60 Years After Coleman: Research Shows Family Structure Is Still Needed

Education policy has long operated under a convenient assumption: if schools improve, student outcomes will improve with them. Better standards, stronger accountability, more effective teachers, and greater transparency have all been central to reform efforts over the last three decades. Those priorities have mattered. But they have also obscured a harder truth: much of educational inequality is formed before a child ever enters a classroom.

A major study released in 2024 by the Thomas B. Fordham Institute reinforces that point—and suggests that one important variable remains conspicuously absent from most state education data systems: family structure.

On August 21, 2024, Fordham published Explaining Achievement Gaps: The Role of Socioeconomic Factors, a rigorous analysis by Eric Hengyu Hu and Paul Morgan examining how much of America’s racial achievement gaps can be explained by differences in family background. Using two nationally representative federal datasets—the National Center for Education Statistics, Early Childhood Longitudinal Study, Kindergarten Class of 1998–99 (ECLS-K: 1998) and the Early Childhood Longitudinal Study, Kindergarten Class of 2010–11 (ECLS-K:2011)—the authors estimated how much of racial differences in reading and mathematics performance can be attributed to measurable household characteristics.

Their central finding was significant: family and socioeconomic conditions explain a substantial share of racial achievement disparities—but not equally.

Fordham found that socioeconomic factors explain 34 to 64 percent of the Black–White achievement gap, but 51 to 77 percent of the Hispanic–White achievement gap. That distinction matters because the Black–White achievement gap is also consistently larger than the Hispanic–White gap across most national datasets, subjects, and grade levels.

Taken together, those findings suggest something deeper: the Black–White gap is not simply a larger version of the Hispanic–White gap. It appears to be a different policy problem altogether.

The smaller Hispanic–White gap appears more responsive to measurable family and assimilation-related variables—household income, parental education, family structure, and likely English-language acquisition. By contrast, the larger Black–White gap remains more persistent even after those same factors are taken into account, suggesting that traditional socioeconomic measures alone do not fully explain the disparity. This implies that policymakers are confronting not one achievement gap, but at least two distinct educational challenges—each requiring different policy responses and more precise diagnostic tools.

That insight alone should force policymakers to rethink simplistic, one-size-fits-all approaches to “closing achievement gaps.” But the study’s most overlooked finding may be buried inside its broader “SES+” model. Included within that framework is a household structure variable measuring whether a child lives in a two-parent household, a single-parent household, or another guardianship arrangement. At first glance, this variable appears secondary; household income and maternal education explain more variation overall. Yet a closer examination shows that family structure is far more consequential than current policy discussions acknowledge.

Fordham estimates that family structure independently explains 10 to 22 percent of the Black–White achievement gap and 1 to 4 percent of the Hispanic–White achievement gap, depending on grade level and subject area. Those percentages may appear modest beside larger socioeconomic variables, but in education policy they are substantial. States routinely redesign programs and allocate millions of dollars based on smaller effects. This is especially notable because according to data from the U.S. Census Bureau American Community Survey, approximately 43 percent of Latino children and only 23 percent of Black children live with their married biological parents. Latino children are therefore nearly twice as likely to live in two-parent biological households.

The significance of this finding is not merely statistical.

Family structure functions as a proxy for a broader set of developmental conditions known to shape educational attainment: household stability, adult supervision, time investment, residential continuity, and what sociologist James Coleman described decades ago as “social capital”—the relationships and routines that support child development. Schools can reinforce these advantages. They do not create them.

The Fordham analysis also raises an uncomfortable but unavoidable question: what happens when race and family structure intersect with gender? Although not the explicit focus of Fordham’s report, one of the most consistent findings in American education data is that Black males perform at the lowest levels of any major race-by-gender subgroup on most academic indicators, from K–12 proficiency to college completion.

At the same time, emerging evidence suggests that boys are disproportionately harmed by family instability relative to girls. Research from the Northwestern Institute for Policy Research and the Russell Sage Foundation found that family disadvantage produces larger negative effects on boys’ behavior, academic achievement, and educational attainment than on girls raised in comparable environments.

That helps explain a persistent but often overlooked pattern in education data: girls generally outperform boys across racial groups, but the gap is especially pronounced among Black students—suggesting that household instability may compound existing racial disadvantage most sharply for Black males.

This should matter for policy.

Most state education dashboards now track dozens of indicators: academic proficiency, chronic absenteeism, graduation rates, school climate, advanced coursework participation, and postsecondary readiness. Some states have gone further, incorporating measures of social-emotional development and student well-being.

But almost none systematically report indicators related to family stability.

That omission reflects a longstanding discomfort in education policy. Family structure is often treated as politically sensitive and therefore analytically untouchable. But avoiding measurement does not make a variable irrelevant. It simply makes policy less accurate.

No serious scholar argues that family structure determines destiny. Many children raised in single-parent households thrive; many children raised in two-parent households struggle. Family form is not fate. But educational accountability has never required deterministic measures. States already track poverty, absenteeism, and neighborhood disadvantage—not because those variables guarantee outcomes, but because they help identify opportunity and risk.

Family structure should be treated the same way.

If states are serious about measuring educational opportunity rather than merely reporting educational outcomes, they should begin incorporating household stability indicators into their public data systems. That could include the percentage of children living in two-adult households, household mobility rates, residential instability, and related measures of family support. Such indicators should not be used to stigmatize communities. They should be used to improve diagnosis.

For decades, school reform has focused primarily on what happens inside the schoolhouse. That work remains important. But the next generation of education policy must become more honest about what happens outside it. The lesson of Fordham’s study is straightforward: educational inequality is not produced only in classrooms.

It is also produced in homes. And if policymakers are serious about closing achievement gaps, family structure can no longer remain invisible in the nation’s education data architecture.