It's no secret that in searching for a home, parents scrutinize nearby schools. The wealthy can afford to live in neighborhoods with small school districts, where most other students are wealthy, too.
In the country's 100 largest metropolitan areas, "income segregation is nearly twice as high among households with children as among those without," according to the study from USC sociologist Ann Owens, published in the American Sociological Review in late April.
The study's second major finding is that among the families that have children in those large areas, there is more income segregation in areas that encompass more school districts.
Of course, it's difficult to attribute all income segregation to schools.
Almost three-fourths of children attended the school to which the district assigned them in 2007, according to a government report Owens cites, suggesting that parents want to live near their child's school.
Take Los Angeles: Parents could live in the boundaries of the huge Los Angeles Unified School District, but many wealthy families choose to live in cities like Beverly Hills, which are higher-income and have their own school districts with more concentrated resources.
Los Angeles is slightly less segmented than the average large county in Owens' study.
In Los Angeles County in 2010, the average high-income family with children lived in a place where 34% of their neighbors were in the top fifth of earners with $116,466 or more, and 12% were in the bottom fifth of the population, making $22,246 or less, Owens found.
High-income Angelenos living without children are less segregated: on average, they lived in places where 31% of neighbors earned $116,466 or more and 14% earned $22,246 or less.
Average L.A. families with children in the lowest fifth of earners lived in neighborhoods where 28% had similar incomes, and 10% of neighbors were in the highest fifth of earners. Poor people without children were more likely to be exposed to wealthy neighbors: For the average household without children at the poorest income level, 26% of households had similar incomes and 16% of neighbors were in the top fifth.
And it's more than just geography that divides them. Poor parents are spending about the same on children as they have for decades, but wealthy parents are spending more than ever before. That spending can come in the form of tutoring, extracurriculars or a house in a wealthy neighborhood and an in-demand school district, said Sean Reardon, Stanford's professor of poverty and inequality in education.
Families with children are "seeking out more advantageous environments for their kids at a rate that’s higher than it was in earlier decades," said Jennifer Jennings, a New York University sociology professor who studies education policy and income.
So why does this matter? This study can influence housing policies moving forward, said Reardon.
When wealthy people are concentrated in one area, the resources they give to their schools and to their children are also concentrated: Parents might invest in early education opportunities that give their children an advantage before entering school, but their public schools also have better facilities, smaller classes and longer-term teachers that help overall outcomes. Reardon's research has shown that wealthy students perform at much higher academic levels than poorer ones.
Requiring mixed-housing units and making vouchers available in higher-end areas could help reduce income-based housing segregation and could improve educational outcomes for lower-income students, Reardon said.
On the education side, counties might consider consolidating school districts to improve equality, Owens said. This idea might be controversial in places like Los Angeles, where critics have said the size of Los Angeles Unified sometimes makes it dysfunctional.
The problem of income segregation is different from racial segregation, though the two often intersect. Black and Latino families are more likely to be low-income, but even if incomes were equal, racial discrimination and prejudiced housing policies would still negatively affect the educational outcomes of black and Latino students, Reardon said.
It's hard to compare the two because the measures are different, but it is fair to say there's more racial segregation than income segregation, Reardon said.
“The fact that a new problem is showing up," Reardon said, "doesn’t mean that the old ones are going away."