiality regimes in sub-Saharan Africa, where the timing of marriage and the practice of polygyny were explored in great depth. This latter study was path breaking in linking ethnographic data (including measures of dependence on subsistence agriculture, lineage systems, inheritance, and presence of various types of chiefs) to demographic data. Malhotra and Tsui’s (1996) study of the effect of norms about marriage—including the importance of setting up an independent household, the desire to work before marriage, and expectations about arranged marriage—on marriage timing in Sri Lanka is also an important contribution to the literature. To the best of our knowledge, it is the only analysis of marriage that uses panel data; however, while the attitudinal variables included in the event history models are measured prior to marriage, they are still likely to be endogenous to marriage timing.