Looking at Segregation Through the Peer Effects Lens

That our children attend schools that are segregated by race is probably not a surprise for any of us. While, as researchers, we might debate how consequential segregation is, we can likely agree that, on its face, segregation raises some important societal concerns; it challenges our sense of what a moral and fair system looks like. It poses barriers to social cohesion, inclusion, and integration—and their well-known positive impacts on society—and it limits our children’s preparedness for the multicultural world in which they live.

As we mark the 60th anniversary of the Supreme Court’s Brown vs. Board of Education decision, and the declaration that “separate but equal” is unconstitutional, we look back on both progress made in desegregating schools and, more recently, backtracking on those efforts and current initiatives that sideline them.  Although separate but equal is unconstitutional, separate and unequal is very much a reality.

The paper that EPI and BBA released earlier today, Segregation and Peers’ Characteristics in the 2010-2011 Kindergarten Class: 60 Years after Brown versus Board, helps sustain that momentum in three ways: first, it shows how segregated our 2010-2011 kindergarten class is; second, it illustrates the differing socioeconomic characteristics of children  depending on their schools racial composition ; and third, it shows how academic performance ranks at the beginning of the children’s school life and at the end of their kindergarten year. We describe these two aspects for the different population groups (or races/ethnic groups), across schools where the proportion of minority students is very low (up to 10 percent minority) to very high (more than 75 percent). This descriptive analysis is based on the National Center of Education Statistics’ Early Childhood Longitudinal Study, Kindergarten Class of 2010-2011, a nationally representative survey of this young kindergarten cohort.

In essence, our analysis paints a picture of the characteristics of kindergartners and their peers across schools with various levels of segregation. We show that if all our kindergarteners were in a gigantic (hypothetical) classroom, just slightly over half of the class would be white, another quarter would be Hispanic, roughly an eighth black, and the remaining eighth composed of Asian children and children of other races/ethnicities. Because kids are not distributed randomly across schools, however, over one third of kindergartners attend schools in which minority students are less than 25 percent of the school body, while over one fourth of kindergartners attend schools in which more than 75 percent of the students are minorities.

In other words, schools are more likely to be at greater extremes than we might expect, with a student’s race predicting which of those extremes he inhabits. White students are disproportionately in the former schools (61 percent), while most Hispanic and black students are in very heavily minority schools (55 and 56 percent, respectively).

More importantly, segregation by race hides marked segregation by income: 61 percent of white students attend schools with a kindergarten poverty rate of 11 or 12 percent. This contrasts with a poverty rate of 50 percent among kindergarteners in schools serving the majority of black and Hispanic students. These sharp disparities illustrate the extremes noted above and both veer far from the average U.S. kindergarten poverty rate of 25 percent. These differences in peers parallel ones regarding mothers’ levels of educational attainment and the odds that a child speaks English at home.

When we look at relative academic performance at kindergarten entry and again in the spring—i.e., at students’ placement on the distribution of scores in reading, math, and approaches to learning—we see the above disparities mirrored in these outcomes. Students in heavily white schools advance over the year (including black students, who start especially far behind their peers in these schools), while students in the most heavily minority schools lose the most ground, relative to the average of their cohort.

Critically, we demonstrate that segregation by race and segregation by income are intertwined. Our analysis shows how students’ socioeconomic and academic environments are shaped not only by their own race/ethnicity but also of that of their peers, and by differences in those peers’ non-racial characteristics. Consequently, sources of disadvantage are not evenly balanced across schools, and they are unlikely to be addressed and compensated for by policies.

Our analysis confirms what prior research indicates: race is not the real culprit limiting student and school progress. Rather, the socioeconomic disadvantage that is closely correlated with minority status and concentrated in classrooms of black and Hispanic children is at the heart of these impediments to academic progress. As such, while race is not the real culprit, failing to alleviate racial segregation means we are unlikely to substantially narrow opportunity gaps, and that Brown’s goal of schools that are united and equal will remain elusive.