- Black, Hispanic, and Native American children attend schools that are on average at the 35th-40th percentile of performance compared with other schools in the same state. White and Asian children are in schools at close to the 60th percentile.
- There is almost no change across grades in reading scores. There is noticeable improvement in mathematics scores in schools attended by Hispanics, Asians, and Native Americans, though not for Blacks.
- Latent cluster analysis identified six common types of schools: three high-poverty schools (one is disproportionately located in central cities, is more Hispanic than black; another is more heavily black but mixed in location; and the third is also in mixed locations and is mixed in racial composition, though three-quarters of Native American children attend this type of school), and three with lower levels of poverty (one that is almost 90 percent white, lowest in poverty, and predominantly located in the suburbs; another that is almost exclusively white but most commonly found in non-metropolitan areas; and a third that is relatively affluent, disproportionately suburban, with a clear white majority but nontrivial shares of other groups.
- Racial composition of schools is important because schools with more minorities do worse, even after controlling for poverty.
- School poverty has as large an effect as the black of Hispanic share of students.
- In terms of district characteristics, adults educational levels matter much more than issues of family disruption; the percentage of foreign born people has no significant effect; and there is no evidence that metropolitan location independently affects school performance.