The proportion of children living in a traditional nuclear family with their biological mother and father increased from 51 percent in 1991 to 56 percent in 1996, the Commerce Department's Census Bureau.
"While such nuclear families are the majority, a significant additional group of children are living with two parents, one or both of whom are stepparents or adoptive parents," said Jason Fields, author of Living Arrangements of Children: Fall 1996.
These data should not be confused with the results of Census 2000, which are being released over the next three years.
The nuclear family is defined as one that has a mother and a father and their biological children, without any other people in the household. When other family situations are considered (adoptive or stepparents), 50.7 million children (71 percent) lived in two-parent households in 1996.
About 19.7 million children under 18 years of age lived with an unmarried parent in 1996. Of these children, 3.3 million lived with an unmarried parent who was cohabiting with another adult.
Stepparents and adoptive parents were on the increase: In 1996, 5.2 million children lived with one biological parent and either a stepparent or adoptive parent, up from 4.5 million in 1991.
About 6 percent of all children (4.1 million) were living in a household with one or both of their grandparents.
There were 15.3 million children living in households with no brothers or sisters present. Among the 56.1 million children living with siblings, 2.1 million lived with at least one stepsibling (children who identify one parent as biological and one as a stepparent but do not share the same biological parent) and 7.8 million lived with at least one half-sibling (children who share one biological parent but not the other).
The proportion of children who lived with only their father in 1996, 3 percent, was similar to the situation more than 100 years earlier (1880), while the proportion of children living with neither parent declined from 6 percent in 1880 to 4 percent in 1996.
The data were collected from the 1996 panel of the Survey of Income and Program Participation. As in all surveys, the data are subject to sampling variability and other sources of error.