School achievement higher for children in nuclear families than for children in blended or single-parent families

Educational outcomes of children in stable blended families are substantially worse than those of children reared in traditional nuclear families, according to a study published in the most recent issue of the journal Demography.

Both stepchildren and their half-siblings who are the joint children of both parents achieved at similar levels, well below children from traditional nuclear families, according to economists Donna Ginther of the University of Kansas and Robert Pollak of Washington University in St. Louis.

Family structure may effect educational outcomes, according to a study by WUSTL professor Robert Pollak.
Family structure may effect educational outcomes, according to a study by WUSTL professor Robert Pollak.

Policymakers have focused on the differences between two-parent families and single-parent families, but this study finds that a crucial distinction is between children raised in traditional nuclear families (families where all children are the joint children of both parents) and children reared in other family types (single-parent families or blended families).

“Popular discussions often begin with the fact that children from single-parent families experience worse educational outcomes than children from two-parent families, and jump to the conclusion that this difference in educational outcomes was caused by the difference in family structure,” says Pollak, Ph.D., the Hernreich Professor of Economics in Arts & Sciences and the John M. Olin School of Business. “Our results call into question this causal interpretation of the correlations between family structure and outcomes for children. Honest policy debates must rest on beliefs about causal relationships, not just on correlations.”

Ginther and Pollak examined achievement test results and levels of educational attainment (high school completion, college attendance, college graduation) of 11,064 children tracked up to 15 years in two large national studies — the National Longitudinal Survey of Youth 1979 and the Panel Study of Income Dynamics.

They found that stepchildren and their half-siblings who spent their childhoods with their two biological parents achieved at virtually similar levels, significantly below children from traditional nuclear families.

When the researchers took into account family income and mother’s education, the relationship between family type and children’s educational outcomes weakened substantially and was often statistically insignificant for children living with a single-parent.

“Family structure may be associated with other factors that contribute to how well children do in school and whether they go on to college, such as limited time and money,” says Ginther. “Or parents in traditional nuclear families may differ in ways we don’t fully understand from parents in blended families and those in single-parent families.” For example, the presence of stepchildren could be a source of stress, affecting educational outcomes for joint children in blended families.

These findings should make policymakers cautious, Ginther argues.

“Policies that are intended to improve children’s well-being often focus on promoting two-parent families, which is easy to observe and, some believe, relatively easy to influence through tax and welfare policy, couples’ counseling, or laws governing marriage, divorce, and child support,” says Ginther. “If the relationship between educational achievement and family structure is influenced by variables other than family type, then policies that seek to affect family structure may have little or no effect on outcomes for children.”

The full article, “Family Structure and Children’s Educational Outcomes: Blended Families, Stylized Facts, and Descriptive Regressions,” is available on Or call the Center for Public Information on Population Research, 202- 939-5409. The Center, a project of the Population Reference Bureau, is funded by the National Institute on Child Health and Human Development.