The Jefferson Scholarship is one of the most competitive merit scholarship programs in the world. The program, which provides 30 exceptionally talented high school seniors with free tuition, room and board, and a generous stipend to attend the University of Virginia, draws more than 1,300 nominations from schools in the United States and 23 other countries around the globe every year.

The pool of nominees, who are nominated by their schools based on their demonstrated excellence in leadership, scholarship, and citizenship, is unusually diverse in most respects, representing every major continent of the world, a range of political, religious, and ideological views, and interests that span the intellectual and extracurricular spectrum, from a young scientist studying "The Effect of Hydrocarbon Contamination on the Root" to a recent all-Ireland Irish Dancing champion.

But based on my experience as a regular faculty reviewer for the scholarship, the nominees are not diverse in one respect: family structure. The vast majority of the nominees come from intact, two-parent families.

This year, for instance, I found that more than 80 percent of the nominees I reviewed came from a home headed by their own married parents. This is striking because about half of American high school seniors do not live with both of their parents, according to the American Community Survey. My experience here suggests that students are much more likely to excel in school when they grow up in a stable, two-parent home.

But this isn't the full story. It may seem intuitive that children who come from stable homes where there are two adults to care for them, help them with schoolwork, and participate in after-school activities will be more likely to succeed academically. But international data suggests this is not always the case.

According to The World Family Map, a new report I helped edit, whether you're helped by having two parents at home is largely a function of where you live. Children in wealthy or middle-income countries - like the vast majority of those applying for Jefferson Scholarships - are helped enormously by two-parent family units. In the developing world, by contrast, children raised by a single parent are just as likely - in some cases more likely - to succeed.

The results for the developed world were as you might expect: Children from single-parent families were more likely to report lower literacy scores in 14 out of the 20 Organization for Economic Cooperation and Development (OECD) countries examined in the report even after controlling for differences in parental education, employment and wealth. They were also more likely to have repeated a grade in school in 15 out of the 18 OECD countries surveyed for this outcome.

This was true in places as culturally varied as Australia, Israel, Mexico, Spain, Sweden and the United States. Children from single-parent families in Australia are 55 percent more likely to have ever repeated a grade, compared to their peers in two-parent families, even after controlling for socioeconomic differences. Similar patterns obtain for children of single parents in Chile (69 percent more likely), Israel (194 percent), Spain (63 percent), Sweden (78 percent), Turkey (95 percent), and the United States (54 percent).

Children from two-parent homes in OECD countries as varied as Sweden, with a generous welfare state, and the United States, with more miserly government services, are also less likely to fall behind in school and more likely to excel in reading, compared with their peers from single-parent families. This advantage, in turn, is likely to translate into benefits that extend into adulthood, including better work opportunities, higher incomes, and more wealth for children raised in a two-parent family, given what we know about the link between family background, education, and adult outcomes in the United States.

The picture looks quite different in the developing world, especially in Asia, the Middle East and sub-Saharan Africa. Take being behind in school.

The World Family Map examined 15 countries in the developing world on this outcome. Children from single-parent families in Egypt are 34 percent less likely to be behind in school (measured by being below the appropriate grade for one's age), compared with their peers in two-parent families, after controlling for socioeconomic differences.

And similar patterns for children of single parents can be found in Ethiopia (19 percent less likely), India (24 percent less likely), Kenya (24 percent less likely), and Nigeria (28 percent less likely), all places where children are less likely to be behind in school compared to children from two-parent families. More significantly, in none of the 15 countries examined in Asia, Latin America, the Middle East, and sub-Saharan Africa were children from two-parent families significantly less likely than children from single-parent families to be behind grade for their age.

So, at least when it comes to education, a two-parent family seems less important, or even unhelpful in some cases, in the developing world. What accounts for this surprising result?

The report suggests three possible answers to this question. First, children living with single parents in the developing world often live with and depend heavily on extended family members who can buffer against the absence of a second parent, usually a father.

By contrast, children living in the developed world - where extended families are much less salient forces in children's lives - often do not have much access to the time, attention, and money of grandparents and other non-parental kin. For instance, children living with only one parent in India may be more likely to rely on "Daadi" and "Dada" - grandma and grandpa - than children raised in single-parent families in the United States.

Second, schooling effects may drown out family effects in the developing world. Even though there are clear differences in school quality within the developed world, those differences are more dramatic in the developing world.

For instance, "something as basic as whether a teacher comes to class regularly" is an important factor in children's achievement in rural India, according to the report. This is a problem that rarely affects children in developed countries. Children in developed countries can depend upon a minimum level of school competency that may make their family resources more salient in determining their educational fortunes.

Third, fathers in the developed world - especially in Europe, North America and Oceania - appear to take a much more hands-on approach to their children's schooling than fathers in much of the developing world.

By contrast, fathers in the developing world may be less involved or, in some cases, more likely to waste family money on their own pursuits, rather than on their children's education. Research by economist Cynthia Lloyd has found, for instance, that single mothers in sub-Saharan Africa are more likely to devote family money to children's schooling than are father-headed households. Thus, the educational value of two parents may depend, in part, on men's willingness to invest practically and financially in their children's schooling.

The dramatic difference between developed and developing countries in the study is ironic because, in recent years, the two-parent family has lost more ground in Europe, North America and Oceania than it has in East Asia. For instance, just 12 percent of children in Japan live in single-parent homes, compared with 27 percent of children in the United States. Moreover, extended families in the West tend to be comparatively weak, by no means ready to step into the gap created by a nuclear family in retreat.

Thus, at least in the West, parents, and nations, committed to giving their children an educational leg up in the global economy may wish to rethink their commitment to the nuclear family. If they want the rising generation to have every educational advantage in an increasingly competitive global economy, parents and policymakers would be wise to focus not just on the quality of their children's schools, but also the quality and stability of their children's homes.

- - -

W. Bradford Wilcox, director of the National Marriage Project at the University of Virginia, is co-primary investigator of The World Family Map, a new report from Child Trends.