Elite schools, prized by parents and politicians alike, may actually hurt disadvantaged students more than they help, new research shows
Kevin Mahnken | September 9, 2019
Applying to one of Chicago’s 11 selective enrollment high schools is a little like banking on the Bears to triumph at the Super Bowl: probably futile and, at times, downright depressing.
The elite public schools, which admit students on the basis of high grades and exam scores, attract thousands of high-achieving applicants each year. In 2018, roughly 30 percent were ultimately offered a spot, and less than half of those were accepted to their top choice. That level of demand is comparable to the nationwide quest for slots at highly touted universities. It’s also understandable, given that five of the academies were recently listed among U.S. News & World Report’s ranking of the nation’s 100 best public high schools.
To temper that extraordinary competition, Chicago Public Schools have instituted an affirmative action system that lowers admissions criteria for disadvantaged students. By expressly setting aside places for students of low socioeconomic status, Chicago has rendered the sector substantially heterogeneous along class and racial lines — so much so that the district is now held up as a model to others that have struggled to diversify their own exam schools.
New York’s system of exam schools has come under particularly close scrutiny; admissions data show that just seven black students were accepted to the prestigious Stuyvesant High School last spring (out of a freshman class of nearly 900). Last year, after the city proposed changes to the schools’ testing and admissions policies that would result in more racial diversity, Asian-American parents and community activists sued.
Even more recently, a task force convened by Mayor Bill de Blasio delivered a plan to eliminate most of the gifted education programs in New York’s public schools, along with the bulk of the city’s screened and selective admissions schools, in order to desegregate what some believe is a two-track system. The proposal has already drawn fierce criticism.
According to new research, Chicago’s diversity measures haven’t succeeded in lifting academic performance for those they were designed to help. Two recent studies — one released as a working paper and another soon to be published in an academic journal — show that students at Chicago’s selective enrollment high schools see no improvement in their test scores. In fact, academic records show that they earned worse grades and GPAs than their peers who were rejected from the schools.
While all students were likely to see their class ranks adversely affected (a predictable result of gaining a horde of extra-diligent classmates), the most disadvantaged saw the biggest declines. Disturbingly, they were also less likely to enroll at selective colleges than similar kids who weren’t offered admission to an elite high school.
Lisa Barrow, a senior economist at the Federal Reserve Bank of Chicago and a co-author of one study, said she had hoped to find evidence that elite high schools could make a difference for nonwhite and less affluent students. Those hopes were dashed.
“We thought, ‘Well, surely [selective high schools] must matter in Chicago because they have this place-based affirmative action,’” she said. “So we were surprised and disappointed to discover that not only did they not generate better outcomes for students from low-socioeconomic-status neighborhoods — in terms of the academic measures — they’re potentially making them worse off.”
The notion that elite public schools don’t actually improve outcomes for their students is not a new one — though it remains counterintuitive. Prior research has indicated that the academic performance of students at storied institutions like Boston Latin or Stuyvesant was more attributable to their own aptitude (and perhaps their families’ resources) than to anything going on inside classrooms. In other words, the high-flying applicants would likely excel anywhere.
Economists even coined a term for the misapprehension: the “Elite Illusion.”
Chicago’s system was meant to be different, though. In a recent overview of selective high schools in eight major cities, Brookings Institution scholar Richard Reeves noted that “only in Chicago are the [selective school enrollments] close to being representative of the city as a whole, in terms of both race and economic disadvantage.”
The schools’ inclusiveness is the legacy of a long-running desegregation order imposed by a federal court in 1980. When a judge vacated that decree in 2009, the city switched from using a race-based affirmative action plan to one rooted in geography and socioeconomic status. Every census tract in the city was sorted into one of separate tiers according to variables like median family income, homeownership and prevalence of single-parent households. Tier-one areas are the least advantaged, tier-four areas the most.
Admissions rankings for selective high schools are determined using a 900-point scale, equally weighted between students’ seventh-grade GPA in core subjects, their performance on seventh-grade standardized tests and their scores on an admissions exam. For each of Chicago’s 11 selective high schools, only 30 percent of seats are reserved for the students with the highest admissions scores; another 5 percent are left at principals’ discretion, and the rest are evenly divided among top applicants in the four socioeconomic tiers.
To determine the effects of the system, Barrow and her co-authors gathered academic data for all applicants to Chicago’s selective high schools, including their application scores, GPAs, ACT scores and graduation and college-enrollment rates. They were thus able to directly compare students who were admitted to a selective high school with those who narrowly missed the cutoff.
In tracking those students, the study found results echoing those from prior research: There was “no evidence” that attending a selective high school raised ACT scores for students of any socioeconomic tier. High-performing students who scored just below the admissions cutoff learned about as much as those who scored just above. This was even true of rejected tier-one applicants, who would theoretically be consigned to high schools with much lower graduation rates and test scores.
But the authors also highlighted a phenomenon that hadn’t been detected before: Selective high school students earn slightly lower grades and GPAs than their peers at non-selective high schools. And if that weren’t bad enough, those declines are progressively worse for disadvantaged students.
Attending a selective school modestly depressed the scholastic performance of tier-four students, whose ninth-grade GPA was measured at .09 points lower than their peers who weren’t admitted to a selective-admission school. Their class ranking was consequently reduced by 10 percentile points, the authors found. But the effects were larger for their classmates from less affluent tier-one neighborhoods: a drop of .29 points of GPA, and a 17-percentile-point reduction in class rank.
Most dismaying of all, tier-one students were actually six percentage points less likely to attend a selective college than their peers who didn’t attend an elite high school. In other words, for students from Chicago’s worst-off neighborhoods, being accepted to an elite school doesn’t lift them to incredible new heights; it seems to be holding them back.
Barrow said that the study didn’t reveal the cause of the problem.
“It could be they’re not applying to selective colleges because they have somewhat lower GPAs, and they look at this data and it suggests they wouldn’t get in,” she said. “It could be that they’re not getting in. It could be that they’re not getting scholarships. Or it could be that counselors are influencing where they apply in ways that respond to their grades. We don’t know what’s going on.”
Charter schools’ role
A second piece of research may go a long way toward providing the explanation.
Just released as a working paper by the National Bureau of Economic Research, and awaiting peer review, the study was conducted by MIT economists Parag Pathak and Josh Angrist, who originally coined the term “Elite Illusion.” The pair, along with co-author Román Andrés Zárate, aimed to use Chicago’s selective high schools to examine the question of academic “mismatch”: the theory that students admitted to elite institutions through affirmative action aren’t truly prepared for the academic rigors they encounter there.
In conducting their own analysis of academic results, Pathak and Angrist found no evidence of mismatch; whether or not students benefited from affirmative action, they saw no gains in their math and reading scores at elite high schools, the authors write. Rather, they found that the selective schools are, in fact, driving academic outcomes, but in an unexpected way: by accepting students who might otherwise have received a better education at one of Chicago’s high-performing charter school networks.
Specifically, the paper finds that many applicants rejected from selective schools ultimately attend a high school in the Noble Network of charters, where they go on to attain much higher test scores. By admitting some students, therefore, the coveted exam schools are in effect “diverting” them from attending Noble schools, where they would perform better.
“We ask, ‘Where are these kids going to go if they’re not offered a seat at a selective school?’” Angrist said in an interview. “The answer is that most of them go to traditional schools, but a very large fraction goes to Noble.” Selective schools’ comparatively poor test results were the result of a substantial fraction of their unsuccessful applicants going on to enroll — and score better — at Noble charters, he argued.
The finding builds on earlier research on Noble conducted by academics Matthew Davis and Blake Heller, who found that attending a Noble school made students 10 percentage points more likely to complete at least four semesters of college. Angrist said that his new study reconciles the conclusions from both Barrow’s study (that low-income students at exam schools actually realize negative academic impacts) and Davis and Heller’s (that students do particularly well at Noble), suggesting that they are actually “one and the same.”
Not all education observers will agree that elite schools like those in Chicago, New York and elsewhere truly provide no benefit to their students. Many point to the social capital gains that accrue to pupils studying alongside more affluent peers for the first time. Even in Barrow’s otherwise-gloomy study, survey data indicated that students at selective high schools were more positive about their school environments and relationships with their peers. That kind of finding “can’t be dismissed,” Barrow said.
“These differences in high school experiences may in part help explain why the selective-enrollment high schools are so popular despite having limited, and perhaps negative, impacts on academic outcomes,” she and her co-authors wrote.
Still, given the scant evidence of their academic efficacy, and the mounting proof that high-performing charter schools like Noble are showing success closing achievement gaps between student populations, Angrist said that he hoped both parents and policymakers would adjust their perceptions of both sectors.
“So much media attention and political energy goes into discussions of who gets to go to exam schools. They’re way oversubscribed, and there’s a vigorous debate about minority access to those schools. We’re basically saying, ‘You’re arguing over something that may not be worth much. Your view of this is potentially being distorted by a misreading of the facts.’”