Are exam schools really an academic promised land?
Studies suggest students don’t do any better by attending ‘elite’ academies
AFTER MONTHS OF debate amid heightened attention to racial justice issues, the Boston School Committee approved sweeping changes on Wednesday night to admission rules for the city’s three selective-entry 7-12 grade exam schools aimed at increasing diversity in their student populations.
The school committee voted to revamp admission standards in a way that will have all seats at exam schools awarded to students based on grades and test scores within a set of eight tiers based on the socioeconomic status of the census tract where they live. The goal is to promote more equitable access to exam school seats and greater racial and economic diversity at the three schools, particularly Boston Latin School, which is regarded as the premier exam school and where Black and Latino students are dramatically underrepresented compared to their share of the overall district student population. The changes mark a significant departure from the system in place for decades, which ranked all applicants citywide based on test scores and grades.
The three schools — Boston Latin, Boston Latin Academy, and the O’Bryant School of Mathematics and Science — are often referred to as “elite” exam schools, and seats there are coveted as a prized ticket to a great education and a pathway to college success beyond.
A growing body of research evidence suggests they aren’t necessarily.
Students at Boston’s exam schools, like those attending selective-admission high schools in New York City, Chicago, and other cities, unquestionably score higher than their non-exam school peers on all sorts of achievement measures, go on at greater rates to higher education, and gain admission to more selective colleges. But that’s hardly surprising, since they were selected for admission to the schools based on test scores and grades from elementary and middle school that predict such outcomes. In-depth research studies, including a 2014 analysis of the Boston and New York City exam schools, say there is little evidence, however, that attending an exam school has much to do with that success.
The 2014 research, carried out by investigators at Duke University and MIT, sought to tease out any effect of the schools themselves by looking at students who just cleared the cutoff mark for admission and comparing them to those who fell just below that line. The reasoning is that the two groups, for all practical purposes, had the same baseline academic achievement levels. The study found no clear benefit from exam schools on subsequent achievement scores, college enrollment and graduation rates, or the selectivity of colleges that students are accepted to. The one exception appeared to be a modest gain in 10th grade English scores, a finding mainly seen among students at O’Bryant.
Students at Boston Latin School do well there and in college afterwards, “but it’s because they were picked that way,” said Joshua Angrist, an economics professor at MIT and co-author of the report. “It’s not because Boston Latin improves their outcomes.”
The report, “Elite Illusion: Achievement Effects at Boston and New York Exam Schools.” reached similar conclusions about New York City’ selective-admission high schools. Harvard researchers Will Dobbie and Roland Fryer had previously documented similar findings for New York’s exam schools.
“We make these assumptions that these selective or predominately white schools are better because their performance is better. People have a hard time letting go of that,” said Boston School Committee member Hardin Coleman. “I’ve not seen a lot of evidence that the quality of teaching at the exam schools is a lot better than the quality anywhere else in the city.”
In Chicago’s selective-entry high schools, where the use of socioeconomic tiers to allocate 70 percent of seats formed a basis for the changes just adopted in Boston, a recent analysis found a similar lack of achievement effect from the schools. The study actually found evidence of negative effects on GPAs and the likelihood of attending a selective college for those attending exam schools from the lowest socioeconomic tier.
The exam school admissions debate ignited heated passions in Boston, with reform advocates calling for change as part of a move for racial justice and greater equity in education. Meanwhile, a group of white and Asian parents fiercely opposed the move to allocate seats within socioeconomic tiers, which will almost certainly reduce the number of slots at Boston Latin going to white and Asian students.
Coleman supported changing the admission policies to create greater equity when it comes to enrollment of Black, Latino, and low-income students. But he says studies suggesting that the exam school attendance, per se, may do little to change a student’s academic trajectory — coupled with the fact that nearly two-thirds of Boston 7-12 graders attend non-exam schools — underscore the urgency of improving the system’s other 122 schools and not solely focusing on who gets exam school seats.
Coleman said he’s been frustrated by the amount of attention consumed by the exam school debate, including the call by some to foster greater equity by expanding exam schools. “If you really want to make improvement, it’s about improving the quality of teaching and learning in all the schools,” he said. Coleman pointed to a devastating report released three years ago that found nearly 1 in 5 Boston Public School students was not on track to graduate, a figure that had barely moved since a similar report was issued a decade earlier. “If we’re really about closing the achievement gaps and raising our performance as a district, those are the issues that are most important in my mind,” Coleman said.
Will Austin, founder of the Boston Schools Fund, a nonprofit that awards grant money to Boston schools to accelerate improvement at what it calls “high-potential schools” and pushes to increase enrollment at high-performing schools, said there were good arguments for the admission changes. “Obviously there’s a value in making sure the exam schools are diverse and there’s equitable access,” he said. “But you have to make sure they are not the only quality options, real or perceived. The more attention we give to the exam schools, the more it just kind of reinforces this narrative.”
Making changes to the assignment system for exam schools or others in the district “does nothing to change the number of high-quality seats,” Austin said. “It changes who has access to them.”
Austin, who graduated from Boston Latin School 25 years ago, said the findings of the Duke and MIT study match his experience there, where he saw little evidence that the school itself was making a dramatic difference for students who all arrived with generally strong academic skills. “Teach, test, and hope for the best” is how he described the school’s approach when he was there.
Angrist and his colleagues say exam schools may, through clubs and activities not found at other schools, “expose students to ideas and concepts not easily captured by achievement tests or our post-secondary outcomes.” They also write in their paper on the Boston and New York systems that exam school attendance might still be linked to higher earnings in adulthood, a question they plan to explore in future research. But they say any such labor market gains “are likely to come through channels other than peer composition and increased cognitive achievement” at the schools.
Austin says such benefits are likely to come from the hard-to-quantify “social capital” that parents are often actually seeking when they say they’re looking for a high-quality school.
In voting to support the admission changes Wednesday night, School Committee member Ernani DeAraujo, who is a lawyer, recounted his own path as the son of a single mother who brought him to Boston from Colombia and wound up graduating from Boston Latin School. As someone who grew up as “an immigrant kid in poverty,” he said, his story “really is the American dream.”
DeAraujo said his Latin School education was key, but in a nod to the social-capital dimensions of that experience, he said, “more importantly, I think I got a network of folks that mentored me, supported me,“ and opened doors to internships and other opportunities.
Critics say the long-time use of an exam school admission system that ranked all students citywide based on test scores and grades tilted admissions unfairly toward white students and those from better-off families, who could pay for outside test prep courses.
Lisa Green, a North End parent who is active with Boston Education Equity Coalition, a group that pushed for all exam school seats to be allocated under the tiered structure, pointed to the huge endowment Boston Latin School has to supplement district resources as an example of the benefits students enjoy that aren’t necessarily captured in test scores. “Sixty-five million dollars goes a long way,” she said of the current holdings of the Boston Latin School Association, which raises money for the school, much of it from successful alumni.
Green said the coalition is mindful of the need to focus on improvement throughout the system in addition to changes to make exam school admissions more equitable. “You can do both,” she said. “There’s no reason to have this glaring inequity in either place.”In emphasizing the need to focus more broadly on improving school quality throughout the district, Coleman said the exam schools date to a time when only a small share of students were college-bound. “We forget that these schools were created when most secondary schools weren’t college prep,” he said. “It made sense to create schools for the relatively small share of kids who were going to college. That’s not the case anymore.”
Jeri Robinson, the school committee chair, echoed that sentiment after the committee voted unanimously to adopt the exam school admission changes. “Are we ready to provide all of our students what they need?” she asked. “We have 30 high schools. There’s no reason we only have three that people feel they want to go to. So to me this vote is just a step.”