by age group for two censuses or surveys but only for those countries with a separate consensual union category, which is a subset of countries in which consensual unions are common.

Tables 7-1 and 7-2, based on the UN database, show the percentage of women and men ever in a union1 by age group from data collected at two different points in time. For women the age groups are 15-19, 20-24, and 25-29, and for men—for whom marriage during the teenage years is rare—they are 20-24 and 25-29. An annualized rate of change is calculated, since the interval between the two times varies by country. Note that for these tables as well as all subsequent tables that present regional data, the individual country data are provided in appendix tables at the end of the chapter.

For all regions except former Soviet Asia and South America, where early marriage was not that common even 10-20 years ago, teenage marriage has declined among women; whereas 27 percent of 15-19-year-old women in the developing world were married in 1970-1989, 21 percent were married in 1990-2000.2 The reduction in the percentage of married 15-19-year-olds is particularly striking in Africa.3 The percentage married among 20-24-year-olds has also fallen markedly in most regions, with the exception, again, of South America. While the majority of women in developing countries were married by ages 25-29, there are regions in which 15-25 percent of women were still not married by their late 20s, including South America, the Caribbean and Central America, the Middle East, the former Soviet Asia, and Eastern and Southern Africa, as well as certain countries in Asia (e.g., Myanmar, the Philippines, and Thailand).

For the most part, regions with a relatively high percentage married at younger ages also had a relatively high percentage married at older ages (e.g., Africa) and vice versa (e.g., the former Soviet Asia). The exception is


For countries in which consensual union is uncommon, the percentage shown is simply those who ever married.


Given that the rate of growth in cohort size is currently declining in most parts of the developing world including in China, where the rate of decline is accelerating, even if the percentage married at each age were unchanged, it would appear that the percentage of 15-19-year-olds married had increased because the ratio of 15-17-year-olds to 18-19-year-olds would be smaller and, relative to 18-19-year-olds, 15-17-year-olds are less likely to be married. The fact that we observe a reduction in the percentage of 15-19-year-olds married indicates that the true decline in marriage is likely to be slightly larger.


Not all demographers would agree with this observation. Van de Walle and Baker (2004:17) assert that for Africa “there are good reasons to argue that the age at union has changed little.” They base this claim on the belief that visiting unions, in which an individual has “stable noncohabiting” partners, have increased. Because most women in a visiting union would not describe themselves as married, they are not categorized as such in the DHS.

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement