According to the U.S. Census Bureau's 2005 American Community Survey:

• Married-couple families comprised nearly 53 percent of non-Hispanic white households. Additionally, 9 percent of households were maintained by a woman with no husband present and 28 percent in which the householder lived alone.

• Married-couple families comprised nearly 29 percent of black or African-American households. About 30 percent of households were maintained by a woman with no husband present and in 31 percent the householder lived alone.

• Married-couple families comprised about 40 percent of American Indian and Alaska Native households. About 20 percent of households were maintained by a woman with no husband present and in 25 percent the householder lived alone.

• Married-couple families comprised about 59 percent of Asian households. Among the Asian households, 9 percent were maintained by a woman with no husband present and 20 percent in which the householder lived alone.

• Married-couple families comprised 50 percent of Native Hawaiian and Other Pacific Islander households. Among the Native Hawaiian and Other Pacific Islander households, about 15 percent were maintained by a woman with no husband present and in 20 percent the householder lived alone.

• Married-couple families comprised 49 percent of Hispanic households. Among Hispanic households, 19 percent were maintained by a woman with no husband present and in about 16 percent the householder lived alone.

"Only solitary men know the full joys of friendship. Others have their family-but to a solitary and an exile his friends are everything."

The U.S. Bureau of the Census defines a family as "two or more persons related by birth, marriage, or adoption who reside in the same household." Some legal and social theorists, however, would like to see recognition of all sorts of "family-like" relationships - including cohabitation and same-sex partnerships - on the assumption that chosen social ties are becoming more important than given or family connections. Yet a British study of family and friends finds no evidence for such a notion and suggests just the opposite, that one's family is the main source of close relationships or friendships throughout life.

Sociologists Ray Pahl and David Pevalin examined ten years of data from the British Household Panel Survey, which has annually surveyed 5,000 households since 1991, asking respondents in even-number waves to identify their closest friends who do not live in the same household and whether their closest friend is a relative. Then in odd-number waves, the survey asked respondents to identify who they count on the most for emotional support, whether a friend, relative, or spouse/partner.

They found that while younger people, relative to older respondents, were more likely to maintain close friends that were outside the family, the proportion of respondents who named relatives as close friends increased significantly as they grew older. The pattern was pronounced among those who were older than 46, but the shift of choosing friends outside the family to friends within the family (not counting spouses) remained statistically significant with the youngest category in the study, respondents ages 16 to 25 (p < .001).

In the odd-numbered waves of data, the researchers found that the youngest respondents increasingly identified their spouses over a non-kin friend as their closest friend. By middle age, spouses were named their closest friend by the majority of the respondents. Among the older cohorts, this pattern changed, as respondents increasingly named a relative as closest friend, a pattern that the researchers attribute to the death or divorce of one's spouse.

While Pahl and Pevalin do not make the claim, their findings provide evidence that the "family-like" relationships activists favor pale in comparison to the real thing - family relationships.