Conservatives see two-parent families as a solution to improving student achievement. Things are not that simple

I want to study the relationship between family structure and student achievement through family income. Single-parent families are more common in low-income communities, and I don’t want to conflate achievement gaps by income with achievement gaps by family structure. For example, 43% of low-income eighth-graders live with only one parent, compared with 13% of their higher-income peers. I wonder if children who live with only one parent do worse than children with the same family income who live with both parents.

To analyze the latest data on the 2024 NAEP exams, I used NAEP Data Explorer, a public tool developed by the testing organization ETS for the National Center for Education Statistics (NCES). I told an ETS researcher what I wanted to know, he showed me how to generate a crosstab, and then I independently replicated those results on four tests: reading and math in fourth and eighth grades. Finally, I reviewed the results with a former senior official at NCES and current staff on the board of governors responsible for overseeing the NAEP evaluation.

The analysis revealed a surprising pattern.

Among students from low-income families, there is little difference in their performance depending on their family structure. Fourth- and eighth-grade students from low-income families scored about the same whether they lived with both parents or with one parent. Two-parent families confer no measurable academic advantage among this group. Fourth grade reading is a great example. Among students in the bottom third of socioeconomic status, those living with their parents scored 199. A student who lives with his mother gets a score of 200. The results are almost identical, if anything, children of single mothers are taller.

However, as socioeconomic status increases, differences in family structure become more pronounced. Among middle- and upper-income students, those who live with two parents tend to score higher than their peers who live with only one parent. The gap is largest among the wealthiest students. For example, in fourth-grade reading, high-income children living with their parents scored 238, 10 points higher than their peers who lived only with their mothers. Experts debate the meaning of NAEP scores, but some equate 10 NAEP scores to one school year’s worth of learning. This is very important.

Family structure has less impact on low-income students’ performance

Still, it’s better to be rich in a one-parent family than poor in a two-parent family. High-income students raised by a single parent outperformed low-income students living with both parents by at least 20 percentage points, underscoring the importance of money and the advantages it brings (such as access to resources, stable housing and educational support) rather than family composition itself. In other words, income far outweighs family structure when it comes to student outcomes.

Despite NAEP’s data, Jonathan Butcher, acting director of the Heritage Foundation’s Center for Education Policy, maintains the argument that family structure matters to student achievement. He noted that research since the landmark Coleman Report in 1966 has consistently found a link. Most recently, in a 2022 American Enterprise Institute-Brookings Institution report, 15 academics concluded that “children who grow up in stable married families are more likely to do well in school and generally have higher grade point averages” than children who grow up in less stable married families. Two recent books, Brad Wilcox’s Married (2024) and Melissa Kearney’s Parental Privilege (2023), also illustrate this point, noting that children raised by married parents are about twice as likely to graduate from college as unmarried children. However, it is not clear to me whether all of these analyzes disaggregate student achievement by family income, as I did with the NAEP data.

Family structure is an enduring theme among conservatives. Just last week, the Heritage Foundation released a report on strengthening and rebuilding American families. In a July 2025 newsletter, Robert Pondiscio, a senior fellow at the American Enterprise Institute, a conservative think tank, wrote, “The most effective intervention in education is not another literacy coach or an SEL program. It’s dads.” He cited Good Fathers, Thriving Kids, a June 2025 report by scholars and advocates. (Disclosure: A group led by Richard Reeves, one of the authors of this report, was one of the funders of the Hezinger Report.)

NAEP data partially support this conclusion, but only for the relatively small share of students from higher-income families (the share of upper-income children who live with their mothers only is between 7 and 10 percent. Eighth-grade students have a higher share of single-parent parents than fourth-graders.) This is not true for the low-income students with whom Pondiscio and the scholars are primarily concerned.

The data has limitations. The NAEP survey does not distinguish between divorced families, grandparent-led families, or same-sex parents. Shared custody arrangements may be grouped with two-parent families because the children may say they live with their mother and father, even if not at the same time. Even so, these nuances are unlikely to change the core finding: For low-income students, academic performance is broadly similar whether they live with both parents all the time, or sometimes with one parent.

The bottom line is that calling for new federal data collection by family structure, as outlined in Plan 2025, may not reveal what advocates expect. A family’s bank account is more important than a wedding ring.