Study: Most School Districts Have Achievement Gaps
Database sheds new light on achievement disparities
In education, a rising tide of resources does not necessarily lift all boats.
Racial achievement gaps exist in nearly every community across the country with a measurable black or Hispanic population—and many districts with a traditional commitment to education and resources to serve all students instead have the worst inequities, according to new research comparing achievement gaps across state lines.
Using a database of five years of test scores from more than 40 million students nationwide, Stanford University researchers Sean Reardon, Demetra Kalogrides, and Kenneth Shores analyzed how racial achievement gaps look in different parts of the country, and how segregated schools widen those gaps.
"I think we like to think, 'Here we have this problem, but it's fixable. We know we could figure it out.' It's not clear we've figured it out," said Reardon, a professor of poverty and inequality in education at Stanford. "There's some deep ... problems that we as a society haven't faced up to yet."
The Stanford researchers and Harvard University education professor Andrew Ho linked scale scores for state tests to the scales for the National Assessment of Educational Progress in the same grades and subjects, and used that yardstick to compare average achievement gap trends for students in 3rd through 8th grades in more than 12,000 districts across the country, from 2009-2013.
Minding the Gaps
State differences in testing and accountability systems have frustrated researchers and families alike since the early days of the No Child Left Behind Act, but a massive new database provides a link to compare achievement trends in districts across the country.
The new Stanford Education Data Archive, a four-year project by researchers at Stanford and Harvard universities, compiles 215 million state test scores from 40 million students—every test taken by a public school student in grades 3 through 8 from 2009-2012—and disaggregated by race, grade, subject, and proficiency level. The data cover more than 12,000 of the nation’s 14,000-plus school districts and 384 metropolitan areas, encompassing both regular and charter public schools. It is the largest and most comprehensive database of its kind to date.
Sean F. Reardon, a professor of poverty inequality in education, his colleagues at Stanford, and Harvard education professor Andrew Ho, analyzed the performance of different student groups in each district. They ranked districts in each state, both overall and separately, by racial/ethnic categories, by comparing their performance at multiple proficiency levels: “below basic,” “basic,” “proficient,” and “advanced.” This allows them to compare districts within a state, but not across state lines.
To compare districts across states, they “linked” state scores to performance on the nationally representative National Assessment of Educational Progress, in the related subjects and grade levels, Ho said.
Crosswalking the Data
An earlier federal crosswalk study of state tests and the NAEP, which is a set of large, nationally sampled tests in core subjects like math and reading, found that states set their proficiency thresholds at widely disparate levels. A 4th grader who was deemed a proficient reader on state tests in Arizona or Virginia could have significantly lower scores on NAEP than a proficient reader in neighboring New Mexico or West Virginia, for example.
The researchers were able to confirm that the NAEP linking system worked by comparing the differences they predicted among major U.S. urban districts to those districts’ actual performance on the NAEP’s most recent Trial Urban District Assessment, which compares NAEP results in 21 urban districts throughout the country. For example, the system accurately predicted that Chicago 4th graders would perform about a half-grade level, or 10 scale-score points, below New York 4th graders in reading in the 2013 TUDA.
Their validation checks show that this method enables them to conduct research on differences among districts both within and across states. “It’s statistically complicated, but it gives answers that are very accurate,” Reardon said.
The data “don’t tell us which school districts are better than others ... because test scores of kids in a district are the result of everything that’s contributed to the kid’s development since conception,” he warned. “These [data] are really good for big picture stuff, but it shouldn’t be used to rank schools that are statistically very similar to each other.”
A large data set from the studies has been made available to researchers and the public, and Reardon is working to bring in foundations to provide grants for those who want to mine it. However, the statistical method could be used to add and analyze new test data, to help compare educational achievement and progress across the country as states explore different testing regimes and accountability measures under the Every Student Succeeds Act.
That five-year period enabled researchers to focus on districts completing a decade of state and federal accountability initiatives designed to close academic gaps between white students and their black and Hispanic peers. Of the 2,500 school districts with a large enough sample of black students to measure their achievement gaps, Reardon and his colleagues found only one with no black-white gap: Detroit.
"Detroit is not the poster child for reducing the achievement gap," Reardon said in an interview with Education Week. "The achievement gap is zero in Detroit largely because everyone's doing really poorly, not because black students are doing particularly well."
Several of the other districts with the lowest achievement gaps have similar profiles of high poverty and multiple attempts at education overhauls, suggesting their low achievement gaps come from lower performance overall.
Moreover, the researchers found some of the biggest black-white achievement gaps in the country—where black students lag their white peers by more than 1.5 full standard deviations, or four to five grade levels on the NAEP scale—in relatively prosperous university towns, including Berkeley, Calif. (home of University of California, Berkeley), and Evanston, Ill. (home of Northwestern University).
"Richer places have bigger achievement gaps than poorer places, all else being equal—which is quite striking and disturbing, since you'd hope that those places that have the most resources would be most effective at reducing the gaps," Reardon said. He has a theory on why these higher education centers haven't produced more equitable K-12 achievement: "Increased competition … a hyper achievement orientation."
When competition increases, black and Hispanic families often lose out. Prior research shows on average, a black family lives in a poor neighborhood and their children attend schools with fewer resources than white families at the same income level.
Groups like the Education Trust, which work with districts to close achievement gaps, have seen that dynamic in practice.
"Equality doesn't necessarily mean 'the same,'" said Sonja Brookins Santelises, the Education Trust's vice president of K-12 policy and practice and the newly tapped CEO of Baltimore city schools. "Higher-income families are investing more in personal wealth in the education of their children, beyond what comes from the school. If you give a low-income student the exact same resource as an upper-income student, but the need is greater for the low-income student, the gap is going to expand."
Segregation 'Feedback Cycle'
That parental drive to find the best education for their kids could ramp up economic segregation in neighborhoods, too.
A separate study by Ann Owens, an assistant sociology professor at the University of Southern California, found that neighborhoods in the 100 largest cities became steadily more isolated by income between 1990 and 2010—but the segregation was driven by families with school-age children.
"There's really a feedback loop, and it's often framed as, we can never have integrated schools while we have segregated neighborhoods," Owens said. "But the flip side is true, as well: As long as schools are unequal and linked to neighborhoods, that's going to play a big role in neighborhood segregation."
A second study using the Stanford data looked at the effects of 16 different facets of racial segregation, including school and residential isolation, segregation within and between districts, racial or socioeconomic isolation, and differences in students' likelihood of being exposed to peers of other races or income groups.
While all types were associated with wider achievement gaps that worked against black and Hispanic students, the strongest gap predictor was the difference in poverty rates between schools with large populations of white or black and Hispanic students.
The finding "suggests it's something about school quality—not only about racial segregation, but about the fact that racial segregation in America almost inevitably leads to these kind of disparities in [students'] exposure to poverty and differences in the kinds of resources that schools have."
Other studies have found systemic differences in resources and in the types of assignments given to students in high-poverty schools; in a 2015 study, for example, 8th graders in high-poverty schools were given shorter writing assignments calling for less critical thinking.
Going forward, the researchers are trying to identify common factors in districts where achievement gaps for black and Hispanic students are narrowing.
"Really, there are very, very few school districts that serve a large proportion of poor students and that have achievement that's even at the national average," Reardon said. "That suggests we may not be able to just 'school reform' our way out of that kind of inequality."
Vol. 35, Issue 30, Pages 1,12-13