Photo courtesy of San Juan Unified School District.
Diverse group of kindergartners in the San Juan Unified School District with large numbers of both Latino and white students.

While racial and ethnic segregation in the nation’s schools is strongly correlated with gaps in academic achievement, the income level of students’ families in a school rather than its racial or ethnic composition account for those gaps, according to a new study.

The study, based on massive amounts of data from schools attended by nearly all of the nation’s black and Hispanic students, was conducted by Sean Reardon, a professor at Stanford University’s Graduate School of Education, and other researchers from Stanford, Pennsylvania State University and St. John’s University in New York City.

Sean Reardon of Stanford University’s Graduate School of Education

Achievement gaps among black, Hispanic and white students, the study found, is “completely accounted for” by the poverty level of students in a school, as measured by the percentage of students who qualify for free and reduced priced meals.

“While racial segregation is important, it’s not the race of one’s classmates that matters,” the researchers concluded in the study released today. “It’s the fact that in America today, racial segregation brings with it very unequal concentrations of students in high and low poverty schools.”

“Differences in exposure to poverty may be more important for the development of achievement gaps than differences in exposure to minority students,” they state. The study looked at student test performance in math and English language arts between the 3rd and 8th grade.

The study looked at what it called de facto segregation in schools, districts and metropolitan areas across the United States with high concentrations of black and Hispanic students, in contrast to the de jure segregation that occurred in many parts of the United States, especially in the American South, as a result of official laws or ordinances barring access to schools based on race. 

While the study didn’t break down achievement gaps by state or district, it was accompanied by a tool kit that for the first time maps average test scores, how much students’ learning improves each year and trends in test score data for every school and district in the nation. Thus it is possible to drill down into hundreds of districts like Los Angeles Unified, Fresno Unified, West Contra Costa Unified and thousands more around the country to look at how these districts are doing on a range of indicators related to test scores — and the extent to which performance is improving over time.

The study underscores the consequences of the nation having virtually abandoned efforts to desegregate its schools over the last several decades as courts have lifted desegregation orders, and desegregation has become harder to accomplish as school districts in many metropolitan areas have had a declining share of white students.

This is especially the case in California where the demographic makeup of the schools has changed dramatically in recent decades. In the just completed school year (2018-19), white students made up 24 percent of California’s total enrollment of 6.2 million students, compared to 38 percent 20 years ago. Black students declined from 8.8 percent in 1998-99 to 5.4 percent of the total student enrollment, while enrollment of Latinos increased from 41 percent to 54 percent over the same period, as did that of Asian students, from 8.1 percent to 9.3 percent.

A 2016 report by UCLA’s Civil Rights Project asserted that Latino and black students in California attended some of the most segregated schools in the United States, meaning that they were most likely to attend a school with the lowest proportion of white students.

The researchers point out that efforts to desegregate schools have virtually ceased and instead the focus among school reformers and lawmakers is now primarily on trying to improve the quality of schooling “within a system of schooling that is highly segregated by both race and class.”

Reardon and his colleagues found that the more segregated a school system, the larger the average achievement gap, and that the gaps grow faster during the K-8 grades than in less segregated ones.

The study focused on averages for the entire nation and as a result does not highlight where students and schools are “beating the odds,” as described for example in a recent paper by the Learning Policy Institute, or the fact that there are many black and Hispanic students who perform far above average and are outstanding students.

The test scores reflect what the researchers broadly refer to as “educational opportunities,” which include not only what they are taught in school, but learning opportunities in their homes and in neighborhoods, their access to child care, preschool, and after-school programs and a range of other resources typically available to higher income children.

Share Article

Comments (13)

Leave a Comment

Your email address will not be published. Required fields are marked * *

Comments Policy

We welcome your comments. All comments are moderated for civility, relevance and other considerations. Click here for EdSource's Comments Policy.

  1. Mellanie 1 week ago1 week ago

    Having grown up in the ’60s and now working in an inner city school for 26 years, I guess I was mistaken in my belief that schools in large cities had taken care of the problems of inequality because of family economics. This article tells me that there are still inequalities in schools. I don’t understand why these problems have not been addressed in all this time??!!

  2. Ellen Graham 2 weeks ago2 weeks ago

    Zeev Wurman is right and Dr. Conrad is wrong. Home culture and expectations about education are far more critical than any other factor. It's just not politically correct to talk about the massive number of boys in hispanic and black homes without fathers. This does impact students because they themselves have told me. Yes teaching matters and too often, lazy teachers are not held accountable. Part of it is the teach to the test mentality … Read More

    Zeev Wurman is right and Dr. Conrad is wrong. Home culture and expectations about education are far more critical than any other factor. It’s just not politically correct to talk about the massive number of boys in hispanic and black homes without fathers. This does impact students because they themselves have told me.
    Yes teaching matters and too often, lazy teachers are not held accountable. Part of it is the teach to the test mentality that permeates low income schools. In upper class school districts, teachers have more academic freedom. For example, in my school fewer novels are read. I always use whole text but am discouraged from doing so, instead I am pointed in the direction of the horrific Pearson high school English text. I don’t use it and never will.

    School has to be relevant but parents also have to set examples. I have had students whose own parents went back to school as adults and this greatly motivated them to do well in school. Parents matter, home culture matters, fathers matter but standardized tests don’t matter and should be done away with.

  3. Bryan Reece 3 weeks ago3 weeks ago

    Good article. The relationship between income, race and ethnicity is complex. There are a few universities that continue to promote their diversity statistics but primarily recruit middle and upper income students if color. This needs to be addressed across all education.

  4. Brigitt 3 weeks ago3 weeks ago

    I worked in the flatlands of West Contra Costa Unified from 1996-2000. All my students were poor. All my students had free/reduced lunch. When I moved to Southern California, I worked with mostly Hispanic, high poverty students as well. Family life and parent engagement is crucial. Poor children with engaged parents improved academically with me. Poor students who had parents or guardians who did not engage were almost always behind 1-3 grade levels. Research must … Read More

    I worked in the flatlands of West Contra Costa Unified from 1996-2000. All my students were poor. All my students had free/reduced lunch. When I moved to Southern California, I worked with mostly Hispanic, high poverty students as well.
    Family life and parent engagement is crucial. Poor children with engaged parents improved academically with me. Poor students who had parents or guardians who did not engage were almost always behind 1-3 grade levels.
    Research must be done on not just how poor students of color (white, brown, black) perform in school, but how much time parents spend helping them every night. Discussions they have at the dinner table, family expectations.
    I’m tired of people “blaming” teachers in these districts or classrooms. Dr. Conrad’s comment about quality of teachers and how these students get new teachers who are not well qualified is a detail. Teachers like myself who work in these schools are and were MAINLY seasoned teachers. Not novices. We attended 101 workshops and meetings in our district, we went on to get master’s and Ph.D’s. Some of us (like myself) are National Board Certified Teachers which is a rigorous certification to receive. Why? Because we are dedicated to help poor WHITE, BLACK, HISPANIC, ASIAN, etc. students. We study data, modify our teaching to accommodate and reach children everyday.
    It can’t all be because of teachers when we look at the gap. Poverty is a major factor and so is Parent Involvement AND their education.

  5. Ariel 3 weeks ago3 weeks ago

    I still think it must be a combination of factors that all contribute to the same issue of resource allocation. There’s a reason for higher poverty rates among black and brown students, and it’s being culturally different from the white majority that controls curriculum, staffing, lawmaking, and all aspects of education at the leadership level.

  6. Bill Leiter 3 weeks ago3 weeks ago

    Well said Zeev Wurman. The full 60 page report is peppered with complex appearing equations and symbols that few people would understand. And that’s its point I think: to obfuscate and impress. That’s all. Your 5 or 6 sentences posted here contain more actual insight than the whole report. Let’s hope that most people will be able to see it for what it is.

  7. Paul Muench 4 weeks ago4 weeks ago

    How is this different than the Coleman Report?

  8. Zeev Wurman 4 weeks ago4 weeks ago

    Well, when it was the color of skin that was the supposed determinant of achievement, East-Asian students showed this couldn’t have been the whole story. Now when the argument is that it is poverty, the relative success of "poor" East-Asian students—and there are quite a few of those—show that the new glib explanation is still just that — a glib explanation. Large achievement gaps exist within same schools, and frequently within the same classroom. That … Read More

    Well, when it was the color of skin that was the supposed determinant of achievement, East-Asian students showed this couldn’t have been the whole story. Now when the argument is that it is poverty, the relative success of “poor” East-Asian students—and there are quite a few of those—show that the new glib explanation is still just that — a glib explanation.

    Large achievement gaps exist within same schools, and frequently within the same classroom. That can’t be a matter of major differences in resources or teacher quality. In fact, often just the opposite—spending on disadvantaged students is typically higher than on non-disadvantaged students.

    So inquiring minds would like to know when those distinguished researchers direct their attention to home cultures (and the resulting school discipline).

  9. SpecialKinNJ 4 weeks ago4 weeks ago

    A reliable source https://object.cato.org/sites/cato.org/files/pubs/pdf/pa746.pdf provides evidence indicating that per pupil costs/expenditures have increased at a 45 degree angle since the 1970s, but average reading, writing and arithmetic scores have not increased at all (see especially Page. Data for states are provided.) The stability of average performance on tests of reading writing and arithmetic is due not so much to lack of effort to change that pattern as to resistance of such abilities to change — as … Read More

    A reliable source https://object.cato.org/sites/cato.org/files/pubs/pdf/pa746.pdf provides evidence indicating that per pupil costs/expenditures have increased at a 45 degree angle since the 1970s, but average reading, writing and arithmetic scores have not increased at all (see especially Page. Data for states are provided.)

    The stability of average performance on tests of reading writing and arithmetic is due not so much to lack of effort to change that pattern as to resistance of such abilities to change — as suggested by data for a recent (almost) 30-year period showing the average performance of all students as well as students classified by race/ethnicity, taking an internationally recognized test (the SAT). See table below, showing SAT Critical Reading averages for selected years. Note. Data for Asian-Americans indicate that they’re exceptions to that rule. Their average has improved steadily, and they’re now “leaders of the pack”.

    Table 1. SAT Critical Reading average selected years
    1987 ’97 2001 ’06 ’11 ’15 ’16
    507 505 506 503 497 495 494 All students
    524 526 529 527 528 529 528 White
    479 496 501 510 517 525 529 Asian
    …………………………… .. …..436 Hispanic
    457 451 451 454 451 448 Mex-Am
    436 454 457 459 452 448 Puerto R
    464 466 460 458 451 449 Oth Hisp
    471 475 481 487 484 481 447 Amer Ind
    428 434 433 434 428 431 430 Black
    SOURCE: U.S. Department of Education, National Center for Education Statistics.(2012). Digest of Education Statistics, 2011 (NCES 2012-001), Chapter 2.
    SAT averages for college-bound seniors, by race/ethnicity: Selected years,1986-87 through 2010–11 Data for 2015&2016 https://nces.ed.gov/fastfac…
    Note 2016 data were not provided for Hispanic subgroups.
    If SAT averages haven’t changed materially over almost 30 years, despite the effort, time and money expended to improve educational programs for all students, it seems reasonable to assume that we shouldn’t expect any meaningful change in average level of performance in this critically important ability in the foreseeable future.
    And what if the achievement gap is here to stay!! And it is important
    to remember that correlation does not imply causation.

  10. Dr. Bill Conrad 4 weeks ago4 weeks ago

    The chief determinant of whether economically disadvantaged students will succeed in school is the quality of the teachers they get and the ability of those teachers to know their content well and to effectively implement high quality professional practices . Economically disadvantaged students are most likely to get the most novice or TFA teachers out of an overall teacher pool that is derived from color in the lines colleges of education unprepared in content … Read More

    The chief determinant of whether economically disadvantaged students will succeed in school is the quality of the teachers they get and the ability of those teachers to know their content well and to effectively implement high quality professional practices .

    Economically disadvantaged students are most likely to get the most novice or TFA teachers out of an overall teacher pool that is derived from color in the lines colleges of education unprepared in content knowledge and pedagogy. That is their fate and it is the root cause of the achievement gap. Economically disadvantaged parents do not have the economic wherewithal to pay for extra tutoring and after school support line White and Asian parents do! They are the most sensitive to the quality of the teachers.

    Local school boards and milquetoast administrators hold the line on allocating their most qualified teachers from predominantly wealthy white schools to schools with predominantly Brown and economically disadvantaged students!

    Correlation does not mean causation. Scratch the surface. Visit classrooms where the economically poor Brown children are “educated” and you will see for yourself!

    Replies

    • Ann 3 weeks ago3 weeks ago

      Dr. Conrad, I'm afraid I disagree. I have never worked in a district that allowed teachers to choose sites. Teachers usually take the placement at whatever site they get hired, then rarely move. The problem is there are far to many who receive tenure have not shown themselves to be good teachers. Principles simply do not say no often enough. The larger problem and where it begins in the schools of education that 1) admit … Read More

      Dr. Conrad, I’m afraid I disagree. I have never worked in a district that allowed teachers to choose sites. Teachers usually take the placement at whatever site they get hired, then rarely move. The problem is there are far to many who receive tenure have not shown themselves to be good teachers. Principles simply do not say no often enough. The larger problem and where it begins in the schools of education that 1) admit those with lesser academic qualifications and 2) do an abhorrent job teaching pedagogy instead spending too much time on ‘culture’ and ‘social justice’. This has been so for decades. Twenty years back in California we were getting pretty well trained teachers from Fresno State but that has deteriorated as well. BTW the same situation for admin credentials.

      • Bill Conrad 2 weeks ago2 weeks ago

        I don't think that we are in disagreement. The woeful colleges of education are the root cause of producing a pool of teachers in general who are not well prepared in either content or pedagogy. These are the teaches that students of color and economically disadvantaged students are destined to get. If we do a thought experiment advocated by Mutiu Fagbayi from Performance Fact, we might be better able to see the issue. If … Read More

        I don’t think that we are in disagreement. The woeful colleges of education are the root cause of producing a pool of teachers in general who are not well prepared in either content or pedagogy. These are the teaches that students of color and economically disadvantaged students are destined to get.

        If we do a thought experiment advocated by Mutiu Fagbayi from Performance Fact, we might be better able to see the issue. If we randomly selected 30 students of color or economically disadvantaged students and assigned them a teacher. Is it possible that this teacher could ensure that 50% of the students meet standards? If we assigned a different teacher. Is it possible that that teacher could ensure that 80% of the students meet standards? Is it possible to assign a teacher who only gets 25% of the students to meet standards. Of course, all of these scenarios are true. So it is not the children who are the issue, it is the adults and their level of preparedness. No?